netdevice.h 153.5 KB
Newer Older
1
/* SPDX-License-Identifier: GPL-2.0-or-later */
L
Linus Torvalds 已提交
2 3 4 5 6 7 8 9 10
/*
 * 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
 *
11
 * Authors:	Ross Biro
L
Linus Torvalds 已提交
12 13 14
 *		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>
15
 *		Alan Cox, <alan@lxorguk.ukuu.org.uk>
L
Linus Torvalds 已提交
16 17 18 19 20 21 22 23
 *		Bjorn Ekwall. <bj0rn@blox.se>
 *              Pekka Riikonen <priikone@poseidon.pspt.fi>
 *
 *		Moved to /usr/include/linux for NET3
 */
#ifndef _LINUX_NETDEVICE_H
#define _LINUX_NETDEVICE_H

A
Al Viro 已提交
24
#include <linux/timer.h>
25
#include <linux/bug.h>
26
#include <linux/delay.h>
A
Arun Sharma 已提交
27
#include <linux/atomic.h>
28
#include <linux/prefetch.h>
L
Linus Torvalds 已提交
29 30 31 32
#include <asm/cache.h>
#include <asm/byteorder.h>

#include <linux/percpu.h>
33
#include <linux/rculist.h>
34
#include <linux/workqueue.h>
T
Tom Herbert 已提交
35
#include <linux/dynamic_queue_limits.h>
L
Linus Torvalds 已提交
36

37
#include <linux/ethtool.h>
38
#include <net/net_namespace.h>
J
Jeff Kirsher 已提交
39
#ifdef CONFIG_DCB
40 41
#include <net/dcbnl.h>
#endif
42
#include <net/netprio_cgroup.h>
43
#include <net/xdp.h>
44

45
#include <linux/netdev_features.h>
46
#include <linux/neighbour.h>
47
#include <uapi/linux/netdevice.h>
48
#include <uapi/linux/if_bonding.h>
49
#include <uapi/linux/pkt_cls.h>
50
#include <linux/hashtable.h>
51

52
struct netpoll_info;
53
struct device;
54
struct phy_device;
55
struct dsa_port;
56

57
struct sfp_bus;
58 59
/* 802.11 specific */
struct wireless_dev;
60 61
/* 802.15.4 specific */
struct wpan_dev;
R
Robert Shearman 已提交
62
struct mpls_dev;
63 64
/* UDP Tunnel offloads */
struct udp_tunnel_info;
65
struct bpf_prog;
66
struct xdp_buff;
L
Linus Torvalds 已提交
67

68 69
void netdev_set_default_ethtool_ops(struct net_device *dev,
				    const struct ethtool_ops *ops);
70

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
#define MAX_NEST_DEV 8

77 78 79 80 81 82 83 84 85 86 87 88
/*
 * 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
B
Bjorn Helgaas 已提交
89 90
 * higher layers. Virtual network devices transmit synchronously; in this case
 * the driver transmit return codes are consumed by dev_queue_xmit(), and all
91 92 93 94 95
 * others are propagated to higher layers.
 */

/* qdisc ->enqueue() return codes. */
#define NET_XMIT_SUCCESS	0x00
96 97 98
#define NET_XMIT_DROP		0x01	/* skb dropped			*/
#define NET_XMIT_CN		0x02	/* congestion notification	*/
#define NET_XMIT_MASK		0x0f	/* qdisc flags in net/sch_generic.h */
L
Linus Torvalds 已提交
99

100 101 102
/* 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. */
103
#define net_xmit_eval(e)	((e) == NET_XMIT_CN ? 0 : (e))
L
Linus Torvalds 已提交
104 105
#define net_xmit_errno(e)	((e) != NET_XMIT_CN ? -ENOBUFS : 0)

106
/* Driver transmit return codes */
107
#define NETDEV_TX_MASK		0xf0
108

109
enum netdev_tx {
110
	__NETDEV_TX_MIN	 = INT_MIN,	/* make sure enum is signed */
111 112
	NETDEV_TX_OK	 = 0x00,	/* driver took care of packet */
	NETDEV_TX_BUSY	 = 0x10,	/* driver tx path was busy*/
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
/*
B
Bjorn Helgaas 已提交
135
 *	Compute the worst-case header length according to the protocols
L
Linus Torvalds 已提交
136 137
 *	used.
 */
G
Graf Yang 已提交
138

139 140 141
#if defined(CONFIG_HYPERV_NET)
# define LL_MAX_HEADER 128
#elif defined(CONFIG_WLAN) || IS_ENABLED(CONFIG_AX25)
142 143 144 145 146
# if defined(CONFIG_MAC80211_MESH)
#  define LL_MAX_HEADER 128
# else
#  define LL_MAX_HEADER 96
# endif
L
Linus Torvalds 已提交
147
#else
148
# define LL_MAX_HEADER 32
L
Linus Torvalds 已提交
149 150
#endif

151 152
#if !IS_ENABLED(CONFIG_NET_IPIP) && !IS_ENABLED(CONFIG_NET_IPGRE) && \
    !IS_ENABLED(CONFIG_IPV6_SIT) && !IS_ENABLED(CONFIG_IPV6_TUNNEL)
L
Linus Torvalds 已提交
153 154 155 156 157 158
#define MAX_HEADER LL_MAX_HEADER
#else
#define MAX_HEADER (LL_MAX_HEADER + 48)
#endif

/*
159 160
 *	Old network device statistics. Fields are native words
 *	(unsigned long) so they can be read and written atomically.
L
Linus Torvalds 已提交
161
 */
G
Graf Yang 已提交
162

E
Eric Dumazet 已提交
163
struct net_device_stats {
164 165 166 167 168 169 170 171 172
	unsigned long	rx_packets;
	unsigned long	tx_packets;
	unsigned long	rx_bytes;
	unsigned long	tx_bytes;
	unsigned long	rx_errors;
	unsigned long	tx_errors;
	unsigned long	rx_dropped;
	unsigned long	tx_dropped;
	unsigned long	multicast;
L
Linus Torvalds 已提交
173 174
	unsigned long	collisions;
	unsigned long	rx_length_errors;
175 176 177 178 179
	unsigned long	rx_over_errors;
	unsigned long	rx_crc_errors;
	unsigned long	rx_frame_errors;
	unsigned long	rx_fifo_errors;
	unsigned long	rx_missed_errors;
L
Linus Torvalds 已提交
180 181 182 183 184 185 186 187 188 189 190 191 192
	unsigned long	tx_aborted_errors;
	unsigned long	tx_carrier_errors;
	unsigned long	tx_fifo_errors;
	unsigned long	tx_heartbeat_errors;
	unsigned long	tx_window_errors;
	unsigned long	rx_compressed;
	unsigned long	tx_compressed;
};


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

193
#ifdef CONFIG_RPS
194
#include <linux/static_key.h>
195 196
extern struct static_key_false rps_needed;
extern struct static_key_false rfs_needed;
197 198
#endif

L
Linus Torvalds 已提交
199 200 201 202
struct neighbour;
struct neigh_parms;
struct sk_buff;

203 204 205 206
struct netdev_hw_addr {
	struct list_head	list;
	unsigned char		addr[MAX_ADDR_LEN];
	unsigned char		type;
J
Jiri Pirko 已提交
207 208 209 210
#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
211 212
#define NETDEV_HW_ADDR_T_MULTICAST	5
	bool			global_use;
213
	int			sync_cnt;
214
	int			refcount;
215
	int			synced;
216 217 218
	struct rcu_head		rcu_head;
};

219 220 221 222 223
struct netdev_hw_addr_list {
	struct list_head	list;
	int			count;
};

224 225 226 227
#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)
228

229 230 231 232
#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)
233

234 235
#define netdev_mc_count(dev) netdev_hw_addr_list_count(&(dev)->mc)
#define netdev_mc_empty(dev) netdev_hw_addr_list_empty(&(dev)->mc)
236
#define netdev_for_each_mc_addr(ha, dev) \
237
	netdev_hw_addr_list_for_each(ha, &(dev)->mc)
238

E
Eric Dumazet 已提交
239
struct hh_cache {
240
	unsigned int	hh_len;
241
	seqlock_t	hh_lock;
L
Linus Torvalds 已提交
242 243 244 245

	/* cached hardware header; allow for machine alignment needs.        */
#define HH_DATA_MOD	16
#define HH_DATA_OFF(__len) \
J
Jiri Benc 已提交
246
	(HH_DATA_MOD - (((__len - 1) & (HH_DATA_MOD - 1)) + 1))
L
Linus Torvalds 已提交
247 248 249 250 251
#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)];
};

B
Bjorn Helgaas 已提交
252
/* Reserve HH_DATA_MOD byte-aligned hard_header_len, but at least that much.
L
Linus Torvalds 已提交
253 254 255 256 257 258 259 260
 * Alternative is:
 *   dev->hard_header_len ? (dev->hard_header_len +
 *                           (HH_DATA_MOD - 1)) & ~(HH_DATA_MOD - 1) : 0
 *
 * We could use other alignment values, but we must maintain the
 * relationship HH alignment <= LL alignment.
 */
#define LL_RESERVED_SPACE(dev) \
261
	((((dev)->hard_header_len+(dev)->needed_headroom)&~(HH_DATA_MOD - 1)) + HH_DATA_MOD)
L
Linus Torvalds 已提交
262
#define LL_RESERVED_SPACE_EXTRA(dev,extra) \
263
	((((dev)->hard_header_len+(dev)->needed_headroom+(extra))&~(HH_DATA_MOD - 1)) + HH_DATA_MOD)
L
Linus Torvalds 已提交
264

265 266 267
struct header_ops {
	int	(*create) (struct sk_buff *skb, struct net_device *dev,
			   unsigned short type, const void *daddr,
268
			   const void *saddr, unsigned int len);
269
	int	(*parse)(const struct sk_buff *skb, unsigned char *haddr);
270
	int	(*cache)(const struct neighbour *neigh, struct hh_cache *hh, __be16 type);
271 272 273
	void	(*cache_update)(struct hh_cache *hh,
				const struct net_device *dev,
				const unsigned char *haddr);
274
	bool	(*validate)(const char *ll_header, unsigned int len);
275
	__be16	(*parse_protocol)(const struct sk_buff *skb);
276 277
};

L
Linus Torvalds 已提交
278
/* These flag bits are private to the generic network queueing
B
Bjorn Helgaas 已提交
279
 * layer; they may not be explicitly referenced by any other
L
Linus Torvalds 已提交
280 281 282
 * code.
 */

E
Eric Dumazet 已提交
283
enum netdev_state_t {
L
Linus Torvalds 已提交
284 285 286
	__LINK_STATE_START,
	__LINK_STATE_PRESENT,
	__LINK_STATE_NOCARRIER,
S
Stefan Rompf 已提交
287 288
	__LINK_STATE_LINKWATCH_PENDING,
	__LINK_STATE_DORMANT,
L
Linus Torvalds 已提交
289 290 291 292
};


/*
B
Bjorn Helgaas 已提交
293
 * This structure holds boot-time configured netdevice settings. They
G
Graf Yang 已提交
294
 * are then used in the device probing.
L
Linus Torvalds 已提交
295 296 297 298 299 300 301
 */
struct netdev_boot_setup {
	char name[IFNAMSIZ];
	struct ifmap map;
};
#define NETDEV_BOOT_SETUP_MAX 8

302
int __init netdev_boot_setup(char *str);
L
Linus Torvalds 已提交
303

304 305 306 307 308
struct gro_list {
	struct list_head	list;
	int			count;
};

309
/*
L
Li RongQing 已提交
310 311
 * size of gro hash buckets, must less than bit number of
 * napi_struct::gro_bitmask
312
 */
313
#define GRO_HASH_BUCKETS	8
L
Li RongQing 已提交
314 315 316 317

/*
 * Structure for NAPI scheduling similar to tasklet but with weighting
 */
318 319 320 321
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
B
Bjorn Helgaas 已提交
322
	 * to the per-CPU poll_list, and whoever clears that bit
323 324 325 326 327 328
	 * can remove from the list right before clearing the bit.
	 */
	struct list_head	poll_list;

	unsigned long		state;
	int			weight;
L
Li RongQing 已提交
329
	unsigned long		gro_bitmask;
330 331 332 333
	int			(*poll)(struct napi_struct *, int);
#ifdef CONFIG_NETPOLL
	int			poll_owner;
#endif
H
Herbert Xu 已提交
334
	struct net_device	*dev;
335
	struct gro_list		gro_hash[GRO_HASH_BUCKETS];
H
Herbert Xu 已提交
336
	struct sk_buff		*skb;
337 338
	struct list_head	rx_list; /* Pending GRO_NORMAL skbs */
	int			rx_count; /* length of rx_list */
339
	struct hrtimer		timer;
340
	struct list_head	dev_list;
E
Eliezer Tamir 已提交
341 342
	struct hlist_node	napi_hash_node;
	unsigned int		napi_id;
343 344
};

E
Eric Dumazet 已提交
345
enum {
346
	NAPI_STATE_SCHED,	/* Poll is scheduled */
E
Eric Dumazet 已提交
347
	NAPI_STATE_MISSED,	/* reschedule a napi */
D
David S. Miller 已提交
348
	NAPI_STATE_DISABLE,	/* Disable pending */
349
	NAPI_STATE_NPSVC,	/* Netpoll - don't dequeue from poll_list */
E
Eric Dumazet 已提交
350 351
	NAPI_STATE_HASHED,	/* In NAPI hash (busy polling possible) */
	NAPI_STATE_NO_BUSY_POLL,/* Do not add in napi_hash, no busy polling */
352 353 354 355
	NAPI_STATE_IN_BUSY_POLL,/* sk_busy_loop() owns this NAPI */
};

enum {
E
Eric Dumazet 已提交
356 357 358 359 360 361 362
	NAPIF_STATE_SCHED	 = BIT(NAPI_STATE_SCHED),
	NAPIF_STATE_MISSED	 = BIT(NAPI_STATE_MISSED),
	NAPIF_STATE_DISABLE	 = BIT(NAPI_STATE_DISABLE),
	NAPIF_STATE_NPSVC	 = BIT(NAPI_STATE_NPSVC),
	NAPIF_STATE_HASHED	 = BIT(NAPI_STATE_HASHED),
	NAPIF_STATE_NO_BUSY_POLL = BIT(NAPI_STATE_NO_BUSY_POLL),
	NAPIF_STATE_IN_BUSY_POLL = BIT(NAPI_STATE_IN_BUSY_POLL),
363 364
};

365
enum gro_result {
366 367 368 369 370
	GRO_MERGED,
	GRO_MERGED_FREE,
	GRO_HELD,
	GRO_NORMAL,
	GRO_DROP,
371
	GRO_CONSUMED,
372
};
373
typedef enum gro_result gro_result_t;
374

375 376 377 378 379 380 381
/*
 * 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.
B
Bjorn Helgaas 已提交
382
 * @RX_HANDLER_PASS: Do nothing, pass the skb as if no rx_handler was called.
383 384 385 386 387 388 389 390 391 392 393 394 395 396
 *
 * 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.
 *
B
Bjorn Helgaas 已提交
397
 * If the rx_handler consumed the skb in some way, it should return
398
 * RX_HANDLER_CONSUMED. This is appropriate when the rx_handler arranged for
B
Bjorn Helgaas 已提交
399
 * the skb to be delivered in some other way.
400 401 402 403 404
 *
 * 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.
 *
B
Bjorn Helgaas 已提交
405
 * If the rx_handler decides the skb should be ignored, it should return
406
 * RX_HANDLER_EXACT. The skb will only be delivered to protocol handlers that
407
 * are registered on exact device (ptype->dev == skb->dev).
408
 *
B
Bjorn Helgaas 已提交
409
 * If the rx_handler didn't change skb->dev, but wants the skb to be normally
410 411 412 413 414 415 416 417 418 419 420 421 422 423
 * 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);
424

425
void __napi_schedule(struct napi_struct *n);
426
void __napi_schedule_irqoff(struct napi_struct *n);
427

428
static inline bool napi_disable_pending(struct napi_struct *n)
D
David S. Miller 已提交
429 430 431 432
{
	return test_bit(NAPI_STATE_DISABLE, &n->state);
}

E
Eric Dumazet 已提交
433
bool napi_schedule_prep(struct napi_struct *n);
434 435 436

/**
 *	napi_schedule - schedule NAPI poll
B
Bjorn Helgaas 已提交
437
 *	@n: NAPI context
438 439 440 441 442 443 444 445 446 447
 *
 * 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);
}

448 449
/**
 *	napi_schedule_irqoff - schedule NAPI poll
B
Bjorn Helgaas 已提交
450
 *	@n: NAPI context
451 452 453 454 455 456 457 458 459
 *
 * Variant of napi_schedule(), assuming hard irqs are masked.
 */
static inline void napi_schedule_irqoff(struct napi_struct *n)
{
	if (napi_schedule_prep(n))
		__napi_schedule_irqoff(n);
}

460
/* Try to reschedule poll. Called by dev->poll() after napi_complete().  */
461
static inline bool napi_reschedule(struct napi_struct *napi)
462 463 464
{
	if (napi_schedule_prep(napi)) {
		__napi_schedule(napi);
465
		return true;
466
	}
467
	return false;
468 469
}

470
bool napi_complete_done(struct napi_struct *n, int work_done);
471 472
/**
 *	napi_complete - NAPI processing complete
B
Bjorn Helgaas 已提交
473
 *	@n: NAPI context
474 475
 *
 * Mark NAPI processing as complete.
476
 * Consider using napi_complete_done() instead.
477
 * Return false if device should avoid rearming interrupts.
478
 */
479
static inline bool napi_complete(struct napi_struct *n)
480 481 482
{
	return napi_complete_done(n, 0);
}
483

E
Eliezer Tamir 已提交
484 485
/**
 *	napi_hash_del - remove a NAPI from global table
B
Bjorn Helgaas 已提交
486
 *	@napi: NAPI context
E
Eliezer Tamir 已提交
487
 *
B
Bjorn Helgaas 已提交
488
 * Warning: caller must observe RCU grace period
489 490
 * before freeing memory containing @napi, if
 * this function returns true.
491
 * Note: core networking stack automatically calls it
B
Bjorn Helgaas 已提交
492
 * from netif_napi_del().
493
 * Drivers might want to call this helper to combine all
B
Bjorn Helgaas 已提交
494
 * the needed RCU grace periods into a single one.
E
Eliezer Tamir 已提交
495
 */
496
bool napi_hash_del(struct napi_struct *napi);
E
Eliezer Tamir 已提交
497

498 499
/**
 *	napi_disable - prevent NAPI from scheduling
B
Bjorn Helgaas 已提交
500
 *	@n: NAPI context
501 502 503 504
 *
 * Stop NAPI from being scheduled on this context.
 * Waits till any outstanding processing completes.
 */
505
void napi_disable(struct napi_struct *n);
506 507 508

/**
 *	napi_enable - enable NAPI scheduling
B
Bjorn Helgaas 已提交
509
 *	@n: NAPI context
510 511 512 513 514 515 516
 *
 * 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));
517
	smp_mb__before_atomic();
518
	clear_bit(NAPI_STATE_SCHED, &n->state);
519
	clear_bit(NAPI_STATE_NPSVC, &n->state);
520 521
}

522 523
/**
 *	napi_synchronize - wait until NAPI is not running
B
Bjorn Helgaas 已提交
524
 *	@n: NAPI context
525 526 527 528 529 530 531
 *
 * 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)
{
532 533 534 535 536
	if (IS_ENABLED(CONFIG_SMP))
		while (test_bit(NAPI_STATE_SCHED, &n->state))
			msleep(1);
	else
		barrier();
537 538
}

539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564
/**
 *	napi_if_scheduled_mark_missed - if napi is running, set the
 *	NAPIF_STATE_MISSED
 *	@n: NAPI context
 *
 * If napi is running, set the NAPIF_STATE_MISSED, and return true if
 * NAPI is scheduled.
 **/
static inline bool napi_if_scheduled_mark_missed(struct napi_struct *n)
{
	unsigned long val, new;

	do {
		val = READ_ONCE(n->state);
		if (val & NAPIF_STATE_DISABLE)
			return true;

		if (!(val & NAPIF_STATE_SCHED))
			return false;

		new = val | NAPIF_STATE_MISSED;
	} while (cmpxchg(&n->state, val, new) != val);

	return true;
}

E
Eric Dumazet 已提交
565
enum netdev_queue_state_t {
566 567
	__QUEUE_STATE_DRV_XOFF,
	__QUEUE_STATE_STACK_XOFF,
568
	__QUEUE_STATE_FROZEN,
569
};
570 571 572 573 574 575 576 577 578 579 580

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

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

581 582 583 584 585 586 587 588 589
/*
 * __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_*.
 */
590

591
struct netdev_queue {
592
/*
B
Bjorn Helgaas 已提交
593
 * read-mostly part
594
 */
595
	struct net_device	*dev;
596
	struct Qdisc __rcu	*qdisc;
597
	struct Qdisc		*qdisc_sleeping;
598
#ifdef CONFIG_SYSFS
T
Tom Herbert 已提交
599 600
	struct kobject		kobj;
#endif
601 602 603
#if defined(CONFIG_XPS) && defined(CONFIG_NUMA)
	int			numa_node;
#endif
604 605 606 607 608 609
	unsigned long		tx_maxrate;
	/*
	 * Number of TX timeouts for this queue
	 * (/sys/class/net/DEV/Q/trans_timeout)
	 */
	unsigned long		trans_timeout;
610 611 612

	/* Subordinate device that the queue has been assigned to */
	struct net_device	*sb_dev;
613 614 615
#ifdef CONFIG_XDP_SOCKETS
	struct xdp_umem         *umem;
#endif
616
/*
B
Bjorn Helgaas 已提交
617
 * write-mostly part
618 619 620
 */
	spinlock_t		_xmit_lock ____cacheline_aligned_in_smp;
	int			xmit_lock_owner;
621
	/*
F
Florian Westphal 已提交
622
	 * Time (in jiffies) of last Tx
623 624
	 */
	unsigned long		trans_start;
625

T
Tom Herbert 已提交
626 627 628 629 630
	unsigned long		state;

#ifdef CONFIG_BQL
	struct dql		dql;
#endif
631
} ____cacheline_aligned_in_smp;
632

633
extern int sysctl_fb_tunnels_only_for_init_net;
634
extern int sysctl_devconf_inherit_init_net;
635 636 637

static inline bool net_has_fallback_tunnels(const struct net *net)
{
638 639 640
	return net == &init_net ||
	       !IS_ENABLED(CONFIG_SYSCTL) ||
	       !sysctl_fb_tunnels_only_for_init_net;
641 642
}

643 644 645 646 647
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
648
	return NUMA_NO_NODE;
649 650 651 652 653 654 655 656 657 658
#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
659
#ifdef CONFIG_RPS
T
Tom Herbert 已提交
660 661 662 663 664 665 666 667 668
/*
 * 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];
};
669
#define RPS_MAP_SIZE(_num) (sizeof(struct rps_map) + ((_num) * sizeof(u16)))
T
Tom Herbert 已提交
670

T
Tom Herbert 已提交
671
/*
672 673 674
 * 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
675 676 677
 */
struct rps_dev_flow {
	u16 cpu;
678
	u16 filter;
T
Tom Herbert 已提交
679 680
	unsigned int last_qtail;
};
681
#define RPS_NO_FILTER 0xffff
T
Tom Herbert 已提交
682 683 684 685 686 687 688 689 690 691

/*
 * The rps_dev_flow_table structure contains a table of flow mappings.
 */
struct rps_dev_flow_table {
	unsigned int mask;
	struct rcu_head rcu;
	struct rps_dev_flow flows[0];
};
#define RPS_DEV_FLOW_TABLE_SIZE(_num) (sizeof(struct rps_dev_flow_table) + \
692
    ((_num) * sizeof(struct rps_dev_flow)))
T
Tom Herbert 已提交
693 694 695 696

/*
 * The rps_sock_flow_table contains mappings of flows to the last CPU
 * on which they were processed by the application (set in recvmsg).
B
Bjorn Helgaas 已提交
697 698
 * Each entry is a 32bit value. Upper part is the high-order bits
 * of flow hash, lower part is CPU number.
699
 * rps_cpu_mask is used to partition the space, depending on number of
B
Bjorn Helgaas 已提交
700 701
 * possible CPUs : rps_cpu_mask = roundup_pow_of_two(nr_cpu_ids) - 1
 * For example, if 64 CPUs are possible, rps_cpu_mask = 0x3f,
702
 * meaning we use 32-6=26 bits for the hash.
T
Tom Herbert 已提交
703 704
 */
struct rps_sock_flow_table {
705
	u32	mask;
706 707

	u32	ents[0] ____cacheline_aligned_in_smp;
T
Tom Herbert 已提交
708
};
709
#define	RPS_SOCK_FLOW_TABLE_SIZE(_num) (offsetof(struct rps_sock_flow_table, ents[_num]))
T
Tom Herbert 已提交
710 711 712

#define RPS_NO_CPU 0xffff

713 714 715
extern u32 rps_cpu_mask;
extern struct rps_sock_flow_table __rcu *rps_sock_flow_table;

T
Tom Herbert 已提交
716 717 718 719
static inline void rps_record_sock_flow(struct rps_sock_flow_table *table,
					u32 hash)
{
	if (table && hash) {
720 721
		unsigned int index = hash & table->mask;
		u32 val = hash & ~rps_cpu_mask;
T
Tom Herbert 已提交
722

B
Bjorn Helgaas 已提交
723
		/* We only give a hint, preemption can change CPU under us */
724
		val |= raw_smp_processor_id();
T
Tom Herbert 已提交
725

726 727
		if (table->ents[index] != val)
			table->ents[index] = val;
T
Tom Herbert 已提交
728 729 730
	}
}

731
#ifdef CONFIG_RFS_ACCEL
732 733
bool rps_may_expire_flow(struct net_device *dev, u16 rxq_index, u32 flow_id,
			 u16 filter_id);
734
#endif
735
#endif /* CONFIG_RPS */
736

