skbuff.h 62.3 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
/*
 *	Definitions for the 'struct sk_buff' memory handlers.
 *
 *	Authors:
 *		Alan Cox, <gw4pts@gw4pts.ampr.org>
 *		Florian La Roche, <rzsfl@rz.uni-sb.de>
 *
 *	This program is free software; you can redistribute it and/or
 *	modify it under the terms of the GNU General Public License
 *	as published by the Free Software Foundation; either version
 *	2 of the License, or (at your option) any later version.
 */

#ifndef _LINUX_SKBUFF_H
#define _LINUX_SKBUFF_H

#include <linux/kernel.h>
18
#include <linux/kmemcheck.h>
L
Linus Torvalds 已提交
19 20 21 22 23 24 25 26
#include <linux/compiler.h>
#include <linux/time.h>
#include <linux/cache.h>

#include <asm/atomic.h>
#include <asm/types.h>
#include <linux/spinlock.h>
#include <linux/net.h>
27
#include <linux/textsearch.h>
L
Linus Torvalds 已提交
28
#include <net/checksum.h>
29
#include <linux/rcupdate.h>
30
#include <linux/dmaengine.h>
31
#include <linux/hrtimer.h>
L
Linus Torvalds 已提交
32

33
/* Don't change this without changing skb_csum_unnecessary! */
L
Linus Torvalds 已提交
34
#define CHECKSUM_NONE 0
35 36 37
#define CHECKSUM_UNNECESSARY 1
#define CHECKSUM_COMPLETE 2
#define CHECKSUM_PARTIAL 3
L
Linus Torvalds 已提交
38 39 40

#define SKB_DATA_ALIGN(X)	(((X) + (SMP_CACHE_BYTES - 1)) & \
				 ~(SMP_CACHE_BYTES - 1))
41
#define SKB_WITH_OVERHEAD(X)	\
42
	((X) - SKB_DATA_ALIGN(sizeof(struct skb_shared_info)))
43 44
#define SKB_MAX_ORDER(X, ORDER) \
	SKB_WITH_OVERHEAD((PAGE_SIZE << (ORDER)) - (X))
L
Linus Torvalds 已提交
45 46 47 48 49 50 51 52 53 54 55 56 57 58
#define SKB_MAX_HEAD(X)		(SKB_MAX_ORDER((X), 0))
#define SKB_MAX_ALLOC		(SKB_MAX_ORDER(0, 2))

/* A. Checksumming of received packets by device.
 *
 *	NONE: device failed to checksum this packet.
 *		skb->csum is undefined.
 *
 *	UNNECESSARY: device parsed packet and wouldbe verified checksum.
 *		skb->csum is undefined.
 *	      It is bad option, but, unfortunately, many of vendors do this.
 *	      Apparently with secret goal to sell you new device, when you
 *	      will add new protocol to your host. F.e. IPv6. 8)
 *
59
 *	COMPLETE: the most generic way. Device supplied checksum of _all_
L
Linus Torvalds 已提交
60 61
 *	    the packet as seen by netif_rx in skb->csum.
 *	    NOTE: Even if device supports only some protocols, but
62
 *	    is able to produce some skb->csum, it MUST use COMPLETE,
L
Linus Torvalds 已提交
63 64
 *	    not UNNECESSARY.
 *
65 66 67 68 69 70 71
 *	PARTIAL: identical to the case for output below.  This may occur
 *	    on a packet received directly from another Linux OS, e.g.,
 *	    a virtualised Linux kernel on the same host.  The packet can
 *	    be treated in the same way as UNNECESSARY except that on
 *	    output (i.e., forwarding) the checksum must be filled in
 *	    by the OS or the hardware.
 *
L
Linus Torvalds 已提交
72 73 74 75
 * B. Checksumming on output.
 *
 *	NONE: skb is checksummed by protocol or csum is not required.
 *
76
 *	PARTIAL: device is required to csum packet as seen by hard_start_xmit
77 78
 *	from skb->csum_start to the end and to record the checksum
 *	at skb->csum_start + skb->csum_offset.
L
Linus Torvalds 已提交
79 80 81 82 83 84 85 86 87 88
 *
 *	Device must show its capabilities in dev->features, set
 *	at device setup time.
 *	NETIF_F_HW_CSUM	- it is clever device, it is able to checksum
 *			  everything.
 *	NETIF_F_NO_CSUM - loopback or reliable single hop media.
 *	NETIF_F_IP_CSUM - device is dumb. It is able to csum only
 *			  TCP/UDP over IPv4. Sigh. Vendors like this
 *			  way by an unknown reason. Though, see comment above
 *			  about CHECKSUM_UNNECESSARY. 8)
89
 *	NETIF_F_IPV6_CSUM about as dumb as the last one but does IPv6 instead.
L
Linus Torvalds 已提交
90 91 92 93 94
 *
 *	Any questions? No questions, good. 		--ANK
 */

struct net_device;
95
struct scatterlist;
J
Jens Axboe 已提交
96
struct pipe_inode_info;
L
Linus Torvalds 已提交
97