T
Tom Herbert 已提交
737 738
/* This structure contains an instance of an RX queue. */
struct netdev_rx_queue {
739
#ifdef CONFIG_RPS
E
Eric Dumazet 已提交
740 741
	struct rps_map __rcu		*rps_map;
	struct rps_dev_flow_table __rcu	*rps_flow_table;
742
#endif
E
Eric Dumazet 已提交
743
	struct kobject			kobj;
T
Tom Herbert 已提交
744
	struct net_device		*dev;
745
	struct xdp_rxq_info		xdp_rxq;
746 747 748
#ifdef CONFIG_XDP_SOCKETS
	struct xdp_umem                 *umem;
#endif
T
Tom Herbert 已提交
749
} ____cacheline_aligned_in_smp;
750 751 752 753 754 755

/*
 * RX queue sysfs structures and functions.
 */
struct rx_queue_attribute {
	struct attribute attr;
756
	ssize_t (*show)(struct netdev_rx_queue *queue, char *buf);
757
	ssize_t (*store)(struct netdev_rx_queue *queue,
758
			 const char *buf, size_t len);
759
};
760

T
Tom Herbert 已提交
761 762 763 764 765 766 767 768 769 770 771
#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];
};
772
#define XPS_MAP_SIZE(_num) (sizeof(struct xps_map) + ((_num) * sizeof(u16)))
773 774
#define XPS_MIN_MAP_ALLOC ((L1_CACHE_ALIGN(offsetof(struct xps_map, queues[1])) \
       - sizeof(struct xps_map)) / sizeof(u16))
T
Tom Herbert 已提交
775 776 777 778 779 780

/*
 * This structure holds all XPS maps for device.  Maps are indexed by CPU.
 */
struct xps_dev_maps {
	struct rcu_head rcu;
781
	struct xps_map __rcu *attr_map[0]; /* Either CPUs map or RXQs map */
T
Tom Herbert 已提交
782
};
783 784

#define XPS_CPU_DEV_MAPS_SIZE(_tcs) (sizeof(struct xps_dev_maps) +	\
785
	(nr_cpu_ids * (_tcs) * sizeof(struct xps_map *)))
786 787 788 789

#define XPS_RXQ_DEV_MAPS_SIZE(_tcs, _rxqs) (sizeof(struct xps_dev_maps) +\
	(_rxqs * (_tcs) * sizeof(struct xps_map *)))

T
Tom Herbert 已提交
790 791
#endif /* CONFIG_XPS */

792 793 794 795 796 797 798 799
#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;
};

800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816
#if defined(CONFIG_FCOE) || defined(CONFIG_FCOE_MODULE)
/*
 * This structure is to hold information about the device
 * configured to run FCoE protocol stack.
 */
struct netdev_fcoe_hbainfo {
	char	manufacturer[64];
	char	serial_number[64];
	char	hardware_version[64];
	char	driver_version[64];
	char	optionrom_version[64];
	char	firmware_version[64];
	char	model[256];
	char	model_description[256];
};
#endif

817
#define MAX_PHYS_ITEM_ID_LEN 32
818

819 820
/* This structure holds a unique identifier to identify some
 * physical item (port for example) used by a netdevice.
821
 */
822 823
struct netdev_phys_item_id {
	unsigned char id[MAX_PHYS_ITEM_ID_LEN];
824 825 826
	unsigned char id_len;
};

827 828 829 830 831 832 833
static inline bool netdev_phys_item_id_same(struct netdev_phys_item_id *a,
					    struct netdev_phys_item_id *b)
{
	return a->id_len == b->id_len &&
	       memcmp(a->id, b->id, a->id_len) == 0;
}

834
typedef u16 (*select_queue_fallback_t)(struct net_device *dev,
835 836
				       struct sk_buff *skb,
				       struct net_device *sb_dev);
837

838
enum tc_setup_type {
839
	TC_SETUP_QDISC_MQPRIO,
840
	TC_SETUP_CLSU32,
841
	TC_SETUP_CLSFLOWER,
842
	TC_SETUP_CLSMATCHALL,
843
	TC_SETUP_CLSBPF,
844
	TC_SETUP_BLOCK,
845
	TC_SETUP_QDISC_CBS,
846
	TC_SETUP_QDISC_RED,
847
	TC_SETUP_QDISC_PRIO,
848
	TC_SETUP_QDISC_MQ,
849
	TC_SETUP_QDISC_ETF,
850
	TC_SETUP_ROOT_QDISC,
851
	TC_SETUP_QDISC_GRED,
852
	TC_SETUP_QDISC_TAPRIO,
853
	TC_SETUP_FT,
854
	TC_SETUP_QDISC_ETS,
855
	TC_SETUP_QDISC_TBF,
856 857
};

858 859
/* These structures hold the attributes of bpf state that are being passed
 * to the netdevice through the bpf op.
860
 */
861
enum bpf_netdev_command {
862 863 864 865 866 867 868 869
	/* Set or clear a bpf program used in the earliest stages of packet
	 * rx. The prog will have been loaded as BPF_PROG_TYPE_XDP. The callee
	 * is responsible for calling bpf_prog_put on any old progs that are
	 * stored. In case of error, the callee need not release the new prog
	 * reference, but on success it takes ownership and must bpf_prog_put
	 * when it is no longer used.
	 */
	XDP_SETUP_PROG,
870
	XDP_SETUP_PROG_HW,
871
	XDP_QUERY_PROG,
872
	XDP_QUERY_PROG_HW,
873
	/* BPF program for offload callbacks, invoked at program load time. */
874 875
	BPF_OFFLOAD_MAP_ALLOC,
	BPF_OFFLOAD_MAP_FREE,
876
	XDP_SETUP_XSK_UMEM,
877 878
};

879
struct bpf_prog_offload_ops;
880
struct netlink_ext_ack;
881
struct xdp_umem;
882
struct xdp_dev_bulk_queue;
883

884 885
struct netdev_bpf {
	enum bpf_netdev_command command;
886 887
	union {
		/* XDP_SETUP_PROG */
888
		struct {
889
			u32 flags;
890 891 892
			struct bpf_prog *prog;
			struct netlink_ext_ack *extack;
		};
893
		/* XDP_QUERY_PROG, XDP_QUERY_PROG_HW */
M
Martin KaFai Lau 已提交
894 895
		struct {
			u32 prog_id;
896 897
			/* flags with which program was installed */
			u32 prog_flags;
M
Martin KaFai Lau 已提交
898
		};
899 900 901 902
		/* BPF_OFFLOAD_MAP_ALLOC, BPF_OFFLOAD_MAP_FREE */
		struct {
			struct bpf_offloaded_map *offmap;
		};
903
		/* XDP_SETUP_XSK_UMEM */
904
		struct {
905 906
			struct xdp_umem *umem;
			u16 queue_id;
907
		} xsk;
908 909
	};
};
910

911 912 913 914
/* Flags for ndo_xsk_wakeup. */
#define XDP_WAKEUP_RX (1 << 0)
#define XDP_WAKEUP_TX (1 << 1)

915 916 917 918 919 920 921
#ifdef CONFIG_XFRM_OFFLOAD
struct xfrmdev_ops {
	int	(*xdo_dev_state_add) (struct xfrm_state *x);
	void	(*xdo_dev_state_delete) (struct xfrm_state *x);
	void	(*xdo_dev_state_free) (struct xfrm_state *x);
	bool	(*xdo_dev_offload_ok) (struct sk_buff *skb,
				       struct xfrm_state *x);
922
	void	(*xdo_dev_state_advance_esn) (struct xfrm_state *x);
923 924 925
};
#endif

926 927 928 929 930
struct dev_ifalias {
	struct rcu_head rcuhead;
	char ifalias[];
};

931
struct devlink;
932
struct tlsdev_ops;
933

934 935
struct netdev_name_node {
	struct hlist_node hlist;
936
	struct list_head list;
937 938 939 940
	struct net_device *dev;
	const char *name;
};

941 942 943
int netdev_name_node_alt_create(struct net_device *dev, const char *name);
int netdev_name_node_alt_destroy(struct net_device *dev, const char *name);

944 945 946 947 948
struct netdev_net_notifier {
	struct list_head list;
	struct notifier_block *nb;
};

949 950
/*
 * This structure defines the management hooks for network devices.
951 952
 * The following hooks can be defined; unless noted otherwise, they are
 * optional and can be filled with a null pointer.
953 954
 *
 * int (*ndo_init)(struct net_device *dev);
B
Bjorn Helgaas 已提交
955 956 957 958
 *     This function is called once when a network device is registered.
 *     The network device can use this for any late stage initialization
 *     or semantic validation. It can fail with an error code which will
 *     be propagated back to register_netdev.
959 960 961 962 963 964
 *
 * 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);
B
Bjorn Helgaas 已提交
965
 *     This function is called when a network device transitions to the up
966 967 968
 *     state.
 *
 * int (*ndo_stop)(struct net_device *dev);
B
Bjorn Helgaas 已提交
969
 *     This function is called when a network device transitions to the down
970 971
 *     state.
 *
972 973
 * netdev_tx_t (*ndo_start_xmit)(struct sk_buff *skb,
 *                               struct net_device *dev);
974
 *	Called when a packet needs to be transmitted.
975 976 977 978
 *	Returns NETDEV_TX_OK.  Can return NETDEV_TX_BUSY, but you should stop
 *	the queue before that can happen; it's for obsolete devices and weird
 *	corner cases, but the stack really does a non-trivial amount
 *	of useless work if you return NETDEV_TX_BUSY.
B
Bjorn Helgaas 已提交
979
 *	Required; cannot be NULL.
980
 *
981 982 983 984 985 986 987 988 989
 * netdev_features_t (*ndo_features_check)(struct sk_buff *skb,
 *					   struct net_device *dev
 *					   netdev_features_t features);
 *	Called by core transmit path to determine if device is capable of
 *	performing offload operations on a given packet. This is to give
 *	the device an opportunity to implement any restrictions that cannot
 *	be otherwise expressed by feature flags. The check is called with
 *	the set of features that the stack has calculated and it returns
 *	those the driver believes to be appropriate.
990
 *
991
 * u16 (*ndo_select_queue)(struct net_device *dev, struct sk_buff *skb,
992
 *                         struct net_device *sb_dev);
B
Bjorn Helgaas 已提交
993
 *	Called to decide which queue to use when device supports multiple
994 995
 *	transmit queues.
 *
996 997
 * void (*ndo_change_rx_flags)(struct net_device *dev, int flags);
 *	This function is called to allow device receiver to make
B
Bjorn Helgaas 已提交
998
 *	changes to configuration when multicast or promiscuous is enabled.
999 1000 1001
 *
 * void (*ndo_set_rx_mode)(struct net_device *dev);
 *	This function is called device changes address list filtering.
1002
 *	If driver handles unicast address filtering, it should set
B
Bjorn Helgaas 已提交
1003
 *	IFF_UNICAST_FLT in its priv_flags.
1004 1005 1006
 *
 * int (*ndo_set_mac_address)(struct net_device *dev, void *addr);
 *	This function  is called when the Media Access Control address
1007
 *	needs to be changed. If this interface is not defined, the
B
Bjorn Helgaas 已提交
1008
 *	MAC address can not be changed.
1009 1010 1011 1012 1013
 *
 * 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);
B
Bjorn Helgaas 已提交
1014 1015
 *	Called when a user requests an ioctl which can't be handled by
 *	the generic interface code. If not defined ioctls return
1016 1017 1018 1019
 *	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
B
Bjorn Helgaas 已提交
1020
 *	is retained for legacy reasons; new devices should use the bus
1021 1022 1023 1024
 *	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
1025
 *	of a device.
1026
 *
1027
 * void (*ndo_tx_timeout)(struct net_device *dev, unsigned int txqueue);
B
Bjorn Helgaas 已提交
1028
 *	Callback used when the transmitter has not made any progress
1029 1030
 *	for dev->watchdog ticks.
 *
1031 1032
 * void (*ndo_get_stats64)(struct net_device *dev,
 *                         struct rtnl_link_stats64 *storage);
1033
 * struct net_device_stats* (*ndo_get_stats)(struct net_device *dev);
1034
 *	Called when a user wants to get the network device usage
1035
 *	statistics. Drivers must do one of the following:
1036 1037
 *	1. Define @ndo_get_stats64 to fill in a zero-initialised
 *	   rtnl_link_stats64 structure passed by the caller.
1038
 *	2. Define @ndo_get_stats to update a net_device_stats structure
1039 1040 1041 1042 1043
 *	   (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.
1044
 *
1045
 * bool (*ndo_has_offload_stats)(const struct net_device *dev, int attr_id)
1046 1047 1048 1049 1050 1051 1052
 *	Return true if this device supports offload stats of this attr_id.
 *
 * int (*ndo_get_offload_stats)(int attr_id, const struct net_device *dev,
 *	void *attr_data)
 *	Get statistics for offload operations by attr_id. Write it into the
 *	attr_data pointer.
 *
1053
 * int (*ndo_vlan_rx_add_vid)(struct net_device *dev, __be16 proto, u16 vid);
B
Bjorn Helgaas 已提交
1054
 *	If device supports VLAN filtering this function is called when a
1055
 *	VLAN id is registered.
1056
 *
1057
 * int (*ndo_vlan_rx_kill_vid)(struct net_device *dev, __be16 proto, u16 vid);
B
Bjorn Helgaas 已提交
1058
 *	If device supports VLAN filtering this function is called when a
1059
 *	VLAN id is unregistered.
1060 1061
 *
 * void (*ndo_poll_controller)(struct net_device *dev);
1062 1063 1064
 *
 *	SR-IOV management functions.
 * int (*ndo_set_vf_mac)(struct net_device *dev, int vf, u8* mac);
1065 1066
 * int (*ndo_set_vf_vlan)(struct net_device *dev, int vf, u16 vlan,
 *			  u8 qos, __be16 proto);
1067 1068
 * int (*ndo_set_vf_rate)(struct net_device *dev, int vf, int min_tx_rate,
 *			  int max_tx_rate);
1069
 * int (*ndo_set_vf_spoofchk)(struct net_device *dev, int vf, bool setting);
H
Hiroshi Shimamoto 已提交
1070
 * int (*ndo_set_vf_trust)(struct net_device *dev, int vf, bool setting);
1071 1072
 * int (*ndo_get_vf_config)(struct net_device *dev,
 *			    int vf, struct ifla_vf_info *ivf);
1073
 * int (*ndo_set_vf_link_state)(struct net_device *dev, int vf, int link_state);
1074 1075
 * int (*ndo_set_vf_port)(struct net_device *dev, int vf,
 *			  struct nlattr *port[]);
1076 1077 1078
 *
 *      Enable or disable the VF ability to query its RSS Redirection Table and
 *      Hash Key. This is needed since on some devices VF share this information
B
Bjorn Helgaas 已提交
1079
 *      with PF and querying it may introduce a theoretical security risk.
1080
 * int (*ndo_set_vf_rss_query_en)(struct net_device *dev, int vf, bool setting);
1081
 * int (*ndo_get_vf_port)(struct net_device *dev, int vf, struct sk_buff *skb);
1082
 * int (*ndo_setup_tc)(struct net_device *dev, enum tc_setup_type type,
1083
 *		       void *type_data);
F
Florian Fainelli 已提交
1084 1085 1086 1087
 *	Called to setup any 'tc' scheduler, classifier or action on @dev.
 *	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.
1088
 *
1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118
 *	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.
 *
1119 1120 1121 1122 1123 1124 1125
 * int (*ndo_fcoe_get_hbainfo)(struct net_device *dev,
 *			       struct netdev_fcoe_hbainfo *hbainfo);
 *	Called when the FCoE Protocol stack wants information on the underlying
 *	device. This information is utilized by the FCoE protocol stack to
 *	register attributes with Fiber Channel management service as per the
 *	FC-GS Fabric Device Management Information(FDMI) specification.
 *
1126 1127 1128 1129 1130 1131
 * 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.
 *
1132 1133 1134 1135 1136 1137
 *	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.
1138
 *
1139
 *	Slave management functions (for bridge, bonding, etc).
1140 1141 1142 1143 1144
 * 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.
1145 1146
 *
 *      Feature/offload setting functions.
1147 1148 1149 1150 1151 1152
 * netdev_features_t (*ndo_fix_features)(struct net_device *dev,
 *		netdev_features_t features);
 *	Adjusts the requested feature flags according to device-specific
 *	constraints, and returns the resulting flags. Must not modify
 *	the device state.
 *
1153
 * int (*ndo_set_features)(struct net_device *dev, netdev_features_t features);
1154 1155 1156 1157
 *	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.
 *
1158 1159
 * int (*ndo_fdb_add)(struct ndmsg *ndm, struct nlattr *tb[],
 *		      struct net_device *dev,
1160 1161
 *		      const unsigned char *addr, u16 vid, u16 flags,
 *		      struct netlink_ext_ack *extack);
1162
 *	Adds an FDB entry to dev for addr.
1163 1164
 * int (*ndo_fdb_del)(struct ndmsg *ndm, struct nlattr *tb[],
 *		      struct net_device *dev,
1165
 *		      const unsigned char *addr, u16 vid)
1166 1167
 *	Deletes the FDB entry from dev coresponding to addr.
 * int (*ndo_fdb_dump)(struct sk_buff *skb, struct netlink_callback *cb,
1168
 *		       struct net_device *dev, struct net_device *filter_dev,
1169
 *		       int *idx)
1170 1171
 *	Used to add FDB entries to dump requests. Implementers should add
 *	entries to skb and update idx with the number of entries.
J
John Fastabend 已提交
1172
 *
1173
 * int (*ndo_bridge_setlink)(struct net_device *dev, struct nlmsghdr *nlh,
P
Petr Machata 已提交
1174
 *			     u16 flags, struct netlink_ext_ack *extack)
J
John Fastabend 已提交
1175
 * int (*ndo_bridge_getlink)(struct sk_buff *skb, u32 pid, u32 seq,
1176 1177
 *			     struct net_device *dev, u32 filter_mask,
 *			     int nlflags)
1178 1179
 * int (*ndo_bridge_dellink)(struct net_device *dev, struct nlmsghdr *nlh,
 *			     u16 flags);
J
Jiri Pirko 已提交
1180 1181 1182 1183 1184 1185 1186 1187
 *
 * int (*ndo_change_carrier)(struct net_device *dev, bool new_carrier);
 *	Called to change device carrier. Soft-devices (like dummy, team, etc)
 *	which do not represent real hardware may define this to allow their
 *	userspace components to manage their virtual carrier state. Devices
 *	that determine carrier state from physical hardware properties (eg
 *	network cables) or protocol-dependent mechanisms (eg
 *	USB_CDC_NOTIFY_NETWORK_CONNECTION) should NOT implement this function.
1188 1189
 *
 * int (*ndo_get_phys_port_id)(struct net_device *dev,
1190
 *			       struct netdev_phys_item_id *ppid);
1191 1192 1193
 *	Called to get ID of physical port of this device. If driver does
 *	not implement this, it is assumed that the hw is not able to have
 *	multiple net devices on single physical port.
1194
 *
1195 1196 1197 1198
 * int (*ndo_get_port_parent_id)(struct net_device *dev,
 *				 struct netdev_phys_item_id *ppid)
 *	Called to get the parent ID of the physical port of this device.
 *
1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211
 * void (*ndo_udp_tunnel_add)(struct net_device *dev,
 *			      struct udp_tunnel_info *ti);
 *	Called by UDP tunnel to notify a driver about the UDP port and socket
 *	address family that a UDP tunnel is listnening to. It is called only
 *	when a new port starts listening. The operation is protected by the
 *	RTNL.
 *
 * void (*ndo_udp_tunnel_del)(struct net_device *dev,
 *			      struct udp_tunnel_info *ti);
 *	Called by UDP tunnel to notify the driver about a UDP port and socket
 *	address family that the UDP tunnel is not listening to anymore. The
 *	operation is protected by the RTNL.
 *
1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223
 * void* (*ndo_dfwd_add_station)(struct net_device *pdev,
 *				 struct net_device *dev)
 *	Called by upper layer devices to accelerate switching or other
 *	station functionality into hardware. 'pdev is the lowerdev
 *	to use for the offload and 'dev' is the net device that will
 *	back the offload. Returns a pointer to the private structure
 *	the upper layer will maintain.
 * void (*ndo_dfwd_del_station)(struct net_device *pdev, void *priv)
 *	Called by upper layer device to delete the station created
 *	by 'ndo_dfwd_add_station'. 'pdev' is the net device backing
 *	the station and priv is the structure returned by the add
 *	operation.
1224 1225 1226 1227
 * int (*ndo_set_tx_maxrate)(struct net_device *dev,
 *			     int queue_index, u32 maxrate);
 *	Called when a user wants to set a max-rate limitation of specific
 *	TX queue.
1228 1229
 * int (*ndo_get_iflink)(const struct net_device *dev);
 *	Called to get the iflink value of this device.
1230
 * void (*ndo_change_proto_down)(struct net_device *dev,
B
Bjorn Helgaas 已提交
1231
 *				 bool proto_down);
1232 1233 1234
 *	This function is used to pass protocol port error state information
 *	to the switch driver. The switch driver can react to the proto_down
 *      by doing a phys down on the associated switch port.
1235 1236 1237 1238
 * int (*ndo_fill_metadata_dst)(struct net_device *dev, struct sk_buff *skb);
 *	This function is used to get egress tunnel information for given skb.
 *	This is useful for retrieving outer tunnel header parameters while
 *	sampling packet.
1239 1240 1241 1242
 * void (*ndo_set_rx_headroom)(struct net_device *dev, int needed_headroom);
 *	This function is used to specify the headroom that the skb must
 *	consider when allocation skb during packet reception. Setting
 *	appropriate rx headroom value allows avoiding skb head copy on
B
Bjorn Helgaas 已提交
1243
 *	forward. Setting a negative value resets the rx headroom to the
1244
 *	default value.
1245
 * int (*ndo_bpf)(struct net_device *dev, struct netdev_bpf *bpf);
1246
 *	This function is used to set or query state related to XDP on the
1247 1248
 *	netdevice and manage BPF offload. See definition of
 *	enum bpf_netdev_command for details.
1249 1250
 * int (*ndo_xdp_xmit)(struct net_device *dev, int n, struct xdp_frame **xdp,
 *			u32 flags);
1251 1252 1253 1254 1255
 *	This function is used to submit @n XDP packets for transmit on a
 *	netdevice. Returns number of frames successfully transmitted, frames
 *	that got dropped are freed/returned via xdp_return_frame().
 *	Returns negative number, means general error invoking ndo, meaning
 *	no frames were xmit'ed and core-caller will free all frames.
1256 1257 1258 1259 1260 1261
 * int (*ndo_xsk_wakeup)(struct net_device *dev, u32 queue_id, u32 flags);
 *      This function is used to wake up the softirq, ksoftirqd or kthread
 *	responsible for sending and/or receiving packets on a specific
 *	queue id bound to an AF_XDP socket. The flags field specifies if
 *	only RX, only Tx, or both should be woken up using the flags
 *	XDP_WAKEUP_RX and XDP_WAKEUP_TX.
1262 1263
 * struct devlink_port *(*ndo_get_devlink_port)(struct net_device *dev);
 *	Get devlink port instance associated with a given netdev.
1264 1265
 *	Called with a reference on the netdevice and devlink locks only,
 *	rtnl_lock is not held.
1266 1267 1268 1269 1270 1271
 */
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);
1272 1273 1274 1275 1276
	netdev_tx_t		(*ndo_start_xmit)(struct sk_buff *skb,
						  struct net_device *dev);
	netdev_features_t	(*ndo_features_check)(struct sk_buff *skb,
						      struct net_device *dev,
						      netdev_features_t features);
1277
	u16			(*ndo_select_queue)(struct net_device *dev,
1278
						    struct sk_buff *skb,
1279
						    struct net_device *sb_dev);
1280 1281 1282 1283 1284 1285 1286 1287 1288 1289
	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);
1290 1291 1292 1293
	int			(*ndo_change_mtu)(struct net_device *dev,
						  int new_mtu);
	int			(*ndo_neigh_setup)(struct net_device *dev,
						   struct neigh_parms *);
1294 1295
	void			(*ndo_tx_timeout) (struct net_device *dev,
						   unsigned int txqueue);
1296

1297 1298
	void			(*ndo_get_stats64)(struct net_device *dev,
						   struct rtnl_link_stats64 *storage);
1299
	bool			(*ndo_has_offload_stats)(const struct net_device *dev, int attr_id);
1300 1301 1302
	int			(*ndo_get_offload_stats)(int attr_id,
							 const struct net_device *dev,
							 void *attr_data);
1303 1304
	struct net_device_stats* (*ndo_get_stats)(struct net_device *dev);

1305
	int			(*ndo_vlan_rx_add_vid)(struct net_device *dev,
1306
						       __be16 proto, u16 vid);
1307
	int			(*ndo_vlan_rx_kill_vid)(struct net_device *dev,
1308
						        __be16 proto, u16 vid);
1309 1310
#ifdef CONFIG_NET_POLL_CONTROLLER
	void                    (*ndo_poll_controller)(struct net_device *dev);
H
Herbert Xu 已提交
1311
	int			(*ndo_netpoll_setup)(struct net_device *dev,
1312
						     struct netpoll_info *info);
1313
	void			(*ndo_netpoll_cleanup)(struct net_device *dev);
1314
#endif
1315 1316 1317
	int			(*ndo_set_vf_mac)(struct net_device *dev,
						  int queue, u8 *mac);
	int			(*ndo_set_vf_vlan)(struct net_device *dev,
1318 1319
						   int queue, u16 vlan,
						   u8 qos, __be16 proto);
1320 1321 1322
	int			(*ndo_set_vf_rate)(struct net_device *dev,
						   int vf, int min_tx_rate,
						   int max_tx_rate);
1323 1324
	int			(*ndo_set_vf_spoofchk)(struct net_device *dev,
						       int vf, bool setting);
H
Hiroshi Shimamoto 已提交
1325 1326
	int			(*ndo_set_vf_trust)(struct net_device *dev,
						    int vf, bool setting);
1327 1328 1329
	int			(*ndo_get_vf_config)(struct net_device *dev,
						     int vf,
						     struct ifla_vf_info *ivf);
1330 1331
	int			(*ndo_set_vf_link_state)(struct net_device *dev,
							 int vf, int link_state);
1332 1333 1334 1335
	int			(*ndo_get_vf_stats)(struct net_device *dev,
						    int vf,
						    struct ifla_vf_stats
						    *vf_stats);
1336 1337 1338 1339 1340
	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);
1341 1342 1343 1344
	int			(*ndo_get_vf_guid)(struct net_device *dev,
						   int vf,
						   struct ifla_vf_guid *node_guid,
						   struct ifla_vf_guid *port_guid);
1345 1346 1347
	int			(*ndo_set_vf_guid)(struct net_device *dev,
						   int vf, u64 guid,
						   int guid_type);
1348 1349 1350
	int			(*ndo_set_vf_rss_query_en)(
						   struct net_device *dev,
						   int vf, bool setting);
1351
	int			(*ndo_setup_tc)(struct net_device *dev,
1352
						enum tc_setup_type type,
1353
						void *type_data);
1354
#if IS_ENABLED(CONFIG_FCOE)
1355 1356
	int			(*ndo_fcoe_enable)(struct net_device *dev);
	int			(*ndo_fcoe_disable)(struct net_device *dev);
1357 1358 1359 1360 1361 1362
	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);
1363 1364 1365 1366
	int			(*ndo_fcoe_ddp_target)(struct net_device *dev,
						       u16 xid,
						       struct scatterlist *sgl,
						       unsigned int sgc);
1367 1368
	int			(*ndo_fcoe_get_hbainfo)(struct net_device *dev,
							struct netdev_fcoe_hbainfo *hbainfo);
1369 1370
#endif

1371
#if IS_ENABLED(CONFIG_LIBFCOE)
1372 1373 1374 1375
#define NETDEV_FCOE_WWNN 0
#define NETDEV_FCOE_WWPN 1
	int			(*ndo_fcoe_get_wwn)(struct net_device *dev,
						    u64 *wwn, int type);
1376
#endif
1377

1378 1379 1380 1381 1382 1383
#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
1384
	int			(*ndo_add_slave)(struct net_device *dev,
D
David Ahern 已提交
1385 1386
						 struct net_device *slave_dev,
						 struct netlink_ext_ack *extack);
1387 1388
	int			(*ndo_del_slave)(struct net_device *dev,
						 struct net_device *slave_dev);
1389 1390
	netdev_features_t	(*ndo_fix_features)(struct net_device *dev,
						    netdev_features_t features);
1391
	int			(*ndo_set_features)(struct net_device *dev,
1392
						    netdev_features_t features);
1393 1394 1395 1396
	int			(*ndo_neigh_construct)(struct net_device *dev,
						       struct neighbour *n);
	void			(*ndo_neigh_destroy)(struct net_device *dev,
						     struct neighbour *n);
1397 1398

	int			(*ndo_fdb_add)(struct ndmsg *ndm,
1399
					       struct nlattr *tb[],
1400
					       struct net_device *dev,
1401
					       const unsigned char *addr,
1402
					       u16 vid,
1403 1404
					       u16 flags,
					       struct netlink_ext_ack *extack);
1405
	int			(*ndo_fdb_del)(struct ndmsg *ndm,
1406
					       struct nlattr *tb[],
1407
					       struct net_device *dev,
1408 1409
					       const unsigned char *addr,
					       u16 vid);
1410 1411 1412
	int			(*ndo_fdb_dump)(struct sk_buff *skb,
						struct netlink_callback *cb,
						struct net_device *dev,
1413
						struct net_device *filter_dev,
1414
						int *idx);
R
Roopa Prabhu 已提交
1415 1416 1417 1418 1419 1420
	int			(*ndo_fdb_get)(struct sk_buff *skb,
					       struct nlattr *tb[],
					       struct net_device *dev,
					       const unsigned char *addr,
					       u16 vid, u32 portid, u32 seq,
					       struct netlink_ext_ack *extack);
J
John Fastabend 已提交
1421
	int			(*ndo_bridge_setlink)(struct net_device *dev,
1422
						      struct nlmsghdr *nlh,
P
Petr Machata 已提交
1423 1424
						      u16 flags,
						      struct netlink_ext_ack *extack);
J
John Fastabend 已提交
1425 1426
	int			(*ndo_bridge_getlink)(struct sk_buff *skb,
						      u32 pid, u32 seq,
1427
						      struct net_device *dev,
1428 1429
						      u32 filter_mask,
						      int nlflags);
1430
	int			(*ndo_bridge_dellink)(struct net_device *dev,
1431 1432
						      struct nlmsghdr *nlh,
						      u16 flags);
J
Jiri Pirko 已提交
1433 1434
	int			(*ndo_change_carrier)(struct net_device *dev,
						      bool new_carrier);
1435
	int			(*ndo_get_phys_port_id)(struct net_device *dev,
1436
							struct netdev_phys_item_id *ppid);
1437 1438
	int			(*ndo_get_port_parent_id)(struct net_device *dev,
							  struct netdev_phys_item_id *ppid);
1439 1440
	int			(*ndo_get_phys_port_name)(struct net_device *dev,
							  char *name, size_t len);
1441 1442 1443 1444
	void			(*ndo_udp_tunnel_add)(struct net_device *dev,
						      struct udp_tunnel_info *ti);
	void			(*ndo_udp_tunnel_del)(struct net_device *dev,
						      struct udp_tunnel_info *ti);
1445 1446 1447 1448 1449
	void*			(*ndo_dfwd_add_station)(struct net_device *pdev,
							struct net_device *dev);
	void			(*ndo_dfwd_del_station)(struct net_device *pdev,
							void *priv);

1450 1451 1452
	int			(*ndo_set_tx_maxrate)(struct net_device *dev,
						      int queue_index,
						      u32 maxrate);
1453
	int			(*ndo_get_iflink)(const struct net_device *dev);
1454 1455
	int			(*ndo_change_proto_down)(struct net_device *dev,
							 bool proto_down);
1456 1457
	int			(*ndo_fill_metadata_dst)(struct net_device *dev,
						       struct sk_buff *skb);
1458 1459
	void			(*ndo_set_rx_headroom)(struct net_device *dev,
						       int needed_headroom);
1460 1461
	int			(*ndo_bpf)(struct net_device *dev,
					   struct netdev_bpf *bpf);
1462
	int			(*ndo_xdp_xmit)(struct net_device *dev, int n,
1463 1464
						struct xdp_frame **xdp,
						u32 flags);
1465 1466
	int			(*ndo_xsk_wakeup)(struct net_device *dev,
						  u32 queue_id, u32 flags);
1467
	struct devlink_port *	(*ndo_get_devlink_port)(struct net_device *dev);
1468 1469
};

1470 1471 1472 1473 1474
/**
 * enum net_device_priv_flags - &struct net_device priv_flags
 *
 * These are the &struct net_device, they are only set internally
 * by drivers and used in the kernel. These flags are invisible to
B
Bjorn Helgaas 已提交
1475
 * userspace; this means that the order of these flags can change
1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498
 * during any kernel release.
 *
 * You should have a pretty good reason to be extending these flags.
 *
 * @IFF_802_1Q_VLAN: 802.1Q VLAN device
 * @IFF_EBRIDGE: Ethernet bridging device
 * @IFF_BONDING: bonding master or slave
 * @IFF_ISATAP: ISATAP interface (RFC4214)
 * @IFF_WAN_HDLC: WAN HDLC device
 * @IFF_XMIT_DST_RELEASE: dev_hard_start_xmit() is allowed to
 *	release skb->dst
 * @IFF_DONT_BRIDGE: disallow bridging this ether dev
 * @IFF_DISABLE_NETPOLL: disable netpoll at run-time
 * @IFF_MACVLAN_PORT: device used as macvlan port
 * @IFF_BRIDGE_PORT: device used as bridge port
 * @IFF_OVS_DATAPATH: device used as Open vSwitch datapath port
 * @IFF_TX_SKB_SHARING: The interface supports sharing skbs on transmit
 * @IFF_UNICAST_FLT: Supports unicast filtering
 * @IFF_TEAM_PORT: device used as team port
 * @IFF_SUPP_NOFCS: device supports sending custom FCS
 * @IFF_LIVE_ADDR_CHANGE: device supports hardware address
 *	change when it's running
 * @IFF_MACVLAN: Macvlan device
1499 1500
 * @IFF_XMIT_DST_RELEASE_PERM: IFF_XMIT_DST_RELEASE not taking into account
 *	underlying stacked devices
1501
 * @IFF_L3MDEV_MASTER: device is an L3 master device
1502
 * @IFF_NO_QUEUE: device can run without qdisc attached
1503
 * @IFF_OPENVSWITCH: device is a Open vSwitch master
D
David Ahern 已提交
1504
 * @IFF_L3MDEV_SLAVE: device is enslaved to an L3 master device
J
Jiri Pirko 已提交
1505
 * @IFF_TEAM: device is a team device
1506
 * @IFF_RXFH_CONFIGURED: device has had Rx Flow indirection table configured
1507 1508
 * @IFF_PHONY_HEADROOM: the headroom value is controlled by an external
 *	entity (i.e. the master device for bridged veth)
1509
 * @IFF_MACSEC: device is a MACsec device
P
Paolo Abeni 已提交
1510
 * @IFF_NO_RX_HANDLER: device doesn't support the rx_handler hook
1511 1512
 * @IFF_FAILOVER: device is a failover master device
 * @IFF_FAILOVER_SLAVE: device is lower dev of a failover master device
1513
 * @IFF_L3MDEV_RX_HANDLER: only invoke the rx handler of L3 master device
1514
 * @IFF_LIVE_RENAME_OK: rename is allowed while device is up and running
1515 1516 1517 1518
 */
enum netdev_priv_flags {
	IFF_802_1Q_VLAN			= 1<<0,
	IFF_EBRIDGE			= 1<<1,
1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534
	IFF_BONDING			= 1<<2,
	IFF_ISATAP			= 1<<3,
	IFF_WAN_HDLC			= 1<<4,
	IFF_XMIT_DST_RELEASE		= 1<<5,
	IFF_DONT_BRIDGE			= 1<<6,
	IFF_DISABLE_NETPOLL		= 1<<7,
	IFF_MACVLAN_PORT		= 1<<8,
	IFF_BRIDGE_PORT			= 1<<9,
	IFF_OVS_DATAPATH		= 1<<10,
	IFF_TX_SKB_SHARING		= 1<<11,
	IFF_UNICAST_FLT			= 1<<12,
	IFF_TEAM_PORT			= 1<<13,
	IFF_SUPP_NOFCS			= 1<<14,
	IFF_LIVE_ADDR_CHANGE		= 1<<15,
	IFF_MACVLAN			= 1<<16,
	IFF_XMIT_DST_RELEASE_PERM	= 1<<17,
P
Paolo Abeni 已提交
1535 1536 1537 1538 1539 1540 1541 1542
	IFF_L3MDEV_MASTER		= 1<<18,
	IFF_NO_QUEUE			= 1<<19,
	IFF_OPENVSWITCH			= 1<<20,
	IFF_L3MDEV_SLAVE		= 1<<21,
	IFF_TEAM			= 1<<22,
	IFF_RXFH_CONFIGURED		= 1<<23,
	IFF_PHONY_HEADROOM		= 1<<24,
	IFF_MACSEC			= 1<<25,
P
Paolo Abeni 已提交
1543
	IFF_NO_RX_HANDLER		= 1<<26,
1544 1545
	IFF_FAILOVER			= 1<<27,
	IFF_FAILOVER_SLAVE		= 1<<28,
1546
	IFF_L3MDEV_RX_HANDLER		= 1<<29,
1547
	IFF_LIVE_RENAME_OK		= 1<<30,
1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566
};

#define IFF_802_1Q_VLAN			IFF_802_1Q_VLAN
#define IFF_EBRIDGE			IFF_EBRIDGE
#define IFF_BONDING			IFF_BONDING
#define IFF_ISATAP			IFF_ISATAP
#define IFF_WAN_HDLC			IFF_WAN_HDLC
#define IFF_XMIT_DST_RELEASE		IFF_XMIT_DST_RELEASE
#define IFF_DONT_BRIDGE			IFF_DONT_BRIDGE
#define IFF_DISABLE_NETPOLL		IFF_DISABLE_NETPOLL
#define IFF_MACVLAN_PORT		IFF_MACVLAN_PORT
#define IFF_BRIDGE_PORT			IFF_BRIDGE_PORT
#define IFF_OVS_DATAPATH		IFF_OVS_DATAPATH
#define IFF_TX_SKB_SHARING		IFF_TX_SKB_SHARING
#define IFF_UNICAST_FLT			IFF_UNICAST_FLT
#define IFF_TEAM_PORT			IFF_TEAM_PORT
#define IFF_SUPP_NOFCS			IFF_SUPP_NOFCS
#define IFF_LIVE_ADDR_CHANGE		IFF_LIVE_ADDR_CHANGE
#define IFF_MACVLAN			IFF_MACVLAN
1567
#define IFF_XMIT_DST_RELEASE_PERM	IFF_XMIT_DST_RELEASE_PERM
1568
#define IFF_L3MDEV_MASTER		IFF_L3MDEV_MASTER
1569
#define IFF_NO_QUEUE			IFF_NO_QUEUE
1570
#define IFF_OPENVSWITCH			IFF_OPENVSWITCH
1571
#define IFF_L3MDEV_SLAVE		IFF_L3MDEV_SLAVE
J
Jiri Pirko 已提交
1572
#define IFF_TEAM			IFF_TEAM
1573
#define IFF_RXFH_CONFIGURED		IFF_RXFH_CONFIGURED
1574
#define IFF_MACSEC			IFF_MACSEC
P
Paolo Abeni 已提交
1575
#define IFF_NO_RX_HANDLER		IFF_NO_RX_HANDLER
1576 1577
#define IFF_FAILOVER			IFF_FAILOVER
#define IFF_FAILOVER_SLAVE		IFF_FAILOVER_SLAVE
1578
#define IFF_L3MDEV_RX_HANDLER		IFF_L3MDEV_RX_HANDLER
1579
#define IFF_LIVE_RENAME_OK		IFF_LIVE_RENAME_OK
1580

1581 1582
/**
 *	struct net_device - The DEVICE structure.
1583 1584 1585 1586
 *
 *	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.
1587 1588 1589
 *
 *	@name:	This is the first field of the "visible" part of this structure
 *		(i.e. as seen by users in the "Space.c" file).  It is the name
1590
 *		of the interface.
1591
 *
1592
 *	@name_node:	Name hashlist node
1593 1594 1595 1596 1597 1598 1599 1600
 *	@ifalias:	SNMP alias
 *	@mem_end:	Shared memory end
 *	@mem_start:	Shared memory start
 *	@base_addr:	Device I/O address
 *	@irq:		Device IRQ number
 *
 *	@state:		Generic network queuing layer state, see netdev_state_t
 *	@dev_list:	The global list of network devices
B
Bjorn Helgaas 已提交
1601 1602 1603 1604
 *	@napi_list:	List entry used for polling NAPI devices
 *	@unreg_list:	List entry  when we are unregistering the
 *			device; see the function unregister_netdev
 *	@close_list:	List entry used when we are closing the device
1605 1606
 *	@ptype_all:     Device-specific packet handlers for all protocols
 *	@ptype_specific: Device-specific, protocol-specific packet handlers
1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620
 *
 *	@adj_list:	Directly linked devices, like slaves for bonding
 *	@features:	Currently active device features
 *	@hw_features:	User-changeable features
 *
 *	@wanted_features:	User-requested features
 *	@vlan_features:		Mask of features inheritable by VLAN devices
 *
 *	@hw_enc_features:	Mask of features inherited by encapsulating devices
 *				This field indicates what encapsulation
 *				offloads the hardware is capable of doing,
 *				and drivers will need to set them appropriately.
 *
 *	@mpls_features:	Mask of features inheritable by MPLS
1621
 *	@gso_partial_features: value(s) from NETIF_F_GSO\*
1622 1623
 *
 *	@ifindex:	interface index
B
Bjorn Helgaas 已提交
1624
 *	@group:		The group the device belongs to
1625 1626 1627 1628 1629 1630 1631 1632
 *
 *	@stats:		Statistics struct, which was left as a legacy, use
 *			rtnl_link_stats64 instead
 *
 *	@rx_dropped:	Dropped packets by core network,
 *			do not use this in drivers
 *	@tx_dropped:	Dropped packets by core network,
 *			do not use this in drivers
1633 1634
 *	@rx_nohandler:	nohandler dropped packets by core network on
 *			inactive devices, do not use this in drivers
1635 1636
 *	@carrier_up_count:	Number of times the carrier has been up
 *	@carrier_down_count:	Number of times the carrier has been down
1637 1638 1639 1640 1641 1642 1643 1644 1645
 *
 *	@wireless_handlers:	List of functions to handle Wireless Extensions,
 *				instead of ioctl,
 *				see <net/iw_handler.h> for details.
 *	@wireless_data:	Instance data managed by the core of wireless extensions
 *
 *	@netdev_ops:	Includes several pointers to callbacks,
 *			if one wants to override the ndo_*() functions
 *	@ethtool_ops:	Management operations
1646
 *	@l3mdev_ops:	Layer 3 master device operations
1647 1648
 *	@ndisc_ops:	Includes callbacks for different IPv6 neighbour
 *			discovery handling. Necessary for e.g. 6LoWPAN.
1649 1650
 *	@xfrmdev_ops:	Transformation offload operations
 *	@tlsdev_ops:	Transport Layer Security offload operations
E
Eric W. Biederman 已提交
1651
 *	@header_ops:	Includes callbacks for creating,parsing,caching,etc
1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663
 *			of Layer 2 headers.
 *
 *	@flags:		Interface flags (a la BSD)
 *	@priv_flags:	Like 'flags' but invisible to userspace,
 *			see if.h for the definitions
 *	@gflags:	Global flags ( kept as legacy )
 *	@padded:	How much padding added by alloc_netdev()
 *	@operstate:	RFC2863 operstate
 *	@link_mode:	Mapping policy to operstate
 *	@if_port:	Selectable AUI, TP, ...
 *	@dma:		DMA channel
 *	@mtu:		Interface MTU value
1664 1665
 *	@min_mtu:	Interface Minimum MTU value
 *	@max_mtu:	Interface Maximum MTU value
1666
 *	@type:		Interface hardware type
1667
 *	@hard_header_len: Maximum hardware header length.
1668
 *	@min_header_len:  Minimum hardware header length
1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680
 *
 *	@needed_headroom: Extra headroom the hardware may need, but not in all
 *			  cases can this be guaranteed
 *	@needed_tailroom: Extra tailroom the hardware may need, but not in all
 *			  cases can this be guaranteed. Some cases also use
 *			  LL_MAX_HEADER instead to allocate the skb
 *
 *	interface address info:
 *
 * 	@perm_addr:		Permanent hw address
 * 	@addr_assign_type:	Hw address assignment type
 * 	@addr_len:		Hardware address length
T
Taehee Yoo 已提交
1681 1682
 *	@upper_level:		Maximum depth level of upper devices.
 *	@lower_level:		Maximum depth level of lower devices.
1683
 *	@neigh_priv_len:	Used in neigh_alloc()
1684 1685 1686 1687 1688
 * 	@dev_id:		Used to differentiate devices that share
 * 				the same link layer address
 * 	@dev_port:		Used to differentiate devices that share
 * 				the same function
 *	@addr_list_lock:	XXX: need comments on this one
1689
 *	@name_assign_type:	network interface name assignment type
B
Bjorn Helgaas 已提交
1690
 *	@uc_promisc:		Counter that indicates promiscuous mode
1691 1692 1693
 *				has been enabled due to the need to listen to
 *				additional unicast addresses in a device that
 *				does not implement ndo_set_rx_mode()
1694 1695 1696 1697
 *	@uc:			unicast mac addresses
 *	@mc:			multicast mac addresses
 *	@dev_addrs:		list of device hw addresses
 *	@queues_kset:		Group of all Kobjects in the Tx and RX queues
B
Bjorn Helgaas 已提交
1698 1699 1700
 *	@promiscuity:		Number of times the NIC is told to work in
 *				promiscuous mode; if it becomes 0 the NIC will
 *				exit promiscuous mode
1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711
 *	@allmulti:		Counter, enables or disables allmulticast mode
 *
 *	@vlan_info:	VLAN info
 *	@dsa_ptr:	dsa specific data
 *	@tipc_ptr:	TIPC specific data
 *	@atalk_ptr:	AppleTalk link
 *	@ip_ptr:	IPv4 specific data
 *	@dn_ptr:	DECnet specific data
 *	@ip6_ptr:	IPv6 specific data
 *	@ax25_ptr:	AX.25 specific data
 *	@ieee80211_ptr:	IEEE 802.11 specific data, assign before registering
1712 1713 1714
 *	@ieee802154_ptr: IEEE 802.15.4 low-rate Wireless Personal Area Network
 *			 device struct
 *	@mpls_ptr:	mpls_dev struct pointer
1715 1716 1717 1718 1719 1720 1721 1722
 *
 *	@dev_addr:	Hw address (before bcast,
 *			because most packets are unicast)
 *
 *	@_rx:			Array of RX queues
 *	@num_rx_queues:		Number of RX queues
 *				allocated at register_netdev() time
 *	@real_num_rx_queues: 	Number of RX queues currently active in device
1723 1724
 *	@xdp_prog:		XDP sockets filter program pointer
 *	@gro_flush_timeout:	timeout for GRO layer in NAPI
1725 1726 1727
 *
 *	@rx_handler:		handler for received packets
 *	@rx_handler_data: 	XXX: need comments on this one
1728 1729
 *	@miniq_ingress:		ingress/clsact qdisc specific data for
 *				ingress processing
1730
 *	@ingress_queue:		XXX: need comments on this one
L
Lukas Wunner 已提交
1731
 *	@nf_hooks_ingress:	netfilter hooks executed for ingress packets
1732 1733
 *	@broadcast:		hw bcast address
 *
1734 1735 1736 1737 1738 1739
 *	@rx_cpu_rmap:	CPU reverse-mapping for RX completion interrupts,
 *			indexed by RX queue number. Assigned by driver.
 *			This must only be set if the ndo_rx_flow_steer
 *			operation is defined
 *	@index_hlist:		Device index hash chain
 *
1740 1741 1742 1743 1744 1745
 *	@_tx:			Array of TX queues
 *	@num_tx_queues:		Number of TX queues allocated at alloc_netdev_mq() time
 *	@real_num_tx_queues: 	Number of TX queues currently active in device
 *	@qdisc:			Root qdisc from userspace point of view
 *	@tx_queue_len:		Max frames per queue allowed
 *	@tx_global_lock: 	XXX: need comments on this one
1746 1747 1748
 *	@xdp_bulkq:		XDP device bulk queue
 *	@xps_cpus_map:		all CPUs map for XPS device
 *	@xps_rxqs_map:		all RXQs map for XPS device
1749 1750
 *
 *	@xps_maps:	XXX: need comments on this one
1751 1752
 *	@miniq_egress:		clsact qdisc specific data for
 *				egress processing
1753
 *	@qdisc_hash:		qdisc hash table
1754
 *	@watchdog_timeo:	Represents the timeout that is used by
B
Bjorn Helgaas 已提交
1755
 *				the watchdog (see dev_watchdog())
1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766
 *	@watchdog_timer:	List of timers
 *
 *	@pcpu_refcnt:		Number of references to this device
 *	@todo_list:		Delayed register/unregister
 *	@link_watch_list:	XXX: need comments on this one
 *
 *	@reg_state:		Register/unregister state machine
 *	@dismantle:		Device is going to be freed
 *	@rtnl_link_state:	This enum represents the phases of creating
 *				a new link
 *
1767 1768
 *	@needs_free_netdev:	Should unregister perform free_netdev?
 *	@priv_destructor:	Called from unregister
1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794
 *	@npinfo:		XXX: need comments on this one
 * 	@nd_net:		Network namespace this network device is inside
 *
 * 	@ml_priv:	Mid-layer private
 * 	@lstats:	Loopback statistics
 * 	@tstats:	Tunnel statistics
 * 	@dstats:	Dummy statistics
 * 	@vstats:	Virtual ethernet statistics
 *
 *	@garp_port:	GARP
 *	@mrp_port:	MRP
 *
 *	@dev:		Class/net/name entry
 *	@sysfs_groups:	Space for optional device, statistics and wireless
 *			sysfs groups
 *
 *	@sysfs_rx_queue_group:	Space for optional per-rx queue attributes
 *	@rtnl_link_ops:	Rtnl_link_ops
 *
 *	@gso_max_size:	Maximum size of generic segmentation offload
 *	@gso_max_segs:	Maximum number of segments that can be passed to the
 *			NIC for GSO
 *
 *	@dcbnl_ops:	Data Center Bridging netlink ops
 *	@num_tc:	Number of traffic classes in the net device
 *	@tc_to_txq:	XXX: need comments on this one
R
Randy Dunlap 已提交
1795
 *	@prio_tc_map:	XXX: need comments on this one
1796 1797 1798 1799 1800 1801
 *
 *	@fcoe_ddp_xid:	Max exchange id for FCoE LRO by ddp
 *
 *	@priomap:	XXX: need comments on this one
 *	@phydev:	Physical device may attach itself
 *			for hardware timestamping
1802
 *	@sfp_bus:	attached &struct sfp_bus structure.
T
Taehee Yoo 已提交
1803
 *	@qdisc_tx_busylock_key: lockdep class annotating Qdisc->busylock
1804
 *				spinlock
T
Taehee Yoo 已提交
1805 1806 1807 1808 1809
 *	@qdisc_running_key:	lockdep class annotating Qdisc->running seqcount
 *	@qdisc_xmit_lock_key:	lockdep class annotating
 *				netdev_queue->_xmit_lock spinlock
 *	@addr_list_lock_key:	lockdep class annotating
 *				net_device->addr_list_lock spinlock
1810
 *
1811 1812 1813 1814
 *	@proto_down:	protocol port state information can be sent to the
 *			switch driver and used to set the phys state of the
 *			switch port.
 *
1815 1816
 *	@wol_enabled:	Wake-on-LAN is enabled
 *
1817 1818 1819 1820
 *	@net_notifier_list:	List of per-net netdev notifier block
 *				that follow this device when it is moved
 *				to another network namespace.
 *
L
Linus Torvalds 已提交
1821 1822 1823 1824
 *	FIXME: cleanup struct net_device such that network protocol info
 *	moves out.
 */