98
#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE)
L
Linus Torvalds 已提交
99 100 101
struct nf_conntrack {
	atomic_t use;
};
102
#endif
L
Linus Torvalds 已提交
103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124

#ifdef CONFIG_BRIDGE_NETFILTER
struct nf_bridge_info {
	atomic_t use;
	struct net_device *physindev;
	struct net_device *physoutdev;
	unsigned int mask;
	unsigned long data[32 / sizeof(unsigned long)];
};
#endif

struct sk_buff_head {
	/* These two members must be first. */
	struct sk_buff	*next;
	struct sk_buff	*prev;

	__u32		qlen;
	spinlock_t	lock;
};

struct sk_buff;

125 126 127 128
/* To allow 64K frame to be packed as single skb without frag_list. Since
 * GRO uses frags we allocate at least 16 regardless of page size.
 */
#if (65536/PAGE_SIZE + 2) < 16
129
#define MAX_SKB_FRAGS 16UL
130
#else
L
Linus Torvalds 已提交
131
#define MAX_SKB_FRAGS (65536/PAGE_SIZE + 2)
132
#endif
L
Linus Torvalds 已提交
133 134 135 136 137

typedef struct skb_frag_struct skb_frag_t;

struct skb_frag_struct {
	struct page *page;
138
#if (BITS_PER_LONG > 32) || (PAGE_SIZE >= 65536)
139 140
	__u32 page_offset;
	__u32 size;
141 142 143 144
#else
	__u16 page_offset;
	__u16 size;
#endif
L
Linus Torvalds 已提交
145 146
};

147 148 149
#define HAVE_HW_TIME_STAMP

/**
150
 * struct skb_shared_hwtstamps - hardware time stamps
151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176
 * @hwtstamp:	hardware time stamp transformed into duration
 *		since arbitrary point in time
 * @syststamp:	hwtstamp transformed to system time base
 *
 * Software time stamps generated by ktime_get_real() are stored in
 * skb->tstamp. The relation between the different kinds of time
 * stamps is as follows:
 *
 * syststamp and tstamp can be compared against each other in
 * arbitrary combinations.  The accuracy of a
 * syststamp/tstamp/"syststamp from other device" comparison is
 * limited by the accuracy of the transformation into system time
 * base. This depends on the device driver and its underlying
 * hardware.
 *
 * hwtstamps can only be compared against other hwtstamps from
 * the same device.
 *
 * This structure is attached to packets as part of the
 * &skb_shared_info. Use skb_hwtstamps() to get a pointer.
 */
struct skb_shared_hwtstamps {
	ktime_t	hwtstamp;
	ktime_t	syststamp;
};

177 178 179 180 181 182 183 184 185 186 187 188 189
/* Definitions for tx_flags in struct skb_shared_info */
enum {
	/* generate hardware time stamp */
	SKBTX_HW_TSTAMP = 1 << 0,

	/* generate software time stamp */
	SKBTX_SW_TSTAMP = 1 << 1,

	/* device driver is going to provide hardware time stamp */
	SKBTX_IN_PROGRESS = 1 << 2,

	/* ensure the originating sk reference is available on driver level */
	SKBTX_DRV_NEEDS_SK_REF = 1 << 3,
190 191
};

L
Linus Torvalds 已提交
192 193 194 195
/* This data is invariant across clones and lives at
 * the end of the header data, ie. at skb->end.
 */
struct skb_shared_info {
B
Benjamin LaHaise 已提交
196
	unsigned short	nr_frags;
197 198 199 200
	unsigned short	gso_size;
	/* Warning: this field is not always filled in (UFO)! */
	unsigned short	gso_segs;
	unsigned short  gso_type;
A
Al Viro 已提交
201
	__be32          ip6_frag_id;
202
	__u8		tx_flags;
L
Linus Torvalds 已提交
203
	struct sk_buff	*frag_list;
204
	struct skb_shared_hwtstamps hwtstamps;
E
Eric Dumazet 已提交
205 206 207 208 209 210

	/*
	 * Warning : all fields before dataref are cleared in __alloc_skb()
	 */
	atomic_t	dataref;

J
Johann Baudy 已提交
211 212 213
	/* Intermediate layers must ensure that destructor_arg
	 * remains valid until skb destructor */
	void *		destructor_arg;
214 215
	/* must be last field, see pskb_expand_head() */
	skb_frag_t	frags[MAX_SKB_FRAGS];
L
Linus Torvalds 已提交
216 217 218 219
};