E
Eric Dumazet 已提交
1825
struct net_device {
L
Linus Torvalds 已提交
1826
	char			name[IFNAMSIZ];
1827
	struct netdev_name_node	*name_node;
1828
	struct dev_ifalias	__rcu *ifalias;
L
Linus Torvalds 已提交
1829 1830 1831 1832
	/*
	 *	I/O specific fields
	 *	FIXME: Merge these and struct ifmap into one
	 */
1833 1834 1835 1836
	unsigned long		mem_end;
	unsigned long		mem_start;
	unsigned long		base_addr;
	int			irq;
L
Linus Torvalds 已提交
1837 1838

	/*
1839 1840
	 *	Some hardware also needs these fields (state,dev_list,
	 *	napi_list,unreg_list,close_list) but they are not
L
Linus Torvalds 已提交
1841 1842 1843 1844 1845
	 *	part of the usual set specified in Space.c.
	 */

	unsigned long		state;

1846
	struct list_head	dev_list;
1847
	struct list_head	napi_list;
1848
	struct list_head	unreg_list;
1849
	struct list_head	close_list;
1850 1851
	struct list_head	ptype_all;
	struct list_head	ptype_specific;
1852 1853 1854 1855 1856 1857

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

1858 1859 1860 1861
	netdev_features_t	features;
	netdev_features_t	hw_features;
	netdev_features_t	wanted_features;
	netdev_features_t	vlan_features;
1862
	netdev_features_t	hw_enc_features;
S
Simon Horman 已提交
1863
	netdev_features_t	mpls_features;
1864
	netdev_features_t	gso_partial_features;
1865

L
Linus Torvalds 已提交
1866
	int			ifindex;
1867
	int			group;
L
Linus Torvalds 已提交
1868

R
Rusty Russell 已提交
1869
	struct net_device_stats	stats;
1870 1871 1872

	atomic_long_t		rx_dropped;
	atomic_long_t		tx_dropped;
1873
	atomic_long_t		rx_nohandler;
L
Linus Torvalds 已提交
1874

1875 1876 1877 1878
	/* Stats to monitor link on/off, flapping */
	atomic_t		carrier_up_count;
	atomic_t		carrier_down_count;

1879
#ifdef CONFIG_WIRELESS_EXT
B
Bjorn Helgaas 已提交
1880 1881
	const struct iw_handler_def *wireless_handlers;
	struct iw_public_data	*wireless_data;
1882
#endif
1883
	const struct net_device_ops *netdev_ops;
1884
	const struct ethtool_ops *ethtool_ops;
1885 1886 1887
#ifdef CONFIG_NET_L3_MASTER_DEV
	const struct l3mdev_ops	*l3mdev_ops;
#endif
1888 1889 1890
#if IS_ENABLED(CONFIG_IPV6)
	const struct ndisc_ops *ndisc_ops;
#endif
L
Linus Torvalds 已提交
1891

1892
#ifdef CONFIG_XFRM_OFFLOAD
1893 1894 1895
	const struct xfrmdev_ops *xfrmdev_ops;
#endif

I
Ilya Lesokhin 已提交
1896 1897 1898 1899
#if IS_ENABLED(CONFIG_TLS_DEVICE)
	const struct tlsdev_ops *tlsdev_ops;
#endif

1900 1901
	const struct header_ops *header_ops;

1902 1903 1904
	unsigned int		flags;
	unsigned int		priv_flags;

L
Linus Torvalds 已提交
1905
	unsigned short		gflags;
1906
	unsigned short		padded;
L
Linus Torvalds 已提交
1907

1908 1909
	unsigned char		operstate;
	unsigned char		link_mode;
S
Stefan Rompf 已提交
1910

1911 1912
	unsigned char		if_port;
	unsigned char		dma;
1913

1914 1915 1916 1917 1918
	/* Note : dev->mtu is often read without holding a lock.
	 * Writers usually hold RTNL.
	 * It is recommended to use READ_ONCE() to annotate the reads,
	 * and to use WRITE_ONCE() to annotate the writes.
	 */
1919
	unsigned int		mtu;
1920 1921
	unsigned int		min_mtu;
	unsigned int		max_mtu;
1922 1923
	unsigned short		type;
	unsigned short		hard_header_len;
1924
	unsigned char		min_header_len;
L
Linus Torvalds 已提交
1925

1926 1927 1928
	unsigned short		needed_headroom;
	unsigned short		needed_tailroom;

L
Linus Torvalds 已提交
1929
	/* Interface address info. */
1930 1931 1932
	unsigned char		perm_addr[MAX_ADDR_LEN];
	unsigned char		addr_assign_type;
	unsigned char		addr_len;
T
Taehee Yoo 已提交
1933 1934
	unsigned char		upper_level;
	unsigned char		lower_level;
1935
	unsigned short		neigh_priv_len;
1936 1937
	unsigned short          dev_id;
	unsigned short          dev_port;
J
Jiri Pirko 已提交
1938
	spinlock_t		addr_list_lock;
1939 1940
	unsigned char		name_assign_type;
	bool			uc_promisc;
1941 1942 1943 1944
	struct netdev_hw_addr_list	uc;
	struct netdev_hw_addr_list	mc;
	struct netdev_hw_addr_list	dev_addrs;

1945 1946 1947
#ifdef CONFIG_SYSFS
	struct kset		*queues_kset;
#endif
1948 1949
	unsigned int		promiscuity;
	unsigned int		allmulti;
L
Linus Torvalds 已提交
1950 1951


B
Bjorn Helgaas 已提交
1952
	/* Protocol-specific pointers */
1953

1954
#if IS_ENABLED(CONFIG_VLAN_8021Q)
1955
	struct vlan_info __rcu	*vlan_info;
1956
#endif
1957
#if IS_ENABLED(CONFIG_NET_DSA)
1958
	struct dsa_port		*dsa_ptr;
1959 1960
#endif
#if IS_ENABLED(CONFIG_TIPC)
1961
	struct tipc_bearer __rcu *tipc_ptr;
1962
#endif
1963
#if IS_ENABLED(CONFIG_IRDA) || IS_ENABLED(CONFIG_ATALK)
1964
	void 			*atalk_ptr;
1965
#endif
1966
	struct in_device __rcu	*ip_ptr;
1967
#if IS_ENABLED(CONFIG_DECNET)
1968
	struct dn_dev __rcu     *dn_ptr;
1969
#endif
1970
	struct inet6_dev __rcu	*ip6_ptr;
1971
#if IS_ENABLED(CONFIG_AX25)
1972
	void			*ax25_ptr;
1973
#endif
1974
	struct wireless_dev	*ieee80211_ptr;
1975
	struct wpan_dev		*ieee802154_ptr;
R
Robert Shearman 已提交
1976 1977 1978
#if IS_ENABLED(CONFIG_MPLS_ROUTING)
	struct mpls_dev __rcu	*mpls_ptr;
#endif
L
Linus Torvalds 已提交
1979

1980
/*
E
Eric Dumazet 已提交
1981
 * Cache lines mostly used on receive path (including eth_type_trans())
1982 1983
 */
	/* Interface address info used in eth_type_trans() */
1984
	unsigned char		*dev_addr;
1985

T
Tom Herbert 已提交
1986 1987
	struct netdev_rx_queue	*_rx;
	unsigned int		num_rx_queues;
1988
	unsigned int		real_num_rx_queues;
T
Tom Herbert 已提交
1989

1990
	struct bpf_prog __rcu	*xdp_prog;
1991
	unsigned long		gro_flush_timeout;
1992 1993
	rx_handler_func_t __rcu	*rx_handler;
	void __rcu		*rx_handler_data;
1994

1995
#ifdef CONFIG_NET_CLS_ACT
1996
	struct mini_Qdisc __rcu	*miniq_ingress;
1997
#endif
1998
	struct netdev_queue __rcu *ingress_queue;
1999
#ifdef CONFIG_NETFILTER_INGRESS
2000
	struct nf_hook_entries __rcu *nf_hooks_ingress;
2001
#endif
2002

2003
	unsigned char		broadcast[MAX_ADDR_LEN];
2004 2005 2006 2007
#ifdef CONFIG_RFS_ACCEL
	struct cpu_rmap		*rx_cpu_rmap;
#endif
	struct hlist_node	index_hlist;
E
Eric Dumazet 已提交
2008 2009 2010 2011

/*
 * Cache lines mostly used on transmit path
 */
2012 2013
	struct netdev_queue	*_tx ____cacheline_aligned_in_smp;
	unsigned int		num_tx_queues;
2014
	unsigned int		real_num_tx_queues;
2015
	struct Qdisc		*qdisc;
2016
	unsigned int		tx_queue_len;
2017
	spinlock_t		tx_global_lock;
2018 2019

	struct xdp_dev_bulk_queue __percpu *xdp_bulkq;
E
Eric Dumazet 已提交
2020

T
Tom Herbert 已提交
2021
#ifdef CONFIG_XPS
2022 2023
	struct xps_dev_maps __rcu *xps_cpus_map;
	struct xps_dev_maps __rcu *xps_rxqs_map;
T
Tom Herbert 已提交
2024
#endif
D
Daniel Borkmann 已提交
2025
#ifdef CONFIG_NET_CLS_ACT
2026
	struct mini_Qdisc __rcu	*miniq_egress;
D
Daniel Borkmann 已提交
2027
#endif
2028

2029 2030 2031
#ifdef CONFIG_NET_SCHED
	DECLARE_HASHTABLE	(qdisc_hash, 4);
#endif
2032 2033
	/* These may be needed for future network-power-down code. */
	struct timer_list	watchdog_timer;
2034
	int			watchdog_timeo;
2035

L
Linus Torvalds 已提交
2036
	struct list_head	todo_list;
2037
	int __percpu		*pcpu_refcnt;
L
Linus Torvalds 已提交
2038

2039
	struct list_head	link_watch_list;
2040

L
Linus Torvalds 已提交
2041
	enum { NETREG_UNINITIALIZED=0,
2042
	       NETREG_REGISTERED,	/* completed register_netdevice */
L
Linus Torvalds 已提交
2043 2044 2045
	       NETREG_UNREGISTERING,	/* called unregister_netdevice */
	       NETREG_UNREGISTERED,	/* completed unregister todo */
	       NETREG_RELEASED,		/* called free_netdev */
2046
	       NETREG_DUMMY,		/* dummy device for NAPI poll */
2047 2048
	} reg_state:8;

2049
	bool dismantle;
2050 2051 2052 2053 2054

	enum {
		RTNL_LINK_INITIALIZED,
		RTNL_LINK_INITIALIZING,
	} rtnl_link_state:16;
L
Linus Torvalds 已提交
2055

2056 2057
	bool needs_free_netdev;
	void (*priv_destructor)(struct net_device *dev);
L
Linus Torvalds 已提交
2058 2059

#ifdef CONFIG_NETPOLL
2060
	struct netpoll_info __rcu	*npinfo;
L
Linus Torvalds 已提交
2061
#endif
2062

2063
	possible_net_t			nd_net;
2064

D
David S. Miller 已提交
2065
	/* mid-layer private */
E
Eric Dumazet 已提交
2066
	union {
2067 2068
		void					*ml_priv;
		struct pcpu_lstats __percpu		*lstats;
2069
		struct pcpu_sw_netstats __percpu	*tstats;
2070
		struct pcpu_dstats __percpu		*dstats;
E
Eric Dumazet 已提交
2071
	};
2072

2073
#if IS_ENABLED(CONFIG_GARP)
E
Eric Dumazet 已提交
2074
	struct garp_port __rcu	*garp_port;
2075 2076
#endif
#if IS_ENABLED(CONFIG_MRP)
2077
	struct mrp_port __rcu	*mrp_port;
2078
#endif
L
Linus Torvalds 已提交
2079

B
Bjorn Helgaas 已提交
2080
	struct device		dev;
2081
	const struct attribute_group *sysfs_groups[4];
2082
	const struct attribute_group *sysfs_rx_queue_group;
P
Patrick McHardy 已提交
2083 2084

	const struct rtnl_link_ops *rtnl_link_ops;
2085

2086 2087 2088
	/* for setting kernel sock attribute on TCP connection setup */
#define GSO_MAX_SIZE		65536
	unsigned int		gso_max_size;
2089 2090
#define GSO_MAX_SEGS		65535
	u16			gso_max_segs;
2091

J
Jeff Kirsher 已提交
2092
#ifdef CONFIG_DCB
2093
	const struct dcbnl_rtnl_ops *dcbnl_ops;
2094
#endif
2095
	s16			num_tc;
B
Bjorn Helgaas 已提交
2096 2097
	struct netdev_tc_txq	tc_to_txq[TC_MAX_QUEUE];
	u8			prio_tc_map[TC_BITMASK + 1];
2098

2099
#if IS_ENABLED(CONFIG_FCOE)
2100
	unsigned int		fcoe_ddp_xid;
2101
#endif
2102
#if IS_ENABLED(CONFIG_CGROUP_NET_PRIO)
2103
	struct netprio_map __rcu *priomap;
2104
#endif
B
Bjorn Helgaas 已提交
2105
	struct phy_device	*phydev;
2106
	struct sfp_bus		*sfp_bus;
T
Taehee Yoo 已提交
2107 2108 2109 2110
	struct lock_class_key	qdisc_tx_busylock_key;
	struct lock_class_key	qdisc_running_key;
	struct lock_class_key	qdisc_xmit_lock_key;
	struct lock_class_key	addr_list_lock_key;
B
Bjorn Helgaas 已提交
2111
	bool			proto_down;
2112
	unsigned		wol_enabled:1;
2113 2114

	struct list_head	net_notifier_list;
L
Linus Torvalds 已提交
2115
};
2116
#define to_net_dev(d) container_of(d, struct net_device, dev)
L
Linus Torvalds 已提交
2117

D
David S. Miller 已提交
2118 2119 2120 2121 2122 2123 2124
static inline bool netif_elide_gro(const struct net_device *dev)
{
	if (!(dev->features & NETIF_F_GRO) || dev->xdp_prog)
		return true;
	return false;
}

L
Linus Torvalds 已提交
2125 2126
#define	NETDEV_ALIGN		32

2127 2128 2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142
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;
}

2143
int netdev_txq_to_tc(struct net_device *dev, unsigned int txq);
2144 2145 2146
void netdev_reset_tc(struct net_device *dev);
int netdev_set_tc_queue(struct net_device *dev, u8 tc, u16 count, u16 offset);
int netdev_set_num_tc(struct net_device *dev, u8 num_tc);
2147 2148 2149 2150 2151 2152 2153

static inline
int netdev_get_num_tc(struct net_device *dev)
{
	return dev->num_tc;
}

2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164
void netdev_unbind_sb_channel(struct net_device *dev,
			      struct net_device *sb_dev);
int netdev_bind_sb_channel_queue(struct net_device *dev,
				 struct net_device *sb_dev,
				 u8 tc, u16 count, u16 offset);
int netdev_set_sb_channel(struct net_device *dev, u16 channel);
static inline int netdev_get_sb_channel(struct net_device *dev)
{
	return max_t(int, -dev->num_tc, 0);
}

2165 2166 2167 2168 2169 2170 2171
static inline
struct netdev_queue *netdev_get_tx_queue(const struct net_device *dev,
					 unsigned int index)
{
	return &dev->_tx[index];
}

2172 2173 2174 2175 2176 2177
static inline struct netdev_queue *skb_get_tx_queue(const struct net_device *dev,
						    const struct sk_buff *skb)
{
	return netdev_get_tx_queue(dev, skb_get_queue_mapping(skb));
}

2178 2179 2180 2181 2182 2183 2184 2185 2186 2187 2188 2189
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);
}

2190 2191
u16 netdev_pick_tx(struct net_device *dev, struct sk_buff *skb,
		     struct net_device *sb_dev);
2192 2193 2194
struct netdev_queue *netdev_core_pick_tx(struct net_device *dev,
					 struct sk_buff *skb,
					 struct net_device *sb_dev);
2195

2196 2197 2198 2199 2200 2201 2202 2203 2204 2205 2206 2207 2208 2209 2210 2211 2212 2213 2214 2215
/* returns the headroom that the master device needs to take in account
 * when forwarding to this dev
 */
static inline unsigned netdev_get_fwd_headroom(struct net_device *dev)
{
	return dev->priv_flags & IFF_PHONY_HEADROOM ? 0 : dev->needed_headroom;
}

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

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

2216 2217 2218 2219 2220 2221
/*
 * Net namespace inlines
 */
static inline
struct net *dev_net(const struct net_device *dev)
{
E
Eric Dumazet 已提交
2222
	return read_pnet(&dev->nd_net);
2223 2224 2225
}

static inline
2226
void dev_net_set(struct net_device *dev, struct net *net)
2227
{
2228
	write_pnet(&dev->nd_net, net);
2229 2230
}

2231 2232 2233 2234 2235 2236
/**
 *	netdev_priv - access network device private data
 *	@dev: network device
 *
 * Get network device private data
 */
2237
static inline void *netdev_priv(const struct net_device *dev)
L
Linus Torvalds 已提交
2238
{
2239
	return (char *)dev + ALIGN(sizeof(struct net_device), NETDEV_ALIGN);
L
Linus Torvalds 已提交
2240 2241 2242 2243 2244
}

/* Set the sysfs physical device reference for the network logical device
 * if set prior to registration will cause a symlink during initialization.
 */
2245
#define SET_NETDEV_DEV(net, pdev)	((net)->dev.parent = (pdev))
L
Linus Torvalds 已提交
2246

2247
/* Set the sysfs device type for the network logical device to allow
2248
 * fine-grained identification of different network device types. For
B
Bjorn Helgaas 已提交
2249
 * example Ethernet, Wireless LAN, Bluetooth, WiMAX etc.
2250 2251 2252
 */
#define SET_NETDEV_DEVTYPE(net, devtype)	((net)->dev.type = (devtype))

E
Eric Dumazet 已提交
2253 2254 2255 2256 2257
/* Default NAPI poll() weight
 * Device drivers are strongly advised to not use bigger value
 */
#define NAPI_POLL_WEIGHT 64

2258
/**
B
Bjorn Helgaas 已提交
2259
 *	netif_napi_add - initialize a NAPI context
2260
 *	@dev:  network device
B
Bjorn Helgaas 已提交
2261
 *	@napi: NAPI context
2262 2263 2264
 *	@poll: polling function
 *	@weight: default weight
 *
B
Bjorn Helgaas 已提交
2265 2266
 * netif_napi_add() must be used to initialize a NAPI context prior to calling
 * *any* of the other NAPI-related functions.
2267
 */
2268 2269
void netif_napi_add(struct net_device *dev, struct napi_struct *napi,
		    int (*poll)(struct napi_struct *, int), int weight);
2270

E
Eric Dumazet 已提交
2271
/**
B
Bjorn Helgaas 已提交
2272
 *	netif_tx_napi_add - initialize a NAPI context
E
Eric Dumazet 已提交
2273
 *	@dev:  network device
B
Bjorn Helgaas 已提交
2274
 *	@napi: NAPI context
E
Eric Dumazet 已提交
2275 2276 2277 2278 2279 2280 2281 2282 2283 2284 2285 2286 2287 2288 2289 2290
 *	@poll: polling function
 *	@weight: default weight
 *
 * This variant of netif_napi_add() should be used from drivers using NAPI
 * to exclusively poll a TX queue.
 * This will avoid we add it into napi_hash[], thus polluting this hash table.
 */
static inline void netif_tx_napi_add(struct net_device *dev,
				     struct napi_struct *napi,
				     int (*poll)(struct napi_struct *, int),
				     int weight)
{
	set_bit(NAPI_STATE_NO_BUSY_POLL, &napi->state);
	netif_napi_add(dev, napi, poll, weight);
}

2291
/**
B
Bjorn Helgaas 已提交
2292 2293
 *  netif_napi_del - remove a NAPI context
 *  @napi: NAPI context
2294
 *
B
Bjorn Helgaas 已提交
2295
 *  netif_napi_del() removes a NAPI context from the network device NAPI list
2296
 */
2297 2298 2299
void netif_napi_del(struct napi_struct *napi);

struct napi_gro_cb {
2300
	/* Virtual address of skb_shinfo(skb)->frags[0].page + offset. */
B
Bjorn Helgaas 已提交
2301
	void	*frag0;
2302

2303 2304 2305
	/* Length of frag0. */
	unsigned int frag0_len;

2306
	/* This indicates where we are processing relative to skb->data. */
B
Bjorn Helgaas 已提交
2307
	int	data_offset;
2308

2309
	/* This is non-zero if the packet cannot be merged with the new skb. */
2310 2311 2312 2313
	u16	flush;

	/* Save the IP ID here and check when we get to the transport layer */
	u16	flush_id;
2314 2315

	/* Number of segments aggregated. */
2316 2317
	u16	count;

2318 2319 2320
	/* Start offset for remote checksum offload */
	u16	gro_remcsum_start;

2321 2322
	/* jiffies when first packet was created/queued */
	unsigned long age;
2323

T
Tom Herbert 已提交
2324
	/* Used in ipv6_gro_receive() and foo-over-udp */
2325 2326
	u16	proto;

2327 2328 2329
	/* This is non-zero if the packet may be of the same flow. */
	u8	same_flow:1;

2330 2331
	/* Used in tunnel GRO receive */
	u8	encap_mark:1;
T
Tom Herbert 已提交
2332 2333 2334 2335

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

2336 2337
	/* Number of checksums via CHECKSUM_UNNECESSARY */
	u8	csum_cnt:3;
2338

2339 2340 2341 2342 2343
	/* Free the skb? */
	u8	free:2;
#define NAPI_GRO_FREE		  1
#define NAPI_GRO_FREE_STOLEN_HEAD 2

2344 2345 2346
	/* Used in foo-over-udp, set in udp[46]_gro_receive */
	u8	is_ipv6:1;

2347 2348 2349
	/* Used in GRE, set in fou/gue_gro_receive */
	u8	is_fou:1;

2350 2351 2352
	/* Used to determine if flush_id can be ignored */
	u8	is_atomic:1;

S
Sabrina Dubroca 已提交
2353 2354 2355
	/* Number of gro_receive callbacks this packet already went through */
	u8 recursion_counter:4;

2356 2357
	/* GRO is done by frag_list pointer chaining. */
	u8	is_flist:1;
2358

2359 2360 2361
	/* used to support CHECKSUM_COMPLETE for tunneling protocols */
	__wsum	csum;

2362 2363
	/* used in skb_gro_receive() slow path */
	struct sk_buff *last;
2364 2365 2366
};

#define NAPI_GRO_CB(skb) ((struct napi_gro_cb *)(skb)->cb)
2367

S
Sabrina Dubroca 已提交
2368 2369 2370 2371 2372 2373
#define GRO_RECURSION_LIMIT 15
static inline int gro_recursion_inc_test(struct sk_buff *skb)
{
	return ++NAPI_GRO_CB(skb)->recursion_counter == GRO_RECURSION_LIMIT;
}

2374 2375 2376 2377
typedef struct sk_buff *(*gro_receive_t)(struct list_head *, struct sk_buff *);
static inline struct sk_buff *call_gro_receive(gro_receive_t cb,
					       struct list_head *head,
					       struct sk_buff *skb)
S
Sabrina Dubroca 已提交
2378 2379 2380 2381 2382 2383 2384 2385 2386
{
	if (unlikely(gro_recursion_inc_test(skb))) {
		NAPI_GRO_CB(skb)->flush |= 1;
		return NULL;
	}

	return cb(head, skb);
}

2387 2388 2389 2390 2391 2392
typedef struct sk_buff *(*gro_receive_sk_t)(struct sock *, struct list_head *,
					    struct sk_buff *);
static inline struct sk_buff *call_gro_receive_sk(gro_receive_sk_t cb,
						  struct sock *sk,
						  struct list_head *head,
						  struct sk_buff *skb)
S
Sabrina Dubroca 已提交
2393 2394 2395 2396 2397 2398 2399 2400 2401
{
	if (unlikely(gro_recursion_inc_test(skb))) {
		NAPI_GRO_CB(skb)->flush |= 1;
		return NULL;
	}

	return cb(sk, head, skb);
}

L
Linus Torvalds 已提交
2402
struct packet_type {
D
David S. Miller 已提交
2403
	__be16			type;	/* This is really htons(ether_type). */
2404
	bool			ignore_outgoing;
D
David S. Miller 已提交
2405 2406 2407 2408 2409
	struct net_device	*dev;	/* NULL is wildcarded here	     */
	int			(*func) (struct sk_buff *,
					 struct net_device *,
					 struct packet_type *,
					 struct net_device *);
2410 2411 2412
	void			(*list_func) (struct list_head *,
					      struct packet_type *,
					      struct net_device *);
2413 2414
	bool			(*id_match)(struct packet_type *ptype,
					    struct sock *sk);
L
Linus Torvalds 已提交
2415 2416 2417 2418
	void			*af_packet_priv;
	struct list_head	list;
};

2419
struct offload_callbacks {
2420
	struct sk_buff		*(*gso_segment)(struct sk_buff *skb,
2421
						netdev_features_t features);
2422 2423
	struct sk_buff		*(*gro_receive)(struct list_head *head,
						struct sk_buff *skb);
2424
	int			(*gro_complete)(struct sk_buff *skb, int nhoff);
2425 2426 2427 2428
};

struct packet_offload {
	__be16			 type;	/* This is really htons(ether_type). */
2429
	u16			 priority;
2430 2431
	struct offload_callbacks callbacks;
	struct list_head	 list;
L
Linus Torvalds 已提交
2432 2433
};

B
Bjorn Helgaas 已提交
2434
/* often modified stats are per-CPU, other are shared (netdev->stats) */
2435 2436 2437 2438 2439 2440
struct pcpu_sw_netstats {
	u64     rx_packets;
	u64     rx_bytes;
	u64     tx_packets;
	u64     tx_bytes;
	struct u64_stats_sync   syncp;
2441
} __aligned(4 * sizeof(u64));
2442 2443

struct pcpu_lstats {
2444 2445
	u64_stats_t packets;
	u64_stats_t bytes;
2446
	struct u64_stats_sync syncp;
2447
} __aligned(2 * sizeof(u64));
2448

2449 2450
void dev_lstats_read(struct net_device *dev, u64 *packets, u64 *bytes);

2451 2452 2453 2454 2455
static inline void dev_lstats_add(struct net_device *dev, unsigned int len)
{
	struct pcpu_lstats *lstats = this_cpu_ptr(dev->lstats);

	u64_stats_update_begin(&lstats->syncp);
2456 2457
	u64_stats_add(&lstats->bytes, len);
	u64_stats_inc(&lstats->packets);
2458 2459 2460
	u64_stats_update_end(&lstats->syncp);
}

2461 2462 2463 2464 2465 2466 2467 2468 2469 2470 2471 2472
#define __netdev_alloc_pcpu_stats(type, gfp)				\
({									\
	typeof(type) __percpu *pcpu_stats = alloc_percpu_gfp(type, gfp);\
	if (pcpu_stats)	{						\
		int __cpu;						\
		for_each_possible_cpu(__cpu) {				\
			typeof(type) *stat;				\
			stat = per_cpu_ptr(pcpu_stats, __cpu);		\
			u64_stats_init(&stat->syncp);			\
		}							\
	}								\
	pcpu_stats;							\
2473 2474
})

2475
#define netdev_alloc_pcpu_stats(type)					\
2476
	__netdev_alloc_pcpu_stats(type, GFP_KERNEL)
2477

2478 2479 2480 2481 2482 2483 2484 2485 2486
enum netdev_lag_tx_type {
	NETDEV_LAG_TX_TYPE_UNKNOWN,
	NETDEV_LAG_TX_TYPE_RANDOM,
	NETDEV_LAG_TX_TYPE_BROADCAST,
	NETDEV_LAG_TX_TYPE_ROUNDROBIN,
	NETDEV_LAG_TX_TYPE_ACTIVEBACKUP,
	NETDEV_LAG_TX_TYPE_HASH,
};

2487 2488 2489 2490 2491 2492 2493 2494 2495 2496
enum netdev_lag_hash {
	NETDEV_LAG_HASH_NONE,
	NETDEV_LAG_HASH_L2,
	NETDEV_LAG_HASH_L34,
	NETDEV_LAG_HASH_L23,
	NETDEV_LAG_HASH_E23,
	NETDEV_LAG_HASH_E34,
	NETDEV_LAG_HASH_UNKNOWN,
};

2497 2498
struct netdev_lag_upper_info {
	enum netdev_lag_tx_type tx_type;
2499
	enum netdev_lag_hash hash_type;
2500 2501
};

2502 2503 2504 2505 2506
struct netdev_lag_lower_state_info {
	u8 link_up : 1,
	   tx_enabled : 1;
};

L
Linus Torvalds 已提交
2507 2508
#include <linux/notifier.h>

2509 2510 2511
/* netdevice notifier chain. Please remember to update netdev_cmd_to_name()
 * and the rtnetlink notification exclusion list in rtnetlink_event() when
 * adding new types.
2512
 */
2513 2514 2515 2516
enum netdev_cmd {
	NETDEV_UP	= 1,	/* For now you can't veto a device up/down */
	NETDEV_DOWN,
	NETDEV_REBOOT,		/* Tell a protocol stack a network interface
2517 2518 2519
				   detected a hardware crash and restarted
				   - we can use this eg to kick tcp sessions
				   once done */
2520 2521 2522 2523
	NETDEV_CHANGE,		/* Notify device state change */
	NETDEV_REGISTER,
	NETDEV_UNREGISTER,
	NETDEV_CHANGEMTU,	/* notify after mtu change happened */
P
Petr Machata 已提交
2524 2525
	NETDEV_CHANGEADDR,	/* notify after the address change */
	NETDEV_PRE_CHANGEADDR,	/* notify before the address change */
2526 2527 2528 2529 2530 2531 2532 2533 2534 2535 2536 2537 2538 2539 2540 2541 2542 2543 2544 2545 2546
	NETDEV_GOING_DOWN,
	NETDEV_CHANGENAME,
	NETDEV_FEAT_CHANGE,
	NETDEV_BONDING_FAILOVER,
	NETDEV_PRE_UP,
	NETDEV_PRE_TYPE_CHANGE,
	NETDEV_POST_TYPE_CHANGE,
	NETDEV_POST_INIT,
	NETDEV_RELEASE,
	NETDEV_NOTIFY_PEERS,
	NETDEV_JOIN,
	NETDEV_CHANGEUPPER,
	NETDEV_RESEND_IGMP,
	NETDEV_PRECHANGEMTU,	/* notify before mtu change happened */
	NETDEV_CHANGEINFODATA,
	NETDEV_BONDING_INFO,
	NETDEV_PRECHANGEUPPER,
	NETDEV_CHANGELOWERSTATE,
	NETDEV_UDP_TUNNEL_PUSH_INFO,
	NETDEV_UDP_TUNNEL_DROP_INFO,
	NETDEV_CHANGE_TX_QUEUE_LEN,
2547 2548 2549 2550
	NETDEV_CVLAN_FILTER_PUSH_INFO,
	NETDEV_CVLAN_FILTER_DROP_INFO,
	NETDEV_SVLAN_FILTER_PUSH_INFO,
	NETDEV_SVLAN_FILTER_DROP_INFO,
2551 2552
};
const char *netdev_cmd_to_name(enum netdev_cmd cmd);
2553

2554 2555
int register_netdevice_notifier(struct notifier_block *nb);
int unregister_netdevice_notifier(struct notifier_block *nb);
2556 2557 2558
int register_netdevice_notifier_net(struct net *net, struct notifier_block *nb);
int unregister_netdevice_notifier_net(struct net *net,
				      struct notifier_block *nb);
2559 2560 2561 2562 2563 2564
int register_netdevice_notifier_dev_net(struct net_device *dev,
					struct notifier_block *nb,
					struct netdev_net_notifier *nn);
int unregister_netdevice_notifier_dev_net(struct net_device *dev,
					  struct notifier_block *nb,
					  struct netdev_net_notifier *nn);
2565 2566

struct netdev_notifier_info {
2567 2568
	struct net_device	*dev;
	struct netlink_ext_ack	*extack;
2569 2570
};

2571 2572 2573 2574 2575 2576 2577
struct netdev_notifier_info_ext {
	struct netdev_notifier_info info; /* must be first */
	union {
		u32 mtu;
	} ext;
};

2578 2579 2580 2581 2582
struct netdev_notifier_change_info {
	struct netdev_notifier_info info; /* must be first */
	unsigned int flags_changed;
};

2583 2584 2585 2586
struct netdev_notifier_changeupper_info {
	struct netdev_notifier_info info; /* must be first */
	struct net_device *upper_dev; /* new upper dev */
	bool master; /* is upper dev master */
B
Bjorn Helgaas 已提交
2587
	bool linking; /* is the notification for link or unlink */
2588
	void *upper_info; /* upper dev info */
2589 2590
};

2591 2592 2593 2594 2595
struct netdev_notifier_changelowerstate_info {
	struct netdev_notifier_info info; /* must be first */
	void *lower_state_info; /* is lower dev state */
};

P
Petr Machata 已提交
2596 2597 2598 2599 2600
struct netdev_notifier_pre_changeaddr_info {
	struct netdev_notifier_info info; /* must be first */
	const unsigned char *dev_addr;
};

2601 2602 2603 2604
static inline void netdev_notifier_info_init(struct netdev_notifier_info *info,
					     struct net_device *dev)
{
	info->dev = dev;
2605
	info->extack = NULL;
2606 2607
}

2608 2609 2610 2611 2612 2613
static inline struct net_device *
netdev_notifier_info_to_dev(const struct netdev_notifier_info *info)
{
	return info->dev;
}

2614 2615 2616 2617 2618 2619
static inline struct netlink_ext_ack *
netdev_notifier_info_to_extack(const struct netdev_notifier_info *info)
{
	return info->extack;
}

2620
int call_netdevice_notifiers(unsigned long val, struct net_device *dev);
2621 2622


L
Linus Torvalds 已提交
2623 2624
extern rwlock_t				dev_base_lock;		/* Device list lock */

2625 2626
#define for_each_netdev(net, d)		\
		list_for_each_entry(d, &(net)->dev_base_head, dev_list)
2627 2628
#define for_each_netdev_reverse(net, d)	\
		list_for_each_entry_reverse(d, &(net)->dev_base_head, dev_list)
2629 2630
#define for_each_netdev_rcu(net, d)		\
		list_for_each_entry_rcu(d, &(net)->dev_base_head, dev_list)
2631 2632 2633 2634
#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)
2635 2636 2637
#define for_each_netdev_continue_reverse(net, d)		\
		list_for_each_entry_continue_reverse(d, &(net)->dev_base_head, \
						     dev_list)
2638 2639
#define for_each_netdev_continue_rcu(net, d)		\
	list_for_each_entry_continue_rcu(d, &(net)->dev_base_head, dev_list)
2640 2641
#define for_each_netdev_in_bond_rcu(bond, slave)	\
		for_each_netdev_rcu(&init_net, slave)	\
2642
			if (netdev_master_upper_dev_get_rcu(slave) == (bond))
2643
#define net_device_entry(lh)	list_entry(lh, struct net_device, dev_list)
2644

2645 2646 2647 2648 2649
static inline struct net_device *next_net_device(struct net_device *dev)
{
	struct list_head *lh;
	struct net *net;

2650
	net = dev_net(dev);
2651 2652 2653 2654
	lh = dev->dev_list.next;
	return lh == &net->dev_base_head ? NULL : net_device_entry(lh);
}

2655 2656 2657 2658 2659 2660
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
2661
	lh = rcu_dereference(list_next_rcu(&dev->dev_list));
2662 2663 2664
	return lh == &net->dev_base_head ? NULL : net_device_entry(lh);
}

2665 2666 2667 2668 2669
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);
}
2670

E
Eric Dumazet 已提交
2671 2672 2673 2674 2675 2676 2677
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);
}

2678 2679 2680 2681 2682 2683 2684 2685 2686 2687 2688 2689
int netdev_boot_setup_check(struct net_device *dev);
unsigned long netdev_boot_base(const char *prefix, int unit);
struct net_device *dev_getbyhwaddr_rcu(struct net *net, unsigned short type,
				       const char *hwaddr);
struct net_device *dev_getfirstbyhwtype(struct net *net, unsigned short type);
struct net_device *__dev_getfirstbyhwtype(struct net *net, unsigned short type);
void dev_add_pack(struct packet_type *pt);
void dev_remove_pack(struct packet_type *pt);
void __dev_remove_pack(struct packet_type *pt);
void dev_add_offload(struct packet_offload *po);
void dev_remove_offload(struct packet_offload *po);

2690
int dev_get_iflink(const struct net_device *dev);
2691
int dev_fill_metadata_dst(struct net_device *dev, struct sk_buff *skb);
2692 2693
struct net_device *__dev_get_by_flags(struct net *net, unsigned short flags,
				      unsigned short mask);
2694 2695 2696 2697
struct net_device *dev_get_by_name(struct net *net, const char *name);
struct net_device *dev_get_by_name_rcu(struct net *net, const char *name);
struct net_device *__dev_get_by_name(struct net *net, const char *name);
int dev_alloc_name(struct net_device *dev, const char *name);
2698
int dev_open(struct net_device *dev, struct netlink_ext_ack *extack);
2699 2700
void dev_close(struct net_device *dev);
void dev_close_many(struct list_head *head, bool unlink);
2701
void dev_disable_lro(struct net_device *dev);
2702
int dev_loopback_xmit(struct net *net, struct sock *sk, struct sk_buff *newskb);
2703
u16 dev_pick_tx_zero(struct net_device *dev, struct sk_buff *skb,
2704
		     struct net_device *sb_dev);
2705
u16 dev_pick_tx_cpu_id(struct net_device *dev, struct sk_buff *skb,
2706
		       struct net_device *sb_dev);
2707
int dev_queue_xmit(struct sk_buff *skb);
2708
int dev_queue_xmit_accel(struct sk_buff *skb, struct net_device *sb_dev);
2709
int dev_direct_xmit(struct sk_buff *skb, u16 queue_id);
2710 2711 2712
int register_netdevice(struct net_device *dev);
void unregister_netdevice_queue(struct net_device *dev, struct list_head *head);
void unregister_netdevice_many(struct list_head *head);
2713 2714 2715 2716 2717
static inline void unregister_netdevice(struct net_device *dev)
{
	unregister_netdevice_queue(dev, NULL);
}

2718 2719
int netdev_refcnt_read(const struct net_device *dev);
void free_netdev(struct net_device *dev);
2720
void netdev_freemem(struct net_device *dev);
2721 2722
void synchronize_net(void);
int init_dummy_netdev(struct net_device *dev);
2723

2724 2725 2726
struct net_device *dev_get_by_index(struct net *net, int ifindex);
struct net_device *__dev_get_by_index(struct net *net, int ifindex);
struct net_device *dev_get_by_index_rcu(struct net *net, int ifindex);
2727
struct net_device *dev_get_by_napi_id(unsigned int napi_id);
2728 2729
int netdev_get_name(struct net *net, char *name, int ifindex);
int dev_restart(struct net_device *dev);
2730
int skb_gro_receive(struct sk_buff *p, struct sk_buff *skb);
2731
int skb_gro_receive_list(struct sk_buff *p, struct sk_buff *skb);
2732 2733 2734 2735 2736 2737 2738 2739 2740 2741 2742 2743 2744 2745 2746 2747

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;
}

2748 2749
static inline void *skb_gro_header_fast(struct sk_buff *skb,
					unsigned int offset)
2750
{
2751 2752
	return NAPI_GRO_CB(skb)->frag0 + offset;
}
2753

2754 2755 2756 2757
static inline int skb_gro_header_hard(struct sk_buff *skb, unsigned int hlen)
{
	return NAPI_GRO_CB(skb)->frag0_len < hlen;
}
2758

2759 2760 2761 2762 2763 2764
static inline void skb_gro_frag0_invalidate(struct sk_buff *skb)
{
	NAPI_GRO_CB(skb)->frag0 = NULL;
	NAPI_GRO_CB(skb)->frag0_len = 0;
}

2765 2766 2767
static inline void *skb_gro_header_slow(struct sk_buff *skb, unsigned int hlen,
					unsigned int offset)
{
2768 2769 2770
	if (!pskb_may_pull(skb, hlen))
		return NULL;

2771
	skb_gro_frag0_invalidate(skb);
2772
	return skb->data + offset;
2773
}
L
Linus Torvalds 已提交
2774

H
Herbert Xu 已提交
2775 2776
static inline void *skb_gro_network_header(struct sk_buff *skb)
{
2777 2778
	return (NAPI_GRO_CB(skb)->frag0 ?: skb->data) +
	       skb_network_offset(skb);
H
Herbert Xu 已提交
2779 2780
}

2781 2782 2783
static inline void skb_gro_postpull_rcsum(struct sk_buff *skb,
					const void *start, unsigned int len)
{
T
Tom Herbert 已提交
2784
	if (NAPI_GRO_CB(skb)->csum_valid)
2785 2786 2787 2788
		NAPI_GRO_CB(skb)->csum = csum_sub(NAPI_GRO_CB(skb)->csum,
						  csum_partial(start, len, 0));
}

T
Tom Herbert 已提交
2789 2790 2791 2792 2793 2794 2795
/* GRO checksum functions. These are logical equivalents of the normal
 * checksum functions (in skbuff.h) except that they operate on the GRO
 * offsets and fields in sk_buff.
 */

__sum16 __skb_gro_checksum_complete(struct sk_buff *skb);

2796 2797
static inline bool skb_at_gro_remcsum_start(struct sk_buff *skb)
{
2798
	return (NAPI_GRO_CB(skb)->gro_remcsum_start == skb_gro_offset(skb));
2799 2800
}

T
Tom Herbert 已提交
2801 2802 2803 2804
static inline bool __skb_gro_checksum_validate_needed(struct sk_buff *skb,
						      bool zero_okay,
						      __sum16 check)
{
2805 2806 2807
	return ((skb->ip_summed != CHECKSUM_PARTIAL ||
		skb_checksum_start_offset(skb) <
		 skb_gro_offset(skb)) &&
2808
		!skb_at_gro_remcsum_start(skb) &&
2809
		NAPI_GRO_CB(skb)->csum_cnt == 0 &&
T
Tom Herbert 已提交
2810 2811 2812 2813 2814 2815 2816 2817 2818 2819 2820 2821 2822 2823 2824 2825 2826
		(!zero_okay || check));
}

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

	NAPI_GRO_CB(skb)->csum = psum;

	return __skb_gro_checksum_complete(skb);
}

static inline void skb_gro_incr_csum_unnecessary(struct sk_buff *skb)
{
2827 2828 2829 2830 2831 2832 2833 2834 2835
	if (NAPI_GRO_CB(skb)->csum_cnt > 0) {
		/* Consume a checksum from CHECKSUM_UNNECESSARY */
		NAPI_GRO_CB(skb)->csum_cnt--;
	} else {
		/* Update skb for CHECKSUM_UNNECESSARY and csum_level when we
		 * verified a new top level checksum or an encapsulated one
		 * during GRO. This saves work if we fallback to normal path.
		 */
		__skb_incr_checksum_unnecessary(skb);
T
Tom Herbert 已提交
2836 2837 2838 2839 2840 2841 2842 2843 2844 2845
	}
}

#define __skb_gro_checksum_validate(skb, proto, zero_okay, check,	\
				    compute_pseudo)			\
({									\
	__sum16 __ret = 0;						\
	if (__skb_gro_checksum_validate_needed(skb, zero_okay, check))	\
		__ret = __skb_gro_checksum_validate_complete(skb,	\
				compute_pseudo(skb, proto));		\
2846
	if (!__ret)							\
T
Tom Herbert 已提交
2847 2848 2849 2850 2851 2852 2853 2854 2855 2856 2857 2858 2859 2860
		skb_gro_incr_csum_unnecessary(skb);			\
	__ret;								\
})

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

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

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

2861 2862 2863 2864 2865 2866 2867
static inline bool __skb_gro_checksum_convert_check(struct sk_buff *skb)
{
	return (NAPI_GRO_CB(skb)->csum_cnt == 0 &&
		!NAPI_GRO_CB(skb)->csum_valid);
}

static inline void __skb_gro_checksum_convert(struct sk_buff *skb,
2868
					      __wsum pseudo)
2869 2870 2871 2872 2873
{
	NAPI_GRO_CB(skb)->csum = ~pseudo;
	NAPI_GRO_CB(skb)->csum_valid = 1;
}

2874
#define skb_gro_checksum_try_convert(skb, proto, compute_pseudo)	\
2875 2876
do {									\
	if (__skb_gro_checksum_convert_check(skb))			\
2877
		__skb_gro_checksum_convert(skb, 			\
2878 2879 2880
					   compute_pseudo(skb, proto));	\
} while (0)

2881 2882 2883 2884 2885 2886 2887
struct gro_remcsum {
	int offset;
	__wsum delta;
};

static inline void skb_gro_remcsum_init(struct gro_remcsum *grc)
{
2888
	grc->offset = 0;
2889 2890 2891
	grc->delta = 0;
}

2892 2893 2894 2895 2896
static inline void *skb_gro_remcsum_process(struct sk_buff *skb, void *ptr,
					    unsigned int off, size_t hdrlen,
					    int start, int offset,
					    struct gro_remcsum *grc,
					    bool nopartial)
2897 2898
{
	__wsum delta;
2899
	size_t plen = hdrlen + max_t(size_t, offset + sizeof(u16), start);
2900 2901 2902

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

2903
	if (!nopartial) {
2904 2905 2906 2907 2908 2909 2910 2911 2912
		NAPI_GRO_CB(skb)->gro_remcsum_start = off + hdrlen + start;
		return ptr;
	}

	ptr = skb_gro_header_fast(skb, off);
	if (skb_gro_header_hard(skb, off + plen)) {
		ptr = skb_gro_header_slow(skb, off + plen, off);
		if (!ptr)
			return NULL;
2913 2914
	}

2915 2916
	delta = remcsum_adjust(ptr + hdrlen, NAPI_GRO_CB(skb)->csum,
			       start, offset);
2917 2918 2919

	/* Adjust skb->csum since we changed the packet */
	NAPI_GRO_CB(skb)->csum = csum_add(NAPI_GRO_CB(skb)->csum, delta);
2920

2921
	grc->offset = off + hdrlen + offset;
2922
	grc->delta = delta;
2923 2924

	return ptr;
2925 2926
}

2927 2928 2929
static inline void skb_gro_remcsum_cleanup(struct sk_buff *skb,
					   struct gro_remcsum *grc)
{
2930 2931 2932
	void *ptr;
	size_t plen = grc->offset + sizeof(u16);

2933 2934 2935
	if (!grc->delta)
		return;

2936 2937 2938 2939 2940 2941 2942 2943
	ptr = skb_gro_header_fast(skb, grc->offset);
	if (skb_gro_header_hard(skb, grc->offset + sizeof(u16))) {
		ptr = skb_gro_header_slow(skb, plen, grc->offset);
		if (!ptr)
			return;
	}

	remcsum_unadjust((__sum16 *)ptr, grc->delta);
2944
}
2945

2946
#ifdef CONFIG_XFRM_OFFLOAD
2947
static inline void skb_gro_flush_final(struct sk_buff *skb, struct sk_buff *pp, int flush)
2948 2949 2950 2951
{
	if (PTR_ERR(pp) != -EINPROGRESS)
		NAPI_GRO_CB(skb)->flush |= flush;
}
2952
static inline void skb_gro_flush_final_remcsum(struct sk_buff *skb,
2953
					       struct sk_buff *pp,
2954 2955 2956 2957 2958 2959 2960 2961 2962
					       int flush,
					       struct gro_remcsum *grc)
{
	if (PTR_ERR(pp) != -EINPROGRESS) {
		NAPI_GRO_CB(skb)->flush |= flush;
		skb_gro_remcsum_cleanup(skb, grc);
		skb->remcsum_offload = 0;
	}
}
2963
#else
2964
static inline void skb_gro_flush_final(struct sk_buff *skb, struct sk_buff *pp, int flush)
2965 2966 2967
{
	NAPI_GRO_CB(skb)->flush |= flush;
}
2968
static inline void skb_gro_flush_final_remcsum(struct sk_buff *skb,
2969
					       struct sk_buff *pp,
2970 2971 2972 2973 2974 2975 2976
					       int flush,
					       struct gro_remcsum *grc)
{
	NAPI_GRO_CB(skb)->flush |= flush;
	skb_gro_remcsum_cleanup(skb, grc);
	skb->remcsum_offload = 0;
}
2977
#endif
2978

2979 2980
static inline int dev_hard_header(struct sk_buff *skb, struct net_device *dev,
				  unsigned short type,
2981
				  const void *daddr, const void *saddr,
2982
				  unsigned int len)
2983
{
2984
	if (!dev->header_ops || !dev->header_ops->create)
2985
		return 0;
2986 2987

	return dev->header_ops->create(skb, dev, type, daddr, saddr, len);
2988 2989
}

S
Stephen Hemminger 已提交
2990 2991 2992 2993 2994
static inline int dev_parse_header(const struct sk_buff *skb,
				   unsigned char *haddr)
{
	const struct net_device *dev = skb->dev;

2995
	if (!dev->header_ops || !dev->header_ops->parse)
S
Stephen Hemminger 已提交
2996
		return 0;
2997
	return dev->header_ops->parse(skb, haddr);
S
Stephen Hemminger 已提交
2998 2999
}

3000 3001 3002 3003 3004 3005 3006 3007 3008
static inline __be16 dev_parse_header_protocol(const struct sk_buff *skb)
{
	const struct net_device *dev = skb->dev;

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

3009 3010 3011 3012 3013 3014
/* ll_header must have at least hard_header_len allocated */
static inline bool dev_validate_header(const struct net_device *dev,
				       char *ll_header, int len)
{
	if (likely(len >= dev->hard_header_len))
		return true;
3015 3016
	if (len < dev->min_header_len)
		return false;
3017 3018 3019 3020 3021 3022 3023 3024 3025 3026 3027 3028

	if (capable(CAP_SYS_RAWIO)) {
		memset(ll_header + len, 0, dev->hard_header_len - len);
		return true;
	}

	if (dev->header_ops && dev->header_ops->validate)
		return dev->header_ops->validate(ll_header, len);

	return false;
}

3029 3030
typedef int gifconf_func_t(struct net_device * dev, char __user * bufptr,
			   int len, int size);
3031
int register_gifconf(unsigned int family, gifconf_func_t *gifconf);
L
Linus Torvalds 已提交
3032 3033 3034 3035 3036
static inline int unregister_gifconf(unsigned int family)
{
	return register_gifconf(family, NULL);
}

3037
#ifdef CONFIG_NET_FLOW_LIMIT
3038
#define FLOW_LIMIT_HISTORY	(1 << 7)  /* must be ^2 and !overflow buckets */
3039 3040 3041 3042 3043 3044 3045 3046 3047 3048 3049
struct sd_flow_limit {
	u64			count;
	unsigned int		num_buckets;
	unsigned int		history_head;
	u16			history[FLOW_LIMIT_HISTORY];
	u8			buckets[];
};

extern int netdev_flow_limit_table_len;
#endif /* CONFIG_NET_FLOW_LIMIT */

L
Linus Torvalds 已提交
3050
/*
B
Bjorn Helgaas 已提交
3051
 * Incoming packets are placed on per-CPU queues
L
Linus Torvalds 已提交
3052
 */
E
Eric Dumazet 已提交
3053
struct softnet_data {
L
Linus Torvalds 已提交
3054
	struct list_head	poll_list;
3055
	struct sk_buff_head	process_queue;
L
Linus Torvalds 已提交
3056

C
Changli Gao 已提交
3057
	/* stats */
3058 3059 3060
	unsigned int		processed;
	unsigned int		time_squeeze;
	unsigned int		received_rps;
3061
#ifdef CONFIG_RPS
E
Eric Dumazet 已提交
3062
	struct softnet_data	*rps_ipi_list;
E
Eric Dumazet 已提交
3063 3064 3065 3066 3067 3068 3069
#endif
#ifdef CONFIG_NET_FLOW_LIMIT
	struct sd_flow_limit __rcu *flow_limit;
#endif
	struct Qdisc		*output_queue;
	struct Qdisc		**output_queue_tailp;
	struct sk_buff		*completion_queue;
3070 3071 3072
#ifdef CONFIG_XFRM_OFFLOAD
	struct sk_buff_head	xfrm_backlog;
#endif
3073 3074 3075 3076 3077
	/* written and read only by owning cpu: */
	struct {
		u16 recursion;
		u8  more;
	} xmit;
E
Eric Dumazet 已提交
3078
#ifdef CONFIG_RPS
3079 3080 3081 3082 3083 3084
	/* input_queue_head should be written by cpu owning this struct,
	 * and only read by other cpus. Worth using a cache line.
	 */
	unsigned int		input_queue_head ____cacheline_aligned_in_smp;