/* We divide dataref into two halves.  The higher 16 bits hold references
 * to the payload part of skb->data.  The lower 16 bits hold references to
220 221
 * the entire skb->data.  A clone of a headerless skb holds the length of
 * the header in skb->hdr_len.
L
Linus Torvalds 已提交
222 223 224 225 226 227 228 229 230 231
 *
 * All users must obey the rule that the skb->data reference count must be
 * greater than or equal to the payload reference count.
 *
 * Holding a reference to the payload part means that the user does not
 * care about modifications to the header part of skb->data.
 */
#define SKB_DATAREF_SHIFT 16
#define SKB_DATAREF_MASK ((1 << SKB_DATAREF_SHIFT) - 1)

232 233 234 235 236 237 238

enum {
	SKB_FCLONE_UNAVAILABLE,
	SKB_FCLONE_ORIG,
	SKB_FCLONE_CLONE,
};

239 240
enum {
	SKB_GSO_TCPV4 = 1 << 0,
H
Herbert Xu 已提交
241
	SKB_GSO_UDP = 1 << 1,
242 243 244

	/* This indicates the skb is from an untrusted source. */
	SKB_GSO_DODGY = 1 << 2,
M
Michael Chan 已提交
245 246

	/* This indicates the tcp segment has CWR set. */
H
Herbert Xu 已提交
247 248 249
	SKB_GSO_TCP_ECN = 1 << 3,

	SKB_GSO_TCPV6 = 1 << 4,
250 251

	SKB_GSO_FCOE = 1 << 5,
252 253
};

254 255 256 257 258 259 260 261 262 263
#if BITS_PER_LONG > 32
#define NET_SKBUFF_DATA_USES_OFFSET 1
#endif

#ifdef NET_SKBUFF_DATA_USES_OFFSET
typedef unsigned int sk_buff_data_t;
#else
typedef unsigned char *sk_buff_data_t;
#endif

264 265 266 267 268
#if defined(CONFIG_NF_DEFRAG_IPV4) || defined(CONFIG_NF_DEFRAG_IPV4_MODULE) || \
    defined(CONFIG_NF_DEFRAG_IPV6) || defined(CONFIG_NF_DEFRAG_IPV6_MODULE)
#define NET_SKBUFF_NF_DEFRAG_NEEDED 1
#endif

L
Linus Torvalds 已提交
269 270 271 272 273
/** 
 *	struct sk_buff - socket buffer
 *	@next: Next buffer in list
 *	@prev: Previous buffer in list
 *	@sk: Socket we are owned by
H
Herbert Xu 已提交
274
 *	@tstamp: Time we arrived
L
Linus Torvalds 已提交
275
 *	@dev: Device we arrived on/are leaving by
R
Randy Dunlap 已提交
276
 *	@transport_header: Transport layer header
277 278
 *	@network_header: Network layer header
 *	@mac_header: Link layer header
E
Eric Dumazet 已提交
279
 *	@_skb_refdst: destination entry (with norefcount bit)
280
 *	@sp: the security path, used for xfrm
L
Linus Torvalds 已提交
281 282 283 284
 *	@cb: Control buffer. Free for use by every layer. Put private vars here
 *	@len: Length of actual data
 *	@data_len: Data length
 *	@mac_len: Length of link layer header
285
 *	@hdr_len: writable header length of cloned skb
286 287 288
 *	@csum: Checksum (must include start/offset pair)
 *	@csum_start: Offset from skb->head where checksumming should start
 *	@csum_offset: Offset from csum_start where checksum should be stored
289
 *	@local_df: allow local fragmentation
L
Linus Torvalds 已提交
290 291 292
 *	@cloned: Head may be cloned (check refcnt to be sure)
 *	@nohdr: Payload reference only, must not modify header
 *	@pkt_type: Packet class
293
 *	@fclone: skbuff clone status
L
Linus Torvalds 已提交
294 295 296 297 298 299 300 301 302 303
 *	@ip_summed: Driver fed us an IP checksum
 *	@priority: Packet queueing priority
 *	@users: User count - see {datagram,tcp}.c
 *	@protocol: Packet protocol from driver
 *	@truesize: Buffer size 
 *	@head: Head of buffer
 *	@data: Data head pointer
 *	@tail: Tail pointer
 *	@end: End pointer
 *	@destructor: Destruct function
T
Thomas Graf 已提交
304
 *	@mark: Generic packet mark
L
Linus Torvalds 已提交
305
 *	@nfct: Associated connection, if any
306
 *	@ipvs_property: skbuff is owned by ipvs
307 308
 *	@peeked: this packet has been seen already, so stats have been
 *		done for it, don't do them again
309
 *	@nf_trace: netfilter packet trace flag
L
Linus Torvalds 已提交
310
 *	@nfctinfo: Relationship of this skb to the connection
R
Randy Dunlap 已提交
311
 *	@nfct_reasm: netfilter conntrack re-assembly pointer
L
Linus Torvalds 已提交
312
 *	@nf_bridge: Saved data about a bridged frame - see br_netfilter.c
313
 *	@skb_iif: ifindex of device we arrived on
T
Tom Herbert 已提交
314
 *	@rxhash: the packet hash computed on receive
315
 *	@queue_mapping: Queue mapping for multiqueue devices
L
Linus Torvalds 已提交
316 317
 *	@tc_index: Traffic control index
 *	@tc_verd: traffic control verdict
318
 *	@ndisc_nodetype: router type (from link layer)
R
Randy Dunlap 已提交
319 320
 *	@dma_cookie: a cookie to one of several possible DMA operations
 *		done by skb DMA functions
321
 *	@secmark: security marking
322
 *	@vlan_tci: vlan tag control information
L
Linus Torvalds 已提交
323 324 325 326 327 328 329
 */