	/* Elements below can be accessed between CPUs for RPS/RFS */
3085
	call_single_data_t	csd ____cacheline_aligned_in_smp;
E
Eric Dumazet 已提交
3086 3087
	struct softnet_data	*rps_ipi_next;
	unsigned int		cpu;
3088
	unsigned int		input_queue_tail;
3089
#endif
3090
	unsigned int		dropped;
T
Tom Herbert 已提交
3091
	struct sk_buff_head	input_pkt_queue;
3092
	struct napi_struct	backlog;
3093

L
Linus Torvalds 已提交
3094 3095
};

3096
static inline void input_queue_head_incr(struct softnet_data *sd)
T
Tom Herbert 已提交
3097 3098
{
#ifdef CONFIG_RPS
3099 3100 3101 3102 3103 3104 3105 3106 3107
	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
3108 3109 3110
#endif
}

T
Tom Herbert 已提交
3111
DECLARE_PER_CPU_ALIGNED(struct softnet_data, softnet_data);
L
Linus Torvalds 已提交
3112

3113 3114
static inline int dev_recursion_level(void)
{
3115
	return this_cpu_read(softnet_data.xmit.recursion);
3116 3117 3118 3119 3120 3121 3122 3123 3124 3125 3126 3127 3128 3129 3130 3131 3132 3133 3134
}

#define XMIT_RECURSION_LIMIT	10
static inline bool dev_xmit_recursion(void)
{
	return unlikely(__this_cpu_read(softnet_data.xmit.recursion) >
			XMIT_RECURSION_LIMIT);
}

static inline void dev_xmit_recursion_inc(void)
{
	__this_cpu_inc(softnet_data.xmit.recursion);
}

static inline void dev_xmit_recursion_dec(void)
{
	__this_cpu_dec(softnet_data.xmit.recursion);
}

3135
void __netif_schedule(struct Qdisc *q);
3136
void netif_schedule_queue(struct netdev_queue *txq);
3137

3138 3139 3140 3141 3142 3143 3144 3145
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));
}

3146
static __always_inline void netif_tx_start_queue(struct netdev_queue *dev_queue)
3147
{
3148
	clear_bit(__QUEUE_STATE_DRV_XOFF, &dev_queue->state);
3149 3150
}

3151 3152 3153 3154 3155 3156
/**
 *	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
3157 3158
static inline void netif_start_queue(struct net_device *dev)
{
3159
	netif_tx_start_queue(netdev_get_tx_queue(dev, 0));
L
Linus Torvalds 已提交
3160 3161
}

3162 3163 3164 3165 3166 3167 3168 3169 3170 3171
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);
	}
}

3172
void netif_tx_wake_queue(struct netdev_queue *dev_queue);
3173

3174 3175 3176 3177 3178 3179 3180
/**
 *	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.
 */
3181 3182
static inline void netif_wake_queue(struct net_device *dev)
{
3183
	netif_tx_wake_queue(netdev_get_tx_queue(dev, 0));
L
Linus Torvalds 已提交
3184 3185
}

3186 3187 3188 3189 3190 3191 3192 3193 3194 3195
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);
	}
}

3196
static __always_inline void netif_tx_stop_queue(struct netdev_queue *dev_queue)
3197
{
3198
	set_bit(__QUEUE_STATE_DRV_XOFF, &dev_queue->state);
3199 3200
}

3201 3202 3203 3204 3205 3206 3207
/**
 *	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
3208 3209
static inline void netif_stop_queue(struct net_device *dev)
{
3210
	netif_tx_stop_queue(netdev_get_tx_queue(dev, 0));
L
Linus Torvalds 已提交
3211 3212
}

3213
void netif_tx_stop_all_queues(struct net_device *dev);
T
Taehee Yoo 已提交
3214
void netdev_update_lockdep_key(struct net_device *dev);
3215

3216
static inline bool netif_tx_queue_stopped(const struct netdev_queue *dev_queue)
3217
{
3218
	return test_bit(__QUEUE_STATE_DRV_XOFF, &dev_queue->state);
3219 3220
}

3221 3222 3223 3224 3225 3226
/**
 *	netif_queue_stopped - test if transmit queue is flowblocked
 *	@dev: network device
 *
 *	Test if transmit queue on device is currently unable to send.
 */
3227
static inline bool netif_queue_stopped(const struct net_device *dev)
L
Linus Torvalds 已提交
3228
{
3229
	return netif_tx_queue_stopped(netdev_get_tx_queue(dev, 0));
L
Linus Torvalds 已提交
3230 3231
}

3232
static inline bool netif_xmit_stopped(const struct netdev_queue *dev_queue)
3233
{
3234 3235 3236
	return dev_queue->state & QUEUE_STATE_ANY_XOFF;
}

3237 3238
static inline bool
netif_xmit_frozen_or_stopped(const struct netdev_queue *dev_queue)
3239 3240 3241 3242
{
	return dev_queue->state & QUEUE_STATE_ANY_XOFF_OR_FROZEN;
}

3243 3244 3245 3246 3247 3248
static inline bool
netif_xmit_frozen_or_drv_stopped(const struct netdev_queue *dev_queue)
{
	return dev_queue->state & QUEUE_STATE_DRV_XOFF_OR_FROZEN;
}

3249 3250 3251 3252 3253
/**
 *	netdev_txq_bql_enqueue_prefetchw - prefetch bql data for write
 *	@dev_queue: pointer to transmit queue
 *
 * BQL enabled drivers might use this helper in their ndo_start_xmit(),
B
Bjorn Helgaas 已提交
3254
 * to give appropriate hint to the CPU.
3255 3256 3257 3258 3259 3260 3261 3262 3263 3264 3265 3266 3267
 */
static inline void netdev_txq_bql_enqueue_prefetchw(struct netdev_queue *dev_queue)
{
#ifdef CONFIG_BQL
	prefetchw(&dev_queue->dql.num_queued);
#endif
}

/**
 *	netdev_txq_bql_complete_prefetchw - prefetch bql data for write
 *	@dev_queue: pointer to transmit queue
 *
 * BQL enabled drivers might use this helper in their TX completion path,
B
Bjorn Helgaas 已提交
3268
 * to give appropriate hint to the CPU.
3269 3270 3271 3272 3273 3274 3275 3276
 */
static inline void netdev_txq_bql_complete_prefetchw(struct netdev_queue *dev_queue)
{
#ifdef CONFIG_BQL
	prefetchw(&dev_queue->dql.limit);
#endif
}

3277 3278 3279
static inline void netdev_tx_sent_queue(struct netdev_queue *dev_queue,
					unsigned int bytes)
{
T
Tom Herbert 已提交
3280 3281
#ifdef CONFIG_BQL
	dql_queued(&dev_queue->dql, bytes);
3282 3283 3284 3285 3286 3287 3288 3289 3290 3291 3292 3293 3294 3295 3296 3297

	if (likely(dql_avail(&dev_queue->dql) >= 0))
		return;

	set_bit(__QUEUE_STATE_STACK_XOFF, &dev_queue->state);

	/*
	 * The XOFF flag must be set before checking the dql_avail below,
	 * because in netdev_tx_completed_queue we update the dql_completed
	 * before checking the XOFF flag.
	 */
	smp_mb();

	/* check again in case another CPU has just made room avail */
	if (unlikely(dql_avail(&dev_queue->dql) >= 0))
		clear_bit(__QUEUE_STATE_STACK_XOFF, &dev_queue->state);
T
Tom Herbert 已提交
3298
#endif
3299 3300
}

3301 3302 3303 3304 3305 3306 3307 3308 3309 3310 3311 3312 3313 3314 3315 3316 3317 3318 3319 3320
/* Variant of netdev_tx_sent_queue() for drivers that are aware
 * that they should not test BQL status themselves.
 * We do want to change __QUEUE_STATE_STACK_XOFF only for the last
 * skb of a batch.
 * Returns true if the doorbell must be used to kick the NIC.
 */
static inline bool __netdev_tx_sent_queue(struct netdev_queue *dev_queue,
					  unsigned int bytes,
					  bool xmit_more)
{
	if (xmit_more) {
#ifdef CONFIG_BQL
		dql_queued(&dev_queue->dql, bytes);
#endif
		return netif_tx_queue_stopped(dev_queue);
	}
	netdev_tx_sent_queue(dev_queue, bytes);
	return true;
}

3321 3322 3323 3324 3325 3326 3327 3328 3329
/**
 * 	netdev_sent_queue - report the number of bytes queued to hardware
 * 	@dev: network device
 * 	@bytes: number of bytes queued to the hardware device queue
 *
 * 	Report the number of bytes queued for sending/completion to the network
 * 	device hardware queue. @bytes should be a good approximation and should
 * 	exactly match netdev_completed_queue() @bytes
 */
3330 3331 3332 3333 3334
static inline void netdev_sent_queue(struct net_device *dev, unsigned int bytes)
{
	netdev_tx_sent_queue(netdev_get_tx_queue(dev, 0), bytes);
}

3335 3336 3337 3338 3339 3340 3341 3342
static inline bool __netdev_sent_queue(struct net_device *dev,
				       unsigned int bytes,
				       bool xmit_more)
{
	return __netdev_tx_sent_queue(netdev_get_tx_queue(dev, 0), bytes,
				      xmit_more);
}

3343
static inline void netdev_tx_completed_queue(struct netdev_queue *dev_queue,
3344
					     unsigned int pkts, unsigned int bytes)
3345
{
T
Tom Herbert 已提交
3346
#ifdef CONFIG_BQL
3347 3348 3349 3350 3351 3352 3353 3354 3355 3356 3357 3358
	if (unlikely(!bytes))
		return;

	dql_completed(&dev_queue->dql, bytes);

	/*
	 * Without the memory barrier there is a small possiblity that
	 * netdev_tx_sent_queue will miss the update and cause the queue to
	 * be stopped forever
	 */
	smp_mb();

3359
	if (unlikely(dql_avail(&dev_queue->dql) < 0))
3360 3361 3362 3363
		return;

	if (test_and_clear_bit(__QUEUE_STATE_STACK_XOFF, &dev_queue->state))
		netif_schedule_queue(dev_queue);
T
Tom Herbert 已提交
3364
#endif
3365 3366
}

3367 3368 3369 3370 3371 3372 3373 3374 3375 3376
/**
 * 	netdev_completed_queue - report bytes and packets completed by device
 * 	@dev: network device
 * 	@pkts: actual number of packets sent over the medium
 * 	@bytes: actual number of bytes sent over the medium
 *
 * 	Report the number of bytes and packets transmitted by the network device
 * 	hardware queue over the physical medium, @bytes must exactly match the
 * 	@bytes amount passed to netdev_sent_queue()
 */
3377
static inline void netdev_completed_queue(struct net_device *dev,
3378
					  unsigned int pkts, unsigned int bytes)
3379 3380 3381 3382 3383 3384
{
	netdev_tx_completed_queue(netdev_get_tx_queue(dev, 0), pkts, bytes);
}

static inline void netdev_tx_reset_queue(struct netdev_queue *q)
{
T
Tom Herbert 已提交
3385
#ifdef CONFIG_BQL
3386
	clear_bit(__QUEUE_STATE_STACK_XOFF, &q->state);
T
Tom Herbert 已提交
3387 3388
	dql_reset(&q->dql);
#endif
3389 3390
}

3391 3392 3393 3394 3395 3396 3397
/**
 * 	netdev_reset_queue - reset the packets and bytes count of a network device
 * 	@dev_queue: network device
 *
 * 	Reset the bytes and packet count of a network device and clear the
 * 	software flow control OFF bit for this network device
 */
3398 3399 3400
static inline void netdev_reset_queue(struct net_device *dev_queue)
{
	netdev_tx_reset_queue(netdev_get_tx_queue(dev_queue, 0));
3401 3402
}

3403 3404 3405 3406 3407 3408 3409 3410 3411 3412 3413 3414 3415 3416 3417 3418 3419 3420 3421 3422
/**
 * 	netdev_cap_txqueue - check if selected tx queue exceeds device queues
 * 	@dev: network device
 * 	@queue_index: given tx queue index
 *
 * 	Returns 0 if given tx queue index >= number of device tx queues,
 * 	otherwise returns the originally passed tx queue index.
 */
static inline u16 netdev_cap_txqueue(struct net_device *dev, u16 queue_index)
{
	if (unlikely(queue_index >= dev->real_num_tx_queues)) {
		net_warn_ratelimited("%s selects TX queue %d, but real number of TX queues is %d\n",
				     dev->name, queue_index,
				     dev->real_num_tx_queues);
		return 0;
	}

	return queue_index;
}

3423 3424 3425 3426 3427 3428
/**
 *	netif_running - test if up
 *	@dev: network device
 *
 *	Test if the device has been brought up.
 */
3429
static inline bool netif_running(const struct net_device *dev)
L
Linus Torvalds 已提交
3430 3431 3432 3433
{
	return test_bit(__LINK_STATE_START, &dev->state);
}

3434
/*
B
Bjorn Helgaas 已提交
3435
 * Routines to manage the subqueues on a device.  We only need start,
3436 3437 3438 3439
 * 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.
 */
3440 3441 3442 3443 3444 3445 3446 3447

/**
 *	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.
 */
3448 3449
static inline void netif_start_subqueue(struct net_device *dev, u16 queue_index)
{
3450
	struct netdev_queue *txq = netdev_get_tx_queue(dev, queue_index);
3451 3452

	netif_tx_start_queue(txq);
3453 3454
}

3455 3456 3457 3458 3459 3460 3461
/**
 *	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.
 */
3462 3463
static inline void netif_stop_subqueue(struct net_device *dev, u16 queue_index)
{
3464
	struct netdev_queue *txq = netdev_get_tx_queue(dev, queue_index);
3465
	netif_tx_stop_queue(txq);
3466 3467
}

3468 3469 3470 3471 3472 3473 3474
/**
 *	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.
 */
3475 3476
static inline bool __netif_subqueue_stopped(const struct net_device *dev,
					    u16 queue_index)
3477
{
3478
	struct netdev_queue *txq = netdev_get_tx_queue(dev, queue_index);
3479 3480

	return netif_tx_queue_stopped(txq);
3481 3482
}

3483 3484
static inline bool netif_subqueue_stopped(const struct net_device *dev,
					  struct sk_buff *skb)
3485 3486 3487
{
	return __netif_subqueue_stopped(dev, skb_get_queue_mapping(skb));
}
3488

3489 3490 3491 3492 3493 3494 3495 3496 3497 3498 3499 3500 3501
/**
 *	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.
 */
static inline void netif_wake_subqueue(struct net_device *dev, u16 queue_index)
{
	struct netdev_queue *txq = netdev_get_tx_queue(dev, queue_index);

	netif_tx_wake_queue(txq);
}
3502

3503
#ifdef CONFIG_XPS
3504
int netif_set_xps_queue(struct net_device *dev, const struct cpumask *mask,
3505
			u16 index);
3506 3507 3508 3509 3510 3511 3512 3513 3514 3515 3516 3517 3518 3519 3520 3521 3522 3523 3524 3525 3526 3527 3528 3529 3530 3531 3532 3533 3534 3535 3536 3537 3538 3539 3540 3541 3542 3543 3544 3545 3546 3547 3548 3549 3550 3551 3552 3553 3554 3555 3556 3557 3558 3559 3560 3561 3562 3563 3564 3565 3566
int __netif_set_xps_queue(struct net_device *dev, const unsigned long *mask,
			  u16 index, bool is_rxqs_map);

/**
 *	netif_attr_test_mask - Test a CPU or Rx queue set in a mask
 *	@j: CPU/Rx queue index
 *	@mask: bitmask of all cpus/rx queues
 *	@nr_bits: number of bits in the bitmask
 *
 * Test if a CPU or Rx queue index is set in a mask of all CPU/Rx queues.
 */
static inline bool netif_attr_test_mask(unsigned long j,
					const unsigned long *mask,
					unsigned int nr_bits)
{
	cpu_max_bits_warn(j, nr_bits);
	return test_bit(j, mask);
}

/**
 *	netif_attr_test_online - Test for online CPU/Rx queue
 *	@j: CPU/Rx queue index
 *	@online_mask: bitmask for CPUs/Rx queues that are online
 *	@nr_bits: number of bits in the bitmask
 *
 * Returns true if a CPU/Rx queue is online.
 */
static inline bool netif_attr_test_online(unsigned long j,
					  const unsigned long *online_mask,
					  unsigned int nr_bits)
{
	cpu_max_bits_warn(j, nr_bits);

	if (online_mask)
		return test_bit(j, online_mask);

	return (j < nr_bits);
}

/**
 *	netif_attrmask_next - get the next CPU/Rx queue in a cpu/Rx queues mask
 *	@n: CPU/Rx queue index
 *	@srcp: the cpumask/Rx queue mask pointer
 *	@nr_bits: number of bits in the bitmask
 *
 * Returns >= nr_bits if no further CPUs/Rx queues set.
 */
static inline unsigned int netif_attrmask_next(int n, const unsigned long *srcp,
					       unsigned int nr_bits)
{
	/* -1 is a legal arg here. */
	if (n != -1)
		cpu_max_bits_warn(n, nr_bits);

	if (srcp)
		return find_next_bit(srcp, nr_bits, n + 1);

	return n + 1;
}

/**
3567
 *	netif_attrmask_next_and - get the next CPU/Rx queue in \*src1p & \*src2p
3568 3569 3570 3571 3572 3573 3574 3575 3576 3577 3578 3579 3580 3581 3582 3583 3584 3585 3586 3587 3588 3589 3590 3591
 *	@n: CPU/Rx queue index
 *	@src1p: the first CPUs/Rx queues mask pointer
 *	@src2p: the second CPUs/Rx queues mask pointer
 *	@nr_bits: number of bits in the bitmask
 *
 * Returns >= nr_bits if no further CPUs/Rx queues set in both.
 */
static inline int netif_attrmask_next_and(int n, const unsigned long *src1p,
					  const unsigned long *src2p,
					  unsigned int nr_bits)
{
	/* -1 is a legal arg here. */
	if (n != -1)
		cpu_max_bits_warn(n, nr_bits);

	if (src1p && src2p)
		return find_next_and_bit(src1p, src2p, nr_bits, n + 1);
	else if (src1p)
		return find_next_bit(src1p, nr_bits, n + 1);
	else if (src2p)
		return find_next_bit(src2p, nr_bits, n + 1);

	return n + 1;
}
3592 3593
#else
static inline int netif_set_xps_queue(struct net_device *dev,
3594
				      const struct cpumask *mask,
3595 3596 3597 3598
				      u16 index)
{
	return 0;
}
3599 3600 3601 3602 3603 3604 3605

static inline int __netif_set_xps_queue(struct net_device *dev,
					const unsigned long *mask,
					u16 index, bool is_rxqs_map)
{
	return 0;
}
3606 3607
#endif

3608 3609 3610 3611 3612 3613
/**
 *	netif_is_multiqueue - test if device has multiple transmit queues
 *	@dev: network device
 *
 * Check if device has multiple transmit queues
 */
3614
static inline bool netif_is_multiqueue(const struct net_device *dev)
3615
{
E
Eric Dumazet 已提交
3616
	return dev->num_tx_queues > 1;
3617
}
L
Linus Torvalds 已提交
3618

3619
int netif_set_real_num_tx_queues(struct net_device *dev, unsigned int txq);
3620

3621
#ifdef CONFIG_SYSFS
3622
int netif_set_real_num_rx_queues(struct net_device *dev, unsigned int rxq);
3623 3624
#else
static inline int netif_set_real_num_rx_queues(struct net_device *dev,
3625
						unsigned int rxqs)
3626
{
3627
	dev->real_num_rx_queues = rxqs;
3628 3629 3630 3631
	return 0;
}
#endif

3632 3633 3634 3635 3636 3637
static inline struct netdev_rx_queue *
__netif_get_rx_queue(struct net_device *dev, unsigned int rxq)
{
	return dev->_rx + rxq;
}

3638 3639 3640 3641 3642 3643 3644 3645 3646 3647 3648 3649
#ifdef CONFIG_SYSFS
static inline unsigned int get_netdev_rx_queue_index(
		struct netdev_rx_queue *queue)
{
	struct net_device *dev = queue->dev;
	int index = queue - dev->_rx;

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

3650
#define DEFAULT_MAX_NUM_RSS_QUEUES	(8)
3651
int netif_get_num_default_rss_queues(void);
3652

3653 3654 3655 3656 3657 3658 3659
enum skb_free_reason {
	SKB_REASON_CONSUMED,
	SKB_REASON_DROPPED,
};

void __dev_kfree_skb_irq(struct sk_buff *skb, enum skb_free_reason reason);
void __dev_kfree_skb_any(struct sk_buff *skb, enum skb_free_reason reason);
L
Linus Torvalds 已提交
3660

3661 3662 3663 3664 3665 3666 3667 3668 3669 3670 3671 3672 3673 3674 3675 3676 3677 3678
/*
 * It is not allowed to call kfree_skb() or consume_skb() from hardware
 * interrupt context or with hardware interrupts being disabled.
 * (in_irq() || irqs_disabled())
 *
 * We provide four helpers that can be used in following contexts :
 *
 * dev_kfree_skb_irq(skb) when caller drops a packet from irq context,
 *  replacing kfree_skb(skb)
 *
 * dev_consume_skb_irq(skb) when caller consumes a packet from irq context.
 *  Typically used in place of consume_skb(skb) in TX completion path
 *
 * dev_kfree_skb_any(skb) when caller doesn't know its current irq context,
 *  replacing kfree_skb(skb)
 *
 * dev_consume_skb_any(skb) when caller doesn't know its current irq context,
 *  and consumed a packet. Used in place of consume_skb(skb)
L
Linus Torvalds 已提交
3679
 */
3680 3681 3682 3683 3684 3685 3686 3687 3688 3689 3690 3691 3692 3693 3694 3695 3696 3697 3698
static inline void dev_kfree_skb_irq(struct sk_buff *skb)
{
	__dev_kfree_skb_irq(skb, SKB_REASON_DROPPED);
}

static inline void dev_consume_skb_irq(struct sk_buff *skb)
{
	__dev_kfree_skb_irq(skb, SKB_REASON_CONSUMED);
}

static inline void dev_kfree_skb_any(struct sk_buff *skb)
{
	__dev_kfree_skb_any(skb, SKB_REASON_DROPPED);
}

static inline void dev_consume_skb_any(struct sk_buff *skb)
{
	__dev_kfree_skb_any(skb, SKB_REASON_CONSUMED);
}
L
Linus Torvalds 已提交
3699

3700 3701
void generic_xdp_tx(struct sk_buff *skb, struct bpf_prog *xdp_prog);
int do_xdp_generic(struct bpf_prog *xdp_prog, struct sk_buff *skb);
3702 3703
int netif_rx(struct sk_buff *skb);
int netif_rx_ni(struct sk_buff *skb);
3704
int netif_receive_skb(struct sk_buff *skb);
3705
int netif_receive_skb_core(struct sk_buff *skb);
3706
void netif_receive_skb_list(struct list_head *head);
3707 3708 3709 3710
gro_result_t napi_gro_receive(struct napi_struct *napi, struct sk_buff *skb);
void napi_gro_flush(struct napi_struct *napi, bool flush_old);
struct sk_buff *napi_get_frags(struct napi_struct *napi);
gro_result_t napi_gro_frags(struct napi_struct *napi);
3711 3712
struct packet_offload *gro_find_receive_by_type(__be16 type);
struct packet_offload *gro_find_complete_by_type(__be16 type);
3713 3714 3715 3716 3717 3718 3719

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

M
Mahesh Bandewar 已提交
3720
bool netdev_is_rx_handler_busy(struct net_device *dev);
3721 3722 3723 3724 3725 3726
int netdev_rx_handler_register(struct net_device *dev,
			       rx_handler_func_t *rx_handler,
			       void *rx_handler_data);
void netdev_rx_handler_unregister(struct net_device *dev);

bool dev_valid_name(const char *name);
3727 3728
int dev_ioctl(struct net *net, unsigned int cmd, struct ifreq *ifr,
		bool *need_copyout);
3729
int dev_ifconf(struct net *net, struct ifconf *, int);
3730 3731
int dev_ethtool(struct net *net, struct ifreq *);
unsigned int dev_get_flags(const struct net_device *);
3732 3733
int __dev_change_flags(struct net_device *dev, unsigned int flags,
		       struct netlink_ext_ack *extack);
3734 3735
int dev_change_flags(struct net_device *dev, unsigned int flags,
		     struct netlink_ext_ack *extack);
3736 3737
void __dev_notify_flags(struct net_device *, unsigned int old_flags,
			unsigned int gchanges);
3738 3739
int dev_change_name(struct net_device *, const char *);
int dev_set_alias(struct net_device *, const char *, size_t);
3740
int dev_get_alias(const struct net_device *, char *, size_t);
3741
int dev_change_net_namespace(struct net_device *, struct net *, const char *);
3742
int __dev_set_mtu(struct net_device *, int);
3743 3744
int dev_validate_mtu(struct net_device *dev, int mtu,
		     struct netlink_ext_ack *extack);
3745 3746
int dev_set_mtu_ext(struct net_device *dev, int mtu,
		    struct netlink_ext_ack *extack);
3747
int dev_set_mtu(struct net_device *, int);
3748
int dev_change_tx_queue_len(struct net_device *, unsigned long);
3749
void dev_set_group(struct net_device *, int);
3750 3751
int dev_pre_changeaddr_notify(struct net_device *dev, const char *addr,
			      struct netlink_ext_ack *extack);
3752 3753
int dev_set_mac_address(struct net_device *dev, struct sockaddr *sa,
			struct netlink_ext_ack *extack);
3754 3755
int dev_change_carrier(struct net_device *, bool new_carrier);
int dev_get_phys_port_id(struct net_device *dev,
3756
			 struct netdev_phys_item_id *ppid);
3757 3758
int dev_get_phys_port_name(struct net_device *dev,
			   char *name, size_t len);
3759 3760 3761
int dev_get_port_parent_id(struct net_device *dev,
			   struct netdev_phys_item_id *ppid, bool recurse);
bool netdev_port_same_parent_id(struct net_device *a, struct net_device *b);
3762
int dev_change_proto_down(struct net_device *dev, bool proto_down);
3763
int dev_change_proto_down_generic(struct net_device *dev, bool proto_down);
3764
struct sk_buff *validate_xmit_skb_list(struct sk_buff *skb, struct net_device *dev, bool *again);
3765 3766
struct sk_buff *dev_hard_start_xmit(struct sk_buff *skb, struct net_device *dev,
				    struct netdev_queue *txq, int *ret);
3767

3768
typedef int (*bpf_op_t)(struct net_device *dev, struct netdev_bpf *bpf);
3769 3770
int dev_change_xdp_fd(struct net_device *dev, struct netlink_ext_ack *extack,
		      int fd, u32 flags);
3771 3772
u32 __dev_xdp_query(struct net_device *dev, bpf_op_t xdp_op,
		    enum bpf_netdev_command cmd);
3773
int xdp_umem_query(struct net_device *dev, u16 queue_id);
3774

H
Herbert Xu 已提交
3775
int __dev_forward_skb(struct net_device *dev, struct sk_buff *skb);
3776
int dev_forward_skb(struct net_device *dev, struct sk_buff *skb);
3777 3778
bool is_skb_forwardable(const struct net_device *dev,
			const struct sk_buff *skb);
L
Linus Torvalds 已提交
3779

3780 3781 3782 3783 3784 3785 3786 3787 3788 3789 3790 3791 3792 3793 3794
static __always_inline int ____dev_forward_skb(struct net_device *dev,
					       struct sk_buff *skb)
{
	if (skb_orphan_frags(skb, GFP_ATOMIC) ||
	    unlikely(!is_skb_forwardable(dev, skb))) {
		atomic_long_inc(&dev->rx_dropped);
		kfree_skb(skb);
		return NET_RX_DROP;
	}

	skb_scrub_packet(skb, true);
	skb->priority = 0;
	return 0;
}

3795
bool dev_nit_active(struct net_device *dev);
3796 3797
void dev_queue_xmit_nit(struct sk_buff *skb, struct net_device *dev);

3798
extern int		netdev_budget;
3799
extern unsigned int	netdev_budget_usecs;
L
Linus Torvalds 已提交
3800 3801

/* Called by rtnetlink.c:rtnl_unlock() */
3802
void netdev_run_todo(void);
L
Linus Torvalds 已提交
3803

3804 3805 3806 3807
/**
 *	dev_put - release reference to device
 *	@dev: network device
 *
3808
 * Release reference to device to allow it to be freed.
3809
 */
L
Linus Torvalds 已提交
3810 3811
static inline void dev_put(struct net_device *dev)
{
3812
	this_cpu_dec(*dev->pcpu_refcnt);
L
Linus Torvalds 已提交
3813 3814
}

3815 3816 3817 3818
/**
 *	dev_hold - get reference to device
 *	@dev: network device
 *
3819
 * Hold reference to device to keep it from being freed.
3820
 */
3821 3822
static inline void dev_hold(struct net_device *dev)
{
3823
	this_cpu_inc(*dev->pcpu_refcnt);
3824
}
L
Linus Torvalds 已提交
3825 3826 3827 3828

/* 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
3829 3830 3831 3832
 *
 * 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
3833 3834
 */

3835 3836 3837
void linkwatch_init_dev(struct net_device *dev);
void linkwatch_fire_event(struct net_device *dev);
void linkwatch_forget_dev(struct net_device *dev);
L
Linus Torvalds 已提交
3838

3839 3840 3841 3842 3843 3844
/**
 *	netif_carrier_ok - test if carrier present
 *	@dev: network device
 *
 * Check if carrier is present on device
 */
3845
static inline bool netif_carrier_ok(const struct net_device *dev)
L
Linus Torvalds 已提交
3846 3847 3848 3849
{
	return !test_bit(__LINK_STATE_NOCARRIER, &dev->state);
}

3850
unsigned long dev_trans_start(struct net_device *dev);
3851

3852
void __netdev_watchdog_up(struct net_device *dev);
L
Linus Torvalds 已提交
3853

3854
void netif_carrier_on(struct net_device *dev);
L
Linus Torvalds 已提交
3855

3856
void netif_carrier_off(struct net_device *dev);
L
Linus Torvalds 已提交
3857

3858 3859 3860 3861 3862 3863 3864 3865 3866 3867 3868 3869
/**
 *	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
3870 3871 3872 3873 3874 3875
static inline void netif_dormant_on(struct net_device *dev)
{
	if (!test_and_set_bit(__LINK_STATE_DORMANT, &dev->state))
		linkwatch_fire_event(dev);
}

3876 3877 3878 3879 3880 3881
/**
 *	netif_dormant_off - set device as not dormant.
 *	@dev: network device
 *
 * Device is not in dormant state.
 */
S
Stefan Rompf 已提交
3882 3883 3884 3885 3886 3887
static inline void netif_dormant_off(struct net_device *dev)
{
	if (test_and_clear_bit(__LINK_STATE_DORMANT, &dev->state))
		linkwatch_fire_event(dev);
}

3888
/**
3889
 *	netif_dormant - test if device is dormant
3890 3891
 *	@dev: network device
 *
3892
 * Check if device is dormant.
3893
 */
3894
static inline bool netif_dormant(const struct net_device *dev)
S
Stefan Rompf 已提交
3895 3896 3897 3898 3899
{
	return test_bit(__LINK_STATE_DORMANT, &dev->state);
}


3900 3901 3902 3903 3904 3905
/**
 *	netif_oper_up - test if device is operational
 *	@dev: network device
 *
 * Check if carrier is operational
 */
3906
static inline bool netif_oper_up(const struct net_device *dev)
E
Eric Dumazet 已提交
3907
{
S
Stefan Rompf 已提交
3908 3909 3910 3911
	return (dev->operstate == IF_OPER_UP ||
		dev->operstate == IF_OPER_UNKNOWN /* backward compat */);
}

3912 3913 3914 3915 3916 3917
/**
 *	netif_device_present - is device available or removed
 *	@dev: network device
 *
 * Check if device has not been removed from system.
 */
3918
static inline bool netif_device_present(struct net_device *dev)
L
Linus Torvalds 已提交
3919 3920 3921 3922
{
	return test_bit(__LINK_STATE_PRESENT, &dev->state);
}

3923
void netif_device_detach(struct net_device *dev);
L
Linus Torvalds 已提交
3924

3925
void netif_device_attach(struct net_device *dev);
L
Linus Torvalds 已提交
3926 3927 3928 3929 3930 3931

/*
 * Network interface message level settings
 */

enum {
3932 3933 3934 3935 3936 3937 3938 3939 3940 3941 3942 3943 3944 3945 3946 3947 3948 3949 3950 3951
	NETIF_MSG_DRV_BIT,
	NETIF_MSG_PROBE_BIT,
	NETIF_MSG_LINK_BIT,
	NETIF_MSG_TIMER_BIT,
	NETIF_MSG_IFDOWN_BIT,
	NETIF_MSG_IFUP_BIT,
	NETIF_MSG_RX_ERR_BIT,
	NETIF_MSG_TX_ERR_BIT,
	NETIF_MSG_TX_QUEUED_BIT,
	NETIF_MSG_INTR_BIT,
	NETIF_MSG_TX_DONE_BIT,
	NETIF_MSG_RX_STATUS_BIT,
	NETIF_MSG_PKTDATA_BIT,
	NETIF_MSG_HW_BIT,
	NETIF_MSG_WOL_BIT,

	/* When you add a new bit above, update netif_msg_class_names array
	 * in net/ethtool/common.c
	 */
	NETIF_MSG_CLASS_COUNT,
L
Linus Torvalds 已提交
3952
};
3953 3954 3955 3956 3957 3958 3959 3960 3961 3962 3963 3964 3965 3966 3967 3968 3969 3970 3971 3972 3973
/* Both ethtool_ops interface and internal driver implementation use u32 */
static_assert(NETIF_MSG_CLASS_COUNT <= 32);

#define __NETIF_MSG_BIT(bit)	((u32)1 << (bit))
#define __NETIF_MSG(name)	__NETIF_MSG_BIT(NETIF_MSG_ ## name ## _BIT)

#define NETIF_MSG_DRV		__NETIF_MSG(DRV)
#define NETIF_MSG_PROBE		__NETIF_MSG(PROBE)
#define NETIF_MSG_LINK		__NETIF_MSG(LINK)
#define NETIF_MSG_TIMER		__NETIF_MSG(TIMER)
#define NETIF_MSG_IFDOWN	__NETIF_MSG(IFDOWN)
#define NETIF_MSG_IFUP		__NETIF_MSG(IFUP)
#define NETIF_MSG_RX_ERR	__NETIF_MSG(RX_ERR)
#define NETIF_MSG_TX_ERR	__NETIF_MSG(TX_ERR)
#define NETIF_MSG_TX_QUEUED	__NETIF_MSG(TX_QUEUED)
#define NETIF_MSG_INTR		__NETIF_MSG(INTR)
#define NETIF_MSG_TX_DONE	__NETIF_MSG(TX_DONE)
#define NETIF_MSG_RX_STATUS	__NETIF_MSG(RX_STATUS)
#define NETIF_MSG_PKTDATA	__NETIF_MSG(PKTDATA)
#define NETIF_MSG_HW		__NETIF_MSG(HW)
#define NETIF_MSG_WOL		__NETIF_MSG(WOL)
L
Linus Torvalds 已提交
3974 3975 3976 3977 3978 3979 3980 3981 3982 3983 3984 3985 3986 3987 3988 3989 3990 3991 3992 3993 3994 3995 3996 3997 3998

#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 */
3999
	return (1U << debug_value) - 1;
L
Linus Torvalds 已提交
4000 4001
}

4002
static inline void __netif_tx_lock(struct netdev_queue *txq, int cpu)
H
Herbert Xu 已提交
4003
{
4004 4005
	spin_lock(&txq->_xmit_lock);
	txq->xmit_lock_owner = cpu;
4006 4007
}

4008 4009 4010 4011 4012 4013 4014 4015 4016 4017 4018
static inline bool __netif_tx_acquire(struct netdev_queue *txq)
{
	__acquire(&txq->_xmit_lock);
	return true;
}

static inline void __netif_tx_release(struct netdev_queue *txq)
{
	__release(&txq->_xmit_lock);
}

4019 4020 4021 4022 4023 4024
static inline void __netif_tx_lock_bh(struct netdev_queue *txq)
{
	spin_lock_bh(&txq->_xmit_lock);
	txq->xmit_lock_owner = smp_processor_id();
}

4025
static inline bool __netif_tx_trylock(struct netdev_queue *txq)
4026
{
4027
	bool ok = spin_trylock(&txq->_xmit_lock);
4028 4029 4030 4031 4032 4033 4034 4035 4036 4037 4038 4039 4040 4041 4042 4043 4044
	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
4045 4046 4047 4048 4049 4050
static inline void txq_trans_update(struct netdev_queue *txq)
{
	if (txq->xmit_lock_owner != -1)
		txq->trans_start = jiffies;
}

4051 4052 4053
/* legacy drivers only, netdev_start_xmit() sets txq->trans_start */
static inline void netif_trans_update(struct net_device *dev)
{
F
Florian Westphal 已提交
4054 4055 4056 4057
	struct netdev_queue *txq = netdev_get_tx_queue(dev, 0);

	if (txq->trans_start != jiffies)
		txq->trans_start = jiffies;
4058 4059
}

4060 4061 4062 4063 4064 4065
/**
 *	netif_tx_lock - grab network device transmit lock
 *	@dev: network device
 *
 * Get network device transmit lock
 */
4066 4067
static inline void netif_tx_lock(struct net_device *dev)
{
4068
	unsigned int i;
4069
	int cpu;
4070

4071 4072
	spin_lock(&dev->tx_global_lock);
	cpu = smp_processor_id();
4073 4074
	for (i = 0; i < dev->num_tx_queues; i++) {
		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
4075 4076 4077 4078 4079 4080 4081

		/* 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.
		 */
4082
		__netif_tx_lock(txq, cpu);
4083 4084
		set_bit(__QUEUE_STATE_FROZEN, &txq->state);
		__netif_tx_unlock(txq);
4085
	}
H
Herbert Xu 已提交
4086 4087 4088 4089
}

static inline void netif_tx_lock_bh(struct net_device *dev)
{
4090 4091
	local_bh_disable();
	netif_tx_lock(dev);
H
Herbert Xu 已提交
4092 4093 4094 4095
}

static inline void netif_tx_unlock(struct net_device *dev)
{
4096 4097 4098 4099
	unsigned int i;

	for (i = 0; i < dev->num_tx_queues; i++) {
		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
4100

4101 4102 4103 4104 4105
		/* 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);
4106
		netif_schedule_queue(txq);
4107 4108
	}
	spin_unlock(&dev->tx_global_lock);
H
Herbert Xu 已提交
4109 4110 4111 4112
}

static inline void netif_tx_unlock_bh(struct net_device *dev)
{
4113 4114
	netif_tx_unlock(dev);
	local_bh_enable();
H
Herbert Xu 已提交
4115 4116
}

4117
#define HARD_TX_LOCK(dev, txq, cpu) {			\
4118
	if ((dev->features & NETIF_F_LLTX) == 0) {	\
4119
		__netif_tx_lock(txq, cpu);		\
4120 4121
	} else {					\
		__netif_tx_acquire(txq);		\
4122 4123 4124
	}						\
}

4125 4126 4127
#define HARD_TX_TRYLOCK(dev, txq)			\
	(((dev->features & NETIF_F_LLTX) == 0) ?	\
		__netif_tx_trylock(txq) :		\
4128
		__netif_tx_acquire(txq))
4129

4130
#define HARD_TX_UNLOCK(dev, txq) {			\
4131
	if ((dev->features & NETIF_F_LLTX) == 0) {	\
4132
		__netif_tx_unlock(txq);			\
4133 4134
	} else {					\
		__netif_tx_release(txq);		\
4135 4136 4137
	}						\
}

L
Linus Torvalds 已提交
4138 4139
static inline void netif_tx_disable(struct net_device *dev)
{
4140
	unsigned int i;
4141
	int cpu;
4142

4143 4144
	local_bh_disable();
	cpu = smp_processor_id();
4145 4146
	for (i = 0; i < dev->num_tx_queues; i++) {
		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
4147 4148

		__netif_tx_lock(txq, cpu);
4149
		netif_tx_stop_queue(txq);
4150
		__netif_tx_unlock(txq);
4151
	}
4152
	local_bh_enable();
L
Linus Torvalds 已提交
4153 4154
}

4155 4156 4157 4158 4159 4160 4161 4162 4163 4164 4165 4166 4167 4168 4169 4170 4171 4172 4173 4174
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);
}

4175
/*
4176
 * dev_addrs walker. Should be used only for read access. Call with
4177 4178 4179
 * rcu_read_lock held.
 */
#define for_each_dev_addr(dev, ha) \
4180
		list_for_each_entry_rcu(ha, &dev->dev_addrs.list, list)
4181

L
Linus Torvalds 已提交
4182 4183
/* These functions live elsewhere (drivers/net/net_init.c, but related) */

4184
void ether_setup(struct net_device *dev);
L
Linus Torvalds 已提交
4185 4186

/* Support for loadable net-drivers */
4187
struct net_device *alloc_netdev_mqs(int sizeof_priv, const char *name,
4188
				    unsigned char name_assign_type,
4189 4190
				    void (*setup)(struct net_device *),
				    unsigned int txqs, unsigned int rxqs);
4191 4192
#define alloc_netdev(sizeof_priv, name, name_assign_type, setup) \
	alloc_netdev_mqs(sizeof_priv, name, name_assign_type, setup, 1, 1)
T
Tom Herbert 已提交
4193

4194 4195 4196
#define alloc_netdev_mq(sizeof_priv, name, name_assign_type, setup, count) \
	alloc_netdev_mqs(sizeof_priv, name, name_assign_type, setup, count, \
			 count)
T
Tom Herbert 已提交
4197

4198 4199
int register_netdev(struct net_device *dev);
void unregister_netdev(struct net_device *dev);
4200

4201
/* General hardware address lists handling functions */
4202 4203 4204 4205
int __hw_addr_sync(struct netdev_hw_addr_list *to_list,
		   struct netdev_hw_addr_list *from_list, int addr_len);
void __hw_addr_unsync(struct netdev_hw_addr_list *to_list,
		      struct netdev_hw_addr_list *from_list, int addr_len);
4206 4207 4208 4209 4210
int __hw_addr_sync_dev(struct netdev_hw_addr_list *list,
		       struct net_device *dev,
		       int (*sync)(struct net_device *, const unsigned char *),
		       int (*unsync)(struct net_device *,
				     const unsigned char *));
4211 4212 4213 4214 4215 4216 4217 4218 4219 4220
int __hw_addr_ref_sync_dev(struct netdev_hw_addr_list *list,
			   struct net_device *dev,
			   int (*sync)(struct net_device *,
				       const unsigned char *, int),
			   int (*unsync)(struct net_device *,
					 const unsigned char *, int));
void __hw_addr_ref_unsync_dev(struct netdev_hw_addr_list *list,
			      struct net_device *dev,
			      int (*unsync)(struct net_device *,
					    const unsigned char *, int));
4221 4222 4223 4224
void __hw_addr_unsync_dev(struct netdev_hw_addr_list *list,
			  struct net_device *dev,
			  int (*unsync)(struct net_device *,
					const unsigned char *));
4225
void __hw_addr_init(struct netdev_hw_addr_list *list);
4226

4227
/* Functions used for device addresses handling */
4228 4229 4230 4231 4232 4233
int dev_addr_add(struct net_device *dev, const unsigned char *addr,
		 unsigned char addr_type);
int dev_addr_del(struct net_device *dev, const unsigned char *addr,
		 unsigned char addr_type);
void dev_addr_flush(struct net_device *dev);
int dev_addr_init(struct net_device *dev);
4234 4235

/* Functions used for unicast addresses handling */
4236 4237 4238 4239 4240 4241 4242 4243
int dev_uc_add(struct net_device *dev, const unsigned char *addr);
int dev_uc_add_excl(struct net_device *dev, const unsigned char *addr);
int dev_uc_del(struct net_device *dev, const unsigned char *addr);
int dev_uc_sync(struct net_device *to, struct net_device *from);
int dev_uc_sync_multiple(struct net_device *to, struct net_device *from);
void dev_uc_unsync(struct net_device *to, struct net_device *from);
void dev_uc_flush(struct net_device *dev);
void dev_uc_init(struct net_device *dev);
4244

4245 4246 4247 4248 4249 4250 4251 4252
/**
 *  __dev_uc_sync - Synchonize device's unicast list
 *  @dev:  device to sync
 *  @sync: function to call if address should be added
 *  @unsync: function to call if address should be removed
 *
 *  Add newly added addresses to the interface, and release
 *  addresses that have been deleted.
B
Bjorn Helgaas 已提交
4253
 */
4254 4255 4256 4257 4258 4259 4260 4261 4262 4263
static inline int __dev_uc_sync(struct net_device *dev,
				int (*sync)(struct net_device *,
					    const unsigned char *),
				int (*unsync)(struct net_device *,
					      const unsigned char *))
{
	return __hw_addr_sync_dev(&dev->uc, dev, sync, unsync);
}

/**
4264
 *  __dev_uc_unsync - Remove synchronized addresses from device
4265 4266 4267 4268
 *  @dev:  device to sync
 *  @unsync: function to call if address should be removed
 *
 *  Remove all addresses that were added to the device by dev_uc_sync().
B
Bjorn Helgaas 已提交
4269
 */
4270 4271 4272 4273 4274 4275 4276
static inline void __dev_uc_unsync(struct net_device *dev,
				   int (*unsync)(struct net_device *,
						 const unsigned char *))
{
	__hw_addr_unsync_dev(&dev->uc, dev, unsync);
}

4277
/* Functions used for multicast addresses handling */
4278 4279 4280 4281 4282 4283 4284 4285 4286 4287
int dev_mc_add(struct net_device *dev, const unsigned char *addr);
int dev_mc_add_global(struct net_device *dev, const unsigned char *addr);
int dev_mc_add_excl(struct net_device *dev, const unsigned char *addr);
int dev_mc_del(struct net_device *dev, const unsigned char *addr);
int dev_mc_del_global(struct net_device *dev, const unsigned char *addr);
int dev_mc_sync(struct net_device *to, struct net_device *from);
int dev_mc_sync_multiple(struct net_device *to, struct net_device *from);
void dev_mc_unsync(struct net_device *to, struct net_device *from);
void dev_mc_flush(struct net_device *dev);
void dev_mc_init(struct net_device *dev);
4288

4289 4290 4291 4292 4293 4294 4295 4296
/**
 *  __dev_mc_sync - Synchonize device's multicast list
 *  @dev:  device to sync
 *  @sync: function to call if address should be added
 *  @unsync: function to call if address should be removed
 *
 *  Add newly added addresses to the interface, and release
 *  addresses that have been deleted.
B
Bjorn Helgaas 已提交
4297
 */
4298 4299 4300 4301 4302 4303 4304 4305 4306 4307
static inline int __dev_mc_sync(struct net_device *dev,
				int (*sync)(struct net_device *,
					    const unsigned char *),
				int (*unsync)(struct net_device *,
					      const unsigned char *))
{
	return __hw_addr_sync_dev(&dev->mc, dev, sync, unsync);
}

/**
4308
 *  __dev_mc_unsync - Remove synchronized addresses from device
4309 4310 4311 4312
 *  @dev:  device to sync
 *  @unsync: function to call if address should be removed
 *
 *  Remove all addresses that were added to the device by dev_mc_sync().
B
Bjorn Helgaas 已提交
4313
 */
4314 4315 4316 4317 4318 4319 4320
static inline void __dev_mc_unsync(struct net_device *dev,
				   int (*unsync)(struct net_device *,
						 const unsigned char *))
{
	__hw_addr_unsync_dev(&dev->mc, dev, unsync);
}

4321
/* Functions used for secondary unicast and multicast support */
4322 4323 4324 4325 4326 4327 4328
void dev_set_rx_mode(struct net_device *dev);
void __dev_set_rx_mode(struct net_device *dev);
int dev_set_promiscuity(struct net_device *dev, int inc);
int dev_set_allmulti(struct net_device *dev, int inc);
void netdev_state_change(struct net_device *dev);
void netdev_notify_peers(struct net_device *dev);
void netdev_features_change(struct net_device *dev);
L
Linus Torvalds 已提交
4329
/* Load a device via the kmod */
4330 4331 4332 4333 4334
void dev_load(struct net *net, const char *name);
struct rtnl_link_stats64 *dev_get_stats(struct net_device *dev,
					struct rtnl_link_stats64 *storage);
void netdev_stats_to_stats64(struct rtnl_link_stats64 *stats64,
			     const struct net_device_stats *netdev_stats);
4335

L
Linus Torvalds 已提交
4336
extern int		netdev_max_backlog;
E
Eric Dumazet 已提交
4337
extern int		netdev_tstamp_prequeue;
L
Linus Torvalds 已提交
4338
extern int		weight_p;
4339 4340 4341 4342
extern int		dev_weight_rx_bias;
extern int		dev_weight_tx_bias;
extern int		dev_rx_weight;
extern int		dev_tx_weight;
4343
extern int		gro_normal_batch;
J
Jiri Pirko 已提交
4344

4345
bool netdev_has_upper_dev(struct net_device *dev, struct net_device *upper_dev);
4346 4347
struct net_device *netdev_upper_get_next_dev_rcu(struct net_device *dev,
						     struct list_head **iter);
4348 4349
struct net_device *netdev_all_upper_get_next_dev_rcu(struct net_device *dev,
						     struct list_head **iter);
4350

4351 4352 4353 4354 4355 4356 4357
/* iterate through upper list, must be called under RCU read lock */
#define netdev_for_each_upper_dev_rcu(dev, updev, iter) \
	for (iter = &(dev)->adj_list.upper, \
	     updev = netdev_upper_get_next_dev_rcu(dev, &(iter)); \
	     updev; \
	     updev = netdev_upper_get_next_dev_rcu(dev, &(iter)))

4358 4359 4360 4361 4362 4363 4364 4365
int netdev_walk_all_upper_dev_rcu(struct net_device *dev,
				  int (*fn)(struct net_device *upper_dev,
					    void *data),
				  void *data);

bool netdev_has_upper_dev_all_rcu(struct net_device *dev,
				  struct net_device *upper_dev);

4366 4367
bool netdev_has_any_upper_dev(struct net_device *dev);

4368 4369 4370 4371
void *netdev_lower_get_next_private(struct net_device *dev,
				    struct list_head **iter);
void *netdev_lower_get_next_private_rcu(struct net_device *dev,
					struct list_head **iter);
4372 4373 4374 4375 4376 4377 4378 4379 4380 4381 4382 4383 4384

#define netdev_for_each_lower_private(dev, priv, iter) \
	for (iter = (dev)->adj_list.lower.next, \
	     priv = netdev_lower_get_next_private(dev, &(iter)); \
	     priv; \
	     priv = netdev_lower_get_next_private(dev, &(iter)))

#define netdev_for_each_lower_private_rcu(dev, priv, iter) \
	for (iter = &(dev)->adj_list.lower, \
	     priv = netdev_lower_get_next_private_rcu(dev, &(iter)); \
	     priv; \
	     priv = netdev_lower_get_next_private_rcu(dev, &(iter)))

4385 4386
void *netdev_lower_get_next(struct net_device *dev,
				struct list_head **iter);
4387

4388
#define netdev_for_each_lower_dev(dev, ldev, iter) \
4389
	for (iter = (dev)->adj_list.lower.next, \
4390 4391 4392 4393
	     ldev = netdev_lower_get_next(dev, &(iter)); \
	     ldev; \
	     ldev = netdev_lower_get_next(dev, &(iter)))

4394
struct net_device *netdev_next_lower_dev_rcu(struct net_device *dev,
4395
					     struct list_head **iter);
4396 4397 4398 4399 4400 4401 4402 4403 4404
int netdev_walk_all_lower_dev(struct net_device *dev,
			      int (*fn)(struct net_device *lower_dev,
					void *data),
			      void *data);
int netdev_walk_all_lower_dev_rcu(struct net_device *dev,
				  int (*fn)(struct net_device *lower_dev,
					    void *data),
				  void *data);