struct sk_buff {
	/* These two members must be first. */
	struct sk_buff		*next;
	struct sk_buff		*prev;

330
	ktime_t			tstamp;
331 332

	struct sock		*sk;
L
Linus Torvalds 已提交
333 334 335 336 337 338 339 340
	struct net_device	*dev;

	/*
	 * This is the control buffer. It is free to use for every
	 * layer. Please put your private variables there. If you
	 * want to keep them across layers you have to do a skb_clone()
	 * first. This is owned by whoever has the skb queued ATM.
	 */
341
	char			cb[48] __aligned(8);
L
Linus Torvalds 已提交
342

E
Eric Dumazet 已提交
343
	unsigned long		_skb_refdst;
344 345 346
#ifdef CONFIG_XFRM
	struct	sec_path	*sp;
#endif
L
Linus Torvalds 已提交
347
	unsigned int		len,
348 349 350
				data_len;
	__u16			mac_len,
				hdr_len;
A
Al Viro 已提交
351 352
	union {
		__wsum		csum;
353 354 355 356
		struct {
			__u16	csum_start;
			__u16	csum_offset;
		};
A
Al Viro 已提交
357
	};
L
Linus Torvalds 已提交
358
	__u32			priority;
359
	kmemcheck_bitfield_begin(flags1);
360 361 362
	__u8			local_df:1,
				cloned:1,
				ip_summed:2,
363 364
				nohdr:1,
				nfctinfo:3;
365
	__u8			pkt_type:3,
366
				fclone:2,
367
				ipvs_property:1,
368
				peeked:1,
369
				nf_trace:1;
370
	kmemcheck_bitfield_end(flags1);
E
Eric Dumazet 已提交
371
	__be16			protocol;
L
Linus Torvalds 已提交
372 373

	void			(*destructor)(struct sk_buff *skb);
374
#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE)
375
	struct nf_conntrack	*nfct;
376 377
#endif
#ifdef NET_SKBUFF_NF_DEFRAG_NEEDED
378 379
	struct sk_buff		*nfct_reasm;
#endif
L
Linus Torvalds 已提交
380 381 382
#ifdef CONFIG_BRIDGE_NETFILTER
	struct nf_bridge_info	*nf_bridge;
#endif
383

384
	int			skb_iif;
L
Linus Torvalds 已提交
385
#ifdef CONFIG_NET_SCHED
386
	__u16			tc_index;	/* traffic control index */
L
Linus Torvalds 已提交
387
#ifdef CONFIG_NET_CLS_ACT
388
	__u16			tc_verd;	/* traffic control verdict */
L
Linus Torvalds 已提交
389 390
#endif
#endif
391

T
Tom Herbert 已提交
392 393
	__u32			rxhash;

394
	__u16			queue_mapping;
395
	kmemcheck_bitfield_begin(flags2);
396
#ifdef CONFIG_IPV6_NDISC_NODETYPE
397
	__u8			ndisc_nodetype:2;
398
#endif
399
	__u8			ooo_okay:1;
400 401
	kmemcheck_bitfield_end(flags2);

402
	/* 0/13 bit hole */
403

404 405 406
#ifdef CONFIG_NET_DMA
	dma_cookie_t		dma_cookie;
#endif
407 408 409
#ifdef CONFIG_NETWORK_SECMARK
	__u32			secmark;
#endif
410 411 412 413
	union {
		__u32		mark;
		__u32		dropcount;
	};
L
Linus Torvalds 已提交
414

415 416
	__u16			vlan_tci;

417 418 419
	sk_buff_data_t		transport_header;
	sk_buff_data_t		network_header;
	sk_buff_data_t		mac_header;
L
Linus Torvalds 已提交
420
	/* These elements must be at the end, see alloc_skb() for details.  */
421
	sk_buff_data_t		tail;
422
	sk_buff_data_t		end;
L
Linus Torvalds 已提交
423
	unsigned char		*head,
424
				*data;
425 426
	unsigned int		truesize;
	atomic_t		users;
L
Linus Torvalds 已提交
427 428 429 430 431 432 433 434 435 436
};

#ifdef __KERNEL__
/*
 *	Handling routines are only of interest to the kernel
 */
#include <linux/slab.h>

#include <asm/system.h>

E
Eric Dumazet 已提交
437 438 439 440 441 442 443 444 445 446 447 448 449
/*
 * skb might have a dst pointer attached, refcounted or not.
 * _skb_refdst low order bit is set if refcount was _not_ taken
 */
#define SKB_DST_NOREF	1UL
#define SKB_DST_PTRMASK	~(SKB_DST_NOREF)

/**
 * skb_dst - returns skb dst_entry
 * @skb: buffer
 *
 * Returns skb dst_entry, regardless of reference taken or not.
 */
E
Eric Dumazet 已提交
450 451
static inline struct dst_entry *skb_dst(const struct sk_buff *skb)
{
E
Eric Dumazet 已提交
452 453 454 455 456 457 458
	/* If refdst was not refcounted, check we still are in a 
	 * rcu_read_lock section
	 */
	WARN_ON((skb->_skb_refdst & SKB_DST_NOREF) &&
		!rcu_read_lock_held() &&
		!rcu_read_lock_bh_held());
	return (struct dst_entry *)(skb->_skb_refdst & SKB_DST_PTRMASK);
E
Eric Dumazet 已提交
459 460
}

E
Eric Dumazet 已提交
461 462 463 464 465 466 467 468
/**
 * skb_dst_set - sets skb dst
 * @skb: buffer
 * @dst: dst entry
 *
 * Sets skb dst, assuming a reference was taken on dst and should
 * be released by skb_dst_drop()
 */
E
Eric Dumazet 已提交
469 470
static inline void skb_dst_set(struct sk_buff *skb, struct dst_entry *dst)
{
E
Eric Dumazet 已提交
471 472 473
	skb->_skb_refdst = (unsigned long)dst;
}

E
Eric Dumazet 已提交
474
extern void skb_dst_set_noref(struct sk_buff *skb, struct dst_entry *dst);
E
Eric Dumazet 已提交
475 476

/**
L
Lucas De Marchi 已提交
477
 * skb_dst_is_noref - Test if skb dst isn't refcounted
E
Eric Dumazet 已提交
478 479 480 481 482
 * @skb: buffer
 */
static inline bool skb_dst_is_noref(const struct sk_buff *skb)
{
	return (skb->_skb_refdst & SKB_DST_NOREF) && skb_dst(skb);
E
Eric Dumazet 已提交
483 484
}

E
Eric Dumazet 已提交
485 486
static inline struct rtable *skb_rtable(const struct sk_buff *skb)
{
E
Eric Dumazet 已提交
487
	return (struct rtable *)skb_dst(skb);
E
Eric Dumazet 已提交
488 489
}

490
extern void kfree_skb(struct sk_buff *skb);
491
extern void consume_skb(struct sk_buff *skb);
L
Linus Torvalds 已提交
492
extern void	       __kfree_skb(struct sk_buff *skb);
493
extern struct sk_buff *__alloc_skb(unsigned int size,
494
				   gfp_t priority, int fclone, int node);
495
static inline struct sk_buff *alloc_skb(unsigned int size,
A
Al Viro 已提交
496
					gfp_t priority)
497
{
E
Eric Dumazet 已提交
498
	return __alloc_skb(size, priority, 0, NUMA_NO_NODE);
499 500 501
}

static inline struct sk_buff *alloc_skb_fclone(unsigned int size,
A
Al Viro 已提交
502
					       gfp_t priority)
503
{
E
Eric Dumazet 已提交
504
	return __alloc_skb(size, priority, 1, NUMA_NO_NODE);
505 506
}