4405
void *netdev_adjacent_get_private(struct list_head *adj_list);
4406
void *netdev_lower_get_first_private_rcu(struct net_device *dev);
4407 4408
struct net_device *netdev_master_upper_dev_get(struct net_device *dev);
struct net_device *netdev_master_upper_dev_get_rcu(struct net_device *dev);
4409 4410
int netdev_upper_dev_link(struct net_device *dev, struct net_device *upper_dev,
			  struct netlink_ext_ack *extack);
4411
int netdev_master_upper_dev_link(struct net_device *dev,
4412
				 struct net_device *upper_dev,
4413 4414
				 void *upper_priv, void *upper_info,
				 struct netlink_ext_ack *extack);
4415 4416
void netdev_upper_dev_unlink(struct net_device *dev,
			     struct net_device *upper_dev);
4417 4418 4419 4420 4421 4422 4423 4424 4425 4426
int netdev_adjacent_change_prepare(struct net_device *old_dev,
				   struct net_device *new_dev,
				   struct net_device *dev,
				   struct netlink_ext_ack *extack);
void netdev_adjacent_change_commit(struct net_device *old_dev,
				   struct net_device *new_dev,
				   struct net_device *dev);
void netdev_adjacent_change_abort(struct net_device *old_dev,
				  struct net_device *new_dev,
				  struct net_device *dev);
4427
void netdev_adjacent_rename_links(struct net_device *dev, char *oldname);
4428 4429
void *netdev_lower_dev_get_private(struct net_device *dev,
				   struct net_device *lower_dev);
4430 4431
void netdev_lower_state_changed(struct net_device *lower_dev,
				void *lower_state_info);
4432 4433 4434

/* RSS keys are 40 or 52 bytes long */
#define NETDEV_RSS_KEY_LEN 52
4435
extern u8 netdev_rss_key[NETDEV_RSS_KEY_LEN] __read_mostly;
4436 4437
void netdev_rss_key_fill(void *buffer, size_t len);

4438
int skb_checksum_help(struct sk_buff *skb);
4439
int skb_crc32c_csum_help(struct sk_buff *skb);
4440 4441 4442
int skb_csum_hwoffload_help(struct sk_buff *skb,
			    const netdev_features_t features);

4443 4444 4445 4446
struct sk_buff *__skb_gso_segment(struct sk_buff *skb,
				  netdev_features_t features, bool tx_path);
struct sk_buff *skb_mac_gso_segment(struct sk_buff *skb,
				    netdev_features_t features);
4447

4448 4449 4450 4451 4452 4453 4454 4455 4456 4457 4458 4459 4460
struct netdev_bonding_info {
	ifslave	slave;
	ifbond	master;
};

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

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

4461 4462 4463 4464 4465 4466 4467 4468 4469
#if IS_ENABLED(CONFIG_ETHTOOL_NETLINK)
void ethtool_notify(struct net_device *dev, unsigned int cmd, const void *data);
#else
static inline void ethtool_notify(struct net_device *dev, unsigned int cmd,
				  const void *data)
{
}
#endif

4470 4471 4472 4473 4474
static inline
struct sk_buff *skb_gso_segment(struct sk_buff *skb, netdev_features_t features)
{
	return __skb_gso_segment(skb, features, true);
}
4475
__be16 skb_network_protocol(struct sk_buff *skb, int *depth);
4476 4477 4478 4479

static inline bool can_checksum_protocol(netdev_features_t features,
					 __be16 protocol)
{
4480 4481 4482 4483 4484 4485 4486 4487 4488 4489 4490 4491 4492 4493 4494 4495 4496 4497
	if (protocol == htons(ETH_P_FCOE))
		return !!(features & NETIF_F_FCOE_CRC);

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

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

	switch (protocol) {
	case htons(ETH_P_IP):
		return !!(features & NETIF_F_IP_CSUM);
	case htons(ETH_P_IPV6):
		return !!(features & NETIF_F_IPV6_CSUM);
	default:
		return false;
	}
4498
}
4499

4500
#ifdef CONFIG_BUG
4501
void netdev_rx_csum_fault(struct net_device *dev, struct sk_buff *skb);
4502
#else
4503 4504
static inline void netdev_rx_csum_fault(struct net_device *dev,
					struct sk_buff *skb)
4505 4506 4507
{
}
#endif
L
Linus Torvalds 已提交
4508
/* rx skb timestamps */
4509 4510
void net_enable_timestamp(void);
void net_disable_timestamp(void);
L
Linus Torvalds 已提交
4511

4512
#ifdef CONFIG_PROC_FS
4513
int __init dev_proc_init(void);
4514 4515
#else
#define dev_proc_init() 0
4516 4517
#endif

D
David S. Miller 已提交
4518
static inline netdev_tx_t __netdev_start_xmit(const struct net_device_ops *ops,
4519 4520
					      struct sk_buff *skb, struct net_device *dev,
					      bool more)
D
David S. Miller 已提交
4521
{
4522
	__this_cpu_write(softnet_data.xmit.more, more);
4523
	return ops->ndo_start_xmit(skb, dev);
D
David S. Miller 已提交
4524 4525
}

4526 4527 4528 4529 4530
static inline bool netdev_xmit_more(void)
{
	return __this_cpu_read(softnet_data.xmit.more);
}

4531
static inline netdev_tx_t netdev_start_xmit(struct sk_buff *skb, struct net_device *dev,
4532
					    struct netdev_queue *txq, bool more)
D
David S. Miller 已提交
4533 4534
{
	const struct net_device_ops *ops = dev->netdev_ops;
4535
	netdev_tx_t rc;
D
David S. Miller 已提交
4536

4537
	rc = __netdev_start_xmit(ops, skb, dev, more);
4538 4539 4540 4541
	if (rc == NETDEV_TX_OK)
		txq_trans_update(txq);

	return rc;
D
David S. Miller 已提交
4542 4543
}

4544
int netdev_class_create_file_ns(const struct class_attribute *class_attr,
4545
				const void *ns);
4546
void netdev_class_remove_file_ns(const struct class_attribute *class_attr,
4547
				 const void *ns);
4548

4549
static inline int netdev_class_create_file(const struct class_attribute *class_attr)
4550 4551 4552 4553
{
	return netdev_class_create_file_ns(class_attr, NULL);
}

4554
static inline void netdev_class_remove_file(const struct class_attribute *class_attr)
4555 4556 4557
{
	netdev_class_remove_file_ns(class_attr, NULL);
}
4558

4559
extern const struct kobj_ns_type_operations net_ns_type_operations;
4560

4561
const char *netdev_drivername(const struct net_device *dev);
4562

4563
void linkwatch_run_queue(void);
4564

4565 4566 4567
static inline netdev_features_t netdev_intersect_features(netdev_features_t f1,
							  netdev_features_t f2)
{
4568 4569
	if ((f1 ^ f2) & NETIF_F_HW_CSUM) {
		if (f1 & NETIF_F_HW_CSUM)
4570
			f1 |= (NETIF_F_IP_CSUM|NETIF_F_IPV6_CSUM);
4571
		else
4572
			f2 |= (NETIF_F_IP_CSUM|NETIF_F_IPV6_CSUM);
4573
	}
4574

4575
	return f1 & f2;
4576 4577
}

4578 4579
static inline netdev_features_t netdev_get_wanted_features(
	struct net_device *dev)
4580 4581 4582
{
	return (dev->features & ~dev->hw_features) | dev->wanted_features;
}
4583 4584
netdev_features_t netdev_increment_features(netdev_features_t all,
	netdev_features_t one, netdev_features_t mask);
4585 4586 4587 4588 4589 4590 4591 4592 4593 4594 4595

/* Allow TSO being used on stacked device :
 * Performing the GSO segmentation before last device
 * is a performance improvement.
 */
static inline netdev_features_t netdev_add_tso_features(netdev_features_t features,
							netdev_features_t mask)
{
	return netdev_increment_features(features, NETIF_F_ALL_TSO, mask);
}

4596
int __netdev_update_features(struct net_device *dev);
4597
void netdev_update_features(struct net_device *dev);
4598
void netdev_change_features(struct net_device *dev);
4599

4600 4601 4602
void netif_stacked_transfer_operstate(const struct net_device *rootdev,
					struct net_device *dev);

4603 4604 4605
netdev_features_t passthru_features_check(struct sk_buff *skb,
					  struct net_device *dev,
					  netdev_features_t features);
4606
netdev_features_t netif_skb_features(struct sk_buff *skb);
4607

4608
static inline bool net_gso_ok(netdev_features_t features, int gso_type)
4609
{
4610
	netdev_features_t feature = (netdev_features_t)gso_type << NETIF_F_GSO_SHIFT;
4611 4612 4613 4614 4615

	/* check flags correspondence */
	BUILD_BUG_ON(SKB_GSO_TCPV4   != (NETIF_F_TSO >> 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));
4616
	BUILD_BUG_ON(SKB_GSO_TCP_FIXEDID != (NETIF_F_TSO_MANGLEID >> NETIF_F_GSO_SHIFT));
4617 4618
	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));
4619 4620
	BUILD_BUG_ON(SKB_GSO_GRE     != (NETIF_F_GSO_GRE >> NETIF_F_GSO_SHIFT));
	BUILD_BUG_ON(SKB_GSO_GRE_CSUM != (NETIF_F_GSO_GRE_CSUM >> NETIF_F_GSO_SHIFT));
4621 4622
	BUILD_BUG_ON(SKB_GSO_IPXIP4  != (NETIF_F_GSO_IPXIP4 >> NETIF_F_GSO_SHIFT));
	BUILD_BUG_ON(SKB_GSO_IPXIP6  != (NETIF_F_GSO_IPXIP6 >> NETIF_F_GSO_SHIFT));
4623 4624
	BUILD_BUG_ON(SKB_GSO_UDP_TUNNEL != (NETIF_F_GSO_UDP_TUNNEL >> NETIF_F_GSO_SHIFT));
	BUILD_BUG_ON(SKB_GSO_UDP_TUNNEL_CSUM != (NETIF_F_GSO_UDP_TUNNEL_CSUM >> NETIF_F_GSO_SHIFT));
4625
	BUILD_BUG_ON(SKB_GSO_PARTIAL != (NETIF_F_GSO_PARTIAL >> NETIF_F_GSO_SHIFT));
4626
	BUILD_BUG_ON(SKB_GSO_TUNNEL_REMCSUM != (NETIF_F_GSO_TUNNEL_REMCSUM >> NETIF_F_GSO_SHIFT));
M
Marcelo Ricardo Leitner 已提交
4627
	BUILD_BUG_ON(SKB_GSO_SCTP    != (NETIF_F_GSO_SCTP >> NETIF_F_GSO_SHIFT));
S
Steffen Klassert 已提交
4628
	BUILD_BUG_ON(SKB_GSO_ESP != (NETIF_F_GSO_ESP >> NETIF_F_GSO_SHIFT));
4629
	BUILD_BUG_ON(SKB_GSO_UDP != (NETIF_F_GSO_UDP >> NETIF_F_GSO_SHIFT));
4630
	BUILD_BUG_ON(SKB_GSO_UDP_L4 != (NETIF_F_GSO_UDP_L4 >> NETIF_F_GSO_SHIFT));
4631
	BUILD_BUG_ON(SKB_GSO_FRAGLIST != (NETIF_F_GSO_FRAGLIST >> NETIF_F_GSO_SHIFT));
4632

4633
	return (features & feature) == feature;
4634 4635
}

4636
static inline bool skb_gso_ok(struct sk_buff *skb, netdev_features_t features)
4637
{
H
Herbert Xu 已提交
4638
	return net_gso_ok(features, skb_shinfo(skb)->gso_type) &&
4639
	       (!skb_has_frag_list(skb) || (features & NETIF_F_FRAGLIST));
4640 4641
}

4642
static inline bool netif_needs_gso(struct sk_buff *skb,
4643
				   netdev_features_t features)
4644
{
4645
	return skb_is_gso(skb) && (!skb_gso_ok(skb, features) ||
4646 4647
		unlikely((skb->ip_summed != CHECKSUM_PARTIAL) &&
			 (skb->ip_summed != CHECKSUM_UNNECESSARY)));
4648 4649
}

4650 4651 4652 4653 4654 4655
static inline void netif_set_gso_max_size(struct net_device *dev,
					  unsigned int size)
{
	dev->gso_max_size = size;
}

4656 4657 4658 4659 4660 4661 4662 4663 4664 4665 4666 4667 4668
static inline void skb_gso_error_unwind(struct sk_buff *skb, __be16 protocol,
					int pulled_hlen, u16 mac_offset,
					int mac_len)
{
	skb->protocol = protocol;
	skb->encapsulation = 1;
	skb_push(skb, pulled_hlen);
	skb_reset_transport_header(skb);
	skb->mac_header = mac_offset;
	skb->network_header = skb->mac_header + mac_len;
	skb->mac_len = mac_len;
}

4669 4670 4671 4672 4673
static inline bool netif_is_macsec(const struct net_device *dev)
{
	return dev->priv_flags & IFF_MACSEC;
}

4674
static inline bool netif_is_macvlan(const struct net_device *dev)
4675 4676 4677 4678
{
	return dev->priv_flags & IFF_MACVLAN;
}

4679
static inline bool netif_is_macvlan_port(const struct net_device *dev)
4680 4681 4682 4683
{
	return dev->priv_flags & IFF_MACVLAN_PORT;
}

4684
static inline bool netif_is_bond_master(const struct net_device *dev)
4685 4686 4687 4688
{
	return dev->flags & IFF_MASTER && dev->priv_flags & IFF_BONDING;
}

4689
static inline bool netif_is_bond_slave(const struct net_device *dev)
J
Jiri Pirko 已提交
4690 4691 4692 4693
{
	return dev->flags & IFF_SLAVE && dev->priv_flags & IFF_BONDING;
}

4694 4695 4696 4697 4698
static inline bool netif_supports_nofcs(struct net_device *dev)
{
	return dev->priv_flags & IFF_SUPP_NOFCS;
}

4699 4700 4701 4702 4703
static inline bool netif_has_l3_rx_handler(const struct net_device *dev)
{
	return dev->priv_flags & IFF_L3MDEV_RX_HANDLER;
}

4704
static inline bool netif_is_l3_master(const struct net_device *dev)
4705
{
4706
	return dev->priv_flags & IFF_L3MDEV_MASTER;
4707 4708
}

D
David Ahern 已提交
4709 4710 4711 4712 4713
static inline bool netif_is_l3_slave(const struct net_device *dev)
{
	return dev->priv_flags & IFF_L3MDEV_SLAVE;
}

4714 4715 4716 4717 4718
static inline bool netif_is_bridge_master(const struct net_device *dev)
{
	return dev->priv_flags & IFF_EBRIDGE;
}

4719 4720 4721 4722 4723
static inline bool netif_is_bridge_port(const struct net_device *dev)
{
	return dev->priv_flags & IFF_BRIDGE_PORT;
}

4724 4725 4726 4727 4728
static inline bool netif_is_ovs_master(const struct net_device *dev)
{
	return dev->priv_flags & IFF_OPENVSWITCH;
}

J
Jiri Pirko 已提交
4729 4730 4731 4732 4733
static inline bool netif_is_ovs_port(const struct net_device *dev)
{
	return dev->priv_flags & IFF_OVS_DATAPATH;
}

4734
static inline bool netif_is_team_master(const struct net_device *dev)
J
Jiri Pirko 已提交
4735 4736 4737 4738
{
	return dev->priv_flags & IFF_TEAM;
}

4739
static inline bool netif_is_team_port(const struct net_device *dev)
J
Jiri Pirko 已提交
4740 4741 4742 4743
{
	return dev->priv_flags & IFF_TEAM_PORT;
}

4744
static inline bool netif_is_lag_master(const struct net_device *dev)
J
Jiri Pirko 已提交
4745 4746 4747 4748
{
	return netif_is_bond_master(dev) || netif_is_team_master(dev);
}

4749
static inline bool netif_is_lag_port(const struct net_device *dev)
J
Jiri Pirko 已提交
4750 4751 4752 4753
{
	return netif_is_bond_slave(dev) || netif_is_team_port(dev);
}

4754 4755 4756 4757 4758
static inline bool netif_is_rxfh_configured(const struct net_device *dev)
{
	return dev->priv_flags & IFF_RXFH_CONFIGURED;
}

4759 4760 4761 4762 4763 4764 4765 4766 4767 4768
static inline bool netif_is_failover(const struct net_device *dev)
{
	return dev->priv_flags & IFF_FAILOVER;
}

static inline bool netif_is_failover_slave(const struct net_device *dev)
{
	return dev->priv_flags & IFF_FAILOVER_SLAVE;
}

4769 4770 4771 4772 4773 4774
/* This device needs to keep skb dst for qdisc enqueue or ndo_start_xmit() */
static inline void netif_keep_dst(struct net_device *dev)
{
	dev->priv_flags &= ~(IFF_XMIT_DST_RELEASE | IFF_XMIT_DST_RELEASE_PERM);
}

4775 4776 4777 4778 4779 4780 4781
/* return true if dev can't cope with mtu frames that need vlan tag insertion */
static inline bool netif_reduces_vlan_mtu(struct net_device *dev)
{
	/* TODO: reserve and use an additional IFF bit, if we get more users */
	return dev->priv_flags & IFF_MACSEC;
}

4782
extern struct pernet_operations __net_initdata loopback_net_ops;
4783

4784 4785 4786 4787 4788 4789
/* Logging, debugging and troubleshooting/diagnostic helpers. */

/* netdev_printk helpers, similar to dev_printk */

static inline const char *netdev_name(const struct net_device *dev)
{
4790 4791
	if (!dev->name[0] || strchr(dev->name, '%'))
		return "(unnamed net_device)";
4792 4793 4794
	return dev->name;
}

4795 4796 4797 4798 4799
static inline bool netdev_unregistering(const struct net_device *dev)
{
	return dev->reg_state == NETREG_UNREGISTERING;
}

4800 4801 4802 4803 4804 4805 4806 4807 4808 4809 4810 4811 4812 4813 4814
static inline const char *netdev_reg_state(const struct net_device *dev)
{
	switch (dev->reg_state) {
	case NETREG_UNINITIALIZED: return " (uninitialized)";
	case NETREG_REGISTERED: return "";
	case NETREG_UNREGISTERING: return " (unregistering)";
	case NETREG_UNREGISTERED: return " (unregistered)";
	case NETREG_RELEASED: return " (released)";
	case NETREG_DUMMY: return " (dummy)";
	}

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

4815
__printf(3, 4) __cold
4816 4817
void netdev_printk(const char *level, const struct net_device *dev,
		   const char *format, ...);
4818
__printf(2, 3) __cold
4819
void netdev_emerg(const struct net_device *dev, const char *format, ...);
4820
__printf(2, 3) __cold
4821
void netdev_alert(const struct net_device *dev, const char *format, ...);
4822
__printf(2, 3) __cold
4823
void netdev_crit(const struct net_device *dev, const char *format, ...);
4824
__printf(2, 3) __cold
4825
void netdev_err(const struct net_device *dev, const char *format, ...);
4826
__printf(2, 3) __cold
4827
void netdev_warn(const struct net_device *dev, const char *format, ...);
4828
__printf(2, 3) __cold
4829
void netdev_notice(const struct net_device *dev, const char *format, ...);
4830
__printf(2, 3) __cold
4831
void netdev_info(const struct net_device *dev, const char *format, ...);
4832

4833 4834 4835 4836 4837 4838 4839 4840 4841 4842 4843 4844 4845 4846 4847 4848 4849 4850 4851 4852 4853 4854 4855 4856 4857
#define netdev_level_once(level, dev, fmt, ...)			\
do {								\
	static bool __print_once __read_mostly;			\
								\
	if (!__print_once) {					\
		__print_once = true;				\
		netdev_printk(level, dev, fmt, ##__VA_ARGS__);	\
	}							\
} while (0)

#define netdev_emerg_once(dev, fmt, ...) \
	netdev_level_once(KERN_EMERG, dev, fmt, ##__VA_ARGS__)
#define netdev_alert_once(dev, fmt, ...) \
	netdev_level_once(KERN_ALERT, dev, fmt, ##__VA_ARGS__)
#define netdev_crit_once(dev, fmt, ...) \
	netdev_level_once(KERN_CRIT, dev, fmt, ##__VA_ARGS__)
#define netdev_err_once(dev, fmt, ...) \
	netdev_level_once(KERN_ERR, dev, fmt, ##__VA_ARGS__)
#define netdev_warn_once(dev, fmt, ...) \
	netdev_level_once(KERN_WARNING, dev, fmt, ##__VA_ARGS__)
#define netdev_notice_once(dev, fmt, ...) \
	netdev_level_once(KERN_NOTICE, dev, fmt, ##__VA_ARGS__)
#define netdev_info_once(dev, fmt, ...) \
	netdev_level_once(KERN_INFO, dev, fmt, ##__VA_ARGS__)

4858 4859 4860
#define MODULE_ALIAS_NETDEV(device) \
	MODULE_ALIAS("netdev-" device)

4861
#if defined(CONFIG_DYNAMIC_DEBUG)
4862 4863
#define netdev_dbg(__dev, format, args...)			\
do {								\
4864
	dynamic_netdev_dbg(__dev, format, ##args);		\
4865
} while (0)
4866 4867 4868
#elif defined(DEBUG)
#define netdev_dbg(__dev, format, args...)			\
	netdev_printk(KERN_DEBUG, __dev, format, ##args)
4869 4870 4871 4872 4873 4874 4875 4876 4877 4878 4879 4880 4881 4882 4883 4884 4885 4886 4887 4888 4889 4890 4891 4892 4893 4894
#else
#define netdev_dbg(__dev, format, args...)			\
({								\
	if (0)							\
		netdev_printk(KERN_DEBUG, __dev, format, ##args); \
})
#endif

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

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

/*
 * netdev_WARN() acts like dev_printk(), but with the key difference
 * of using a WARN/WARN_ON to get the message out, including the
 * file/line information and a backtrace.
 */
#define netdev_WARN(dev, format, args...)			\
4895
	WARN(1, "netdevice: %s%s: " format, netdev_name(dev),	\
4896
	     netdev_reg_state(dev), ##args)
4897

G
Gal Pressman 已提交
4898
#define netdev_WARN_ONCE(dev, format, args...)				\
4899
	WARN_ONCE(1, "netdevice: %s%s: " format, netdev_name(dev),	\
4900 4901
		  netdev_reg_state(dev), ##args)

4902 4903 4904 4905 4906 4907 4908 4909
/* 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)

4910 4911 4912 4913 4914 4915
#define netif_level(level, priv, type, dev, fmt, args...)	\
do {								\
	if (netif_msg_##type(priv))				\
		netdev_##level(dev, fmt, ##args);		\
} while (0)

4916
#define netif_emerg(priv, type, dev, fmt, args...)		\
4917
	netif_level(emerg, priv, type, dev, fmt, ##args)
4918
#define netif_alert(priv, type, dev, fmt, args...)		\
4919
	netif_level(alert, priv, type, dev, fmt, ##args)
4920
#define netif_crit(priv, type, dev, fmt, args...)		\
4921
	netif_level(crit, priv, type, dev, fmt, ##args)
4922
#define netif_err(priv, type, dev, fmt, args...)		\
4923
	netif_level(err, priv, type, dev, fmt, ##args)
4924
#define netif_warn(priv, type, dev, fmt, args...)		\
4925
	netif_level(warn, priv, type, dev, fmt, ##args)
4926
#define netif_notice(priv, type, dev, fmt, args...)		\
4927
	netif_level(notice, priv, type, dev, fmt, ##args)
4928
#define netif_info(priv, type, dev, fmt, args...)		\
4929
	netif_level(info, priv, type, dev, fmt, ##args)
4930

4931
#if defined(CONFIG_DYNAMIC_DEBUG)
4932 4933 4934
#define netif_dbg(priv, type, netdev, format, args...)		\
do {								\
	if (netif_msg_##type(priv))				\
4935
		dynamic_netdev_dbg(netdev, format, ##args);	\
4936
} while (0)
4937 4938 4939
#elif defined(DEBUG)
#define netif_dbg(priv, type, dev, format, args...)		\
	netif_printk(priv, type, KERN_DEBUG, dev, format, ##args)
4940 4941 4942 4943 4944 4945 4946 4947 4948
#else
#define netif_dbg(priv, type, dev, format, args...)			\
({									\
	if (0)								\
		netif_printk(priv, type, KERN_DEBUG, dev, format, ##args); \
	0;								\
})
#endif

4949 4950 4951 4952 4953 4954 4955 4956 4957
/* if @cond then downgrade to debug, else print at @level */
#define netif_cond_dbg(priv, type, netdev, cond, level, fmt, args...)     \
	do {                                                              \
		if (cond)                                                 \
			netif_dbg(priv, type, netdev, fmt, ##args);       \
		else                                                      \
			netif_ ## level(priv, type, netdev, fmt, ##args); \
	} while (0)

4958
#if defined(VERBOSE_DEBUG)
4959
#define netif_vdbg	netif_dbg
4960 4961 4962 4963
#else
#define netif_vdbg(priv, type, dev, format, args...)		\
({								\
	if (0)							\
4964
		netif_printk(priv, type, KERN_DEBUG, dev, format, ##args); \
4965 4966 4967
	0;							\
})
#endif
4968

4969 4970 4971 4972 4973 4974 4975 4976 4977 4978 4979 4980 4981 4982 4983 4984 4985 4986 4987 4988 4989 4990
/*
 *	The list of packet types we will receive (as opposed to discard)
 *	and the routines to invoke.
 *
 *	Why 16. Because with 16 the only overlap we get on a hash of the
 *	low nibble of the protocol value is RARP/SNAP/X.25.
 *
 *		0800	IP
 *		0001	802.3
 *		0002	AX.25
 *		0004	802.2
 *		8035	RARP
 *		0005	SNAP
 *		0805	X.25
 *		0806	ARP
 *		8137	IPX
 *		0009	Localtalk
 *		86DD	IPv6
 */
#define PTYPE_HASH_SIZE	(16)
#define PTYPE_HASH_MASK	(PTYPE_HASH_SIZE - 1)

4991 4992
extern struct net_device *blackhole_netdev;

4993
#endif	/* _LINUX_NETDEVICE_H */