507
extern bool skb_recycle_check(struct sk_buff *skb, int skb_size);
508

H
Herbert Xu 已提交
509
extern struct sk_buff *skb_morph(struct sk_buff *dst, struct sk_buff *src);
V
Victor Fusco 已提交
510
extern struct sk_buff *skb_clone(struct sk_buff *skb,
A
Al Viro 已提交
511
				 gfp_t priority);
V
Victor Fusco 已提交
512
extern struct sk_buff *skb_copy(const struct sk_buff *skb,
A
Al Viro 已提交
513
				gfp_t priority);
V
Victor Fusco 已提交
514
extern struct sk_buff *pskb_copy(struct sk_buff *skb,
A
Al Viro 已提交
515
				 gfp_t gfp_mask);
L
Linus Torvalds 已提交
516
extern int	       pskb_expand_head(struct sk_buff *skb,
V
Victor Fusco 已提交
517
					int nhead, int ntail,
A
Al Viro 已提交
518
					gfp_t gfp_mask);
L
Linus Torvalds 已提交
519 520 521 522
extern struct sk_buff *skb_realloc_headroom(struct sk_buff *skb,
					    unsigned int headroom);
extern struct sk_buff *skb_copy_expand(const struct sk_buff *skb,
				       int newheadroom, int newtailroom,
A
Al Viro 已提交
523
				       gfp_t priority);
524 525 526 527 528
extern int	       skb_to_sgvec(struct sk_buff *skb,
				    struct scatterlist *sg, int offset,
				    int len);
extern int	       skb_cow_data(struct sk_buff *skb, int tailbits,
				    struct sk_buff **trailer);
529
extern int	       skb_pad(struct sk_buff *skb, int pad);
530
#define dev_kfree_skb(a)	consume_skb(a)
L
Linus Torvalds 已提交
531

532 533 534 535 536
extern int skb_append_datato_frags(struct sock *sk, struct sk_buff *skb,
			int getfrag(void *from, char *to, int offset,
			int len,int odd, struct sk_buff *skb),
			void *from, int length);

E
Eric Dumazet 已提交
537
struct skb_seq_state {
538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553
	__u32		lower_offset;
	__u32		upper_offset;
	__u32		frag_idx;
	__u32		stepped_offset;
	struct sk_buff	*root_skb;
	struct sk_buff	*cur_skb;
	__u8		*frag_data;
};

extern void	      skb_prepare_seq_read(struct sk_buff *skb,
					   unsigned int from, unsigned int to,
					   struct skb_seq_state *st);
extern unsigned int   skb_seq_read(unsigned int consumed, const u8 **data,
				   struct skb_seq_state *st);
extern void	      skb_abort_seq_read(struct skb_seq_state *st);

554 555 556 557
extern unsigned int   skb_find_text(struct sk_buff *skb, unsigned int from,
				    unsigned int to, struct ts_config *config,
				    struct ts_state *state);

558 559 560 561 562 563 564 565 566
extern __u32 __skb_get_rxhash(struct sk_buff *skb);
static inline __u32 skb_get_rxhash(struct sk_buff *skb)
{
	if (!skb->rxhash)
		skb->rxhash = __skb_get_rxhash(skb);

	return skb->rxhash;
}

567 568 569 570 571 572 573 574 575 576 577 578
#ifdef NET_SKBUFF_DATA_USES_OFFSET
static inline unsigned char *skb_end_pointer(const struct sk_buff *skb)
{
	return skb->head + skb->end;
}
#else
static inline unsigned char *skb_end_pointer(const struct sk_buff *skb)
{
	return skb->end;
}
#endif

L
Linus Torvalds 已提交
579
/* Internal */
580
#define skb_shinfo(SKB)	((struct skb_shared_info *)(skb_end_pointer(SKB)))
L
Linus Torvalds 已提交
581

582 583 584 585 586
static inline struct skb_shared_hwtstamps *skb_hwtstamps(struct sk_buff *skb)
{
	return &skb_shinfo(skb)->hwtstamps;
}

L
Linus Torvalds 已提交
587 588 589 590 591 592 593 594 595 596 597
/**
 *	skb_queue_empty - check if a queue is empty
 *	@list: queue head
 *
 *	Returns true if the queue is empty, false otherwise.
 */
static inline int skb_queue_empty(const struct sk_buff_head *list)
{
	return list->next == (struct sk_buff *)list;
}

D
David S. Miller 已提交
598 599 600 601 602 603 604 605 606 607
/**
 *	skb_queue_is_last - check if skb is the last entry in the queue
 *	@list: queue head
 *	@skb: buffer
 *
 *	Returns true if @skb is the last buffer on the list.
 */
static inline bool skb_queue_is_last(const struct sk_buff_head *list,
				     const struct sk_buff *skb)
{
E
Eric Dumazet 已提交
608
	return skb->next == (struct sk_buff *)list;
D
David S. Miller 已提交
609 610
}

611 612 613 614 615 616 617 618 619 620
/**
 *	skb_queue_is_first - check if skb is the first entry in the queue
 *	@list: queue head
 *	@skb: buffer
 *
 *	Returns true if @skb is the first buffer on the list.
 */
static inline bool skb_queue_is_first(const struct sk_buff_head *list,
				      const struct sk_buff *skb)
{
E
Eric Dumazet 已提交
621
	return skb->prev == (struct sk_buff *)list;
622 623
}

D
David S. Miller 已提交
624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641
/**
 *	skb_queue_next - return the next packet in the queue
 *	@list: queue head
 *	@skb: current buffer
 *
 *	Return the next packet in @list after @skb.  It is only valid to
 *	call this if skb_queue_is_last() evaluates to false.
 */
static inline struct sk_buff *skb_queue_next(const struct sk_buff_head *list,
					     const struct sk_buff *skb)
{
	/* This BUG_ON may seem severe, but if we just return then we
	 * are going to dereference garbage.
	 */
	BUG_ON(skb_queue_is_last(list, skb));
	return skb->next;
}

642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659
/**
 *	skb_queue_prev - return the prev packet in the queue
 *	@list: queue head
 *	@skb: current buffer
 *
 *	Return the prev packet in @list before @skb.  It is only valid to
 *	call this if skb_queue_is_first() evaluates to false.
 */
static inline struct sk_buff *skb_queue_prev(const struct sk_buff_head *list,
					     const struct sk_buff *skb)
{
	/* This BUG_ON may seem severe, but if we just return then we
	 * are going to dereference garbage.
	 */
	BUG_ON(skb_queue_is_first(list, skb));
	return skb->prev;
}

L
Linus Torvalds 已提交
660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750
/**
 *	skb_get - reference buffer
 *	@skb: buffer to reference
 *
 *	Makes another reference to a socket buffer and returns a pointer
 *	to the buffer.
 */
static inline struct sk_buff *skb_get(struct sk_buff *skb)
{
	atomic_inc(&skb->users);
	return skb;
}

/*
 * If users == 1, we are the only owner and are can avoid redundant
 * atomic change.
 */

/**
 *	skb_cloned - is the buffer a clone
 *	@skb: buffer to check
 *
 *	Returns true if the buffer was generated with skb_clone() and is
 *	one of multiple shared copies of the buffer. Cloned buffers are
 *	shared data so must not be written to under normal circumstances.
 */
static inline int skb_cloned(const struct sk_buff *skb)
{
	return skb->cloned &&
	       (atomic_read(&skb_shinfo(skb)->dataref) & SKB_DATAREF_MASK) != 1;
}

/**
 *	skb_header_cloned - is the header a clone
 *	@skb: buffer to check
 *
 *	Returns true if modifying the header part of the buffer requires
 *	the data to be copied.
 */
static inline int skb_header_cloned(const struct sk_buff *skb)
{
	int dataref;

	if (!skb->cloned)
		return 0;

	dataref = atomic_read(&skb_shinfo(skb)->dataref);
	dataref = (dataref & SKB_DATAREF_MASK) - (dataref >> SKB_DATAREF_SHIFT);
	return dataref != 1;
}

/**
 *	skb_header_release - release reference to header
 *	@skb: buffer to operate on
 *
 *	Drop a reference to the header part of the buffer.  This is done
 *	by acquiring a payload reference.  You must not read from the header
 *	part of skb->data after this.
 */
static inline void skb_header_release(struct sk_buff *skb)
{
	BUG_ON(skb->nohdr);
	skb->nohdr = 1;
	atomic_add(1 << SKB_DATAREF_SHIFT, &skb_shinfo(skb)->dataref);
}

/**
 *	skb_shared - is the buffer shared
 *	@skb: buffer to check
 *
 *	Returns true if more than one person has a reference to this
 *	buffer.
 */
static inline int skb_shared(const struct sk_buff *skb)
{
	return atomic_read(&skb->users) != 1;
}

/**
 *	skb_share_check - check if buffer is shared and if so clone it
 *	@skb: buffer to check
 *	@pri: priority for memory allocation
 *
 *	If the buffer is shared the buffer is cloned and the old copy
 *	drops a reference. A new clone with a single reference is returned.
 *	If the buffer is not shared the original buffer is returned. When
 *	being called from interrupt status or with spinlocks held pri must
 *	be GFP_ATOMIC.
 *
 *	NULL is returned on a memory allocation failure.
 */
V
Victor Fusco 已提交
751
static inline struct sk_buff *skb_share_check(struct sk_buff *skb,
A
Al Viro 已提交
752
					      gfp_t pri)
L
Linus Torvalds 已提交
753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782
{
	might_sleep_if(pri & __GFP_WAIT);
	if (skb_shared(skb)) {
		struct sk_buff *nskb = skb_clone(skb, pri);
		kfree_skb(skb);
		skb = nskb;
	}
	return skb;
}

/*
 *	Copy shared buffers into a new sk_buff. We effectively do COW on
 *	packets to handle cases where we have a local reader and forward
 *	and a couple of other messy ones. The normal one is tcpdumping
 *	a packet thats being forwarded.
 */

/**
 *	skb_unshare - make a copy of a shared buffer
 *	@skb: buffer to check
 *	@pri: priority for memory allocation
 *
 *	If the socket buffer is a clone then this function creates a new
 *	copy of the data, drops a reference count on the old copy and returns
 *	the new copy with the reference count at 1. If the buffer is not a clone
 *	the original buffer is returned. When called with a spinlock held or
 *	from interrupt state @pri must be %GFP_ATOMIC
 *
 *	%NULL is returned on a memory allocation failure.
 */
783
static inline struct sk_buff *skb_unshare(struct sk_buff *skb,
A
Al Viro 已提交
784
					  gfp_t pri)
L
Linus Torvalds 已提交
785 786 787 788 789 790 791 792 793 794 795
{
	might_sleep_if(pri & __GFP_WAIT);
	if (skb_cloned(skb)) {
		struct sk_buff *nskb = skb_copy(skb, pri);
		kfree_skb(skb);	/* Free our shared copy */
		skb = nskb;
	}
	return skb;
}

/**
796
 *	skb_peek - peek at the head of an &sk_buff_head
L
Linus Torvalds 已提交
797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816
 *	@list_: list to peek at
 *
 *	Peek an &sk_buff. Unlike most other operations you _MUST_
 *	be careful with this one. A peek leaves the buffer on the
 *	list and someone else may run off with it. You must hold
 *	the appropriate locks or have a private queue to do this.
 *
 *	Returns %NULL for an empty list or a pointer to the head element.
 *	The reference count is not incremented and the reference is therefore
 *	volatile. Use with caution.
 */
static inline struct sk_buff *skb_peek(struct sk_buff_head *list_)
{
	struct sk_buff *list = ((struct sk_buff *)list_)->next;
	if (list == (struct sk_buff *)list_)
		list = NULL;
	return list;
}

/**
817
 *	skb_peek_tail - peek at the tail of an &sk_buff_head
L
Linus Torvalds 已提交
818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847
 *	@list_: list to peek at
 *
 *	Peek an &sk_buff. Unlike most other operations you _MUST_
 *	be careful with this one. A peek leaves the buffer on the
 *	list and someone else may run off with it. You must hold
 *	the appropriate locks or have a private queue to do this.
 *
 *	Returns %NULL for an empty list or a pointer to the tail element.
 *	The reference count is not incremented and the reference is therefore
 *	volatile. Use with caution.
 */
static inline struct sk_buff *skb_peek_tail(struct sk_buff_head *list_)
{
	struct sk_buff *list = ((struct sk_buff *)list_)->prev;
	if (list == (struct sk_buff *)list_)
		list = NULL;
	return list;
}

/**
 *	skb_queue_len	- get queue length
 *	@list_: list to measure
 *
 *	Return the length of an &sk_buff queue.
 */
static inline __u32 skb_queue_len(const struct sk_buff_head *list_)
{
	return list_->qlen;
}

848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863
/**
 *	__skb_queue_head_init - initialize non-spinlock portions of sk_buff_head
 *	@list: queue to initialize
 *
 *	This initializes only the list and queue length aspects of
 *	an sk_buff_head object.  This allows to initialize the list
 *	aspects of an sk_buff_head without reinitializing things like
 *	the spinlock.  It can also be used for on-stack sk_buff_head
 *	objects where the spinlock is known to not be used.
 */
static inline void __skb_queue_head_init(struct sk_buff_head *list)
{
	list->prev = list->next = (struct sk_buff *)list;
	list->qlen = 0;
}

864 865 866 867 868 869 870 871
/*
 * This function creates a split out lock class for each invocation;
 * this is needed for now since a whole lot of users of the skb-queue
 * infrastructure in drivers have different locking usage (in hardirq)
 * than the networking core (in softirq only). In the long run either the
 * network layer or drivers should need annotation to consolidate the
 * main types of usage into 3 classes.
 */
L
Linus Torvalds 已提交
872 873 874
static inline void skb_queue_head_init(struct sk_buff_head *list)
{
	spin_lock_init(&list->lock);
875
	__skb_queue_head_init(list);
L
Linus Torvalds 已提交
876 877
}

878 879 880 881 882 883 884
static inline void skb_queue_head_init_class(struct sk_buff_head *list,
		struct lock_class_key *class)
{
	skb_queue_head_init(list);
	lockdep_set_class(&list->lock, class);
}

L
Linus Torvalds 已提交
885
/*
886
 *	Insert an sk_buff on a list.
L
Linus Torvalds 已提交
887 888 889 890
 *
 *	The "__skb_xxxx()" functions are the non-atomic ones that
 *	can only be called with interrupts disabled.
 */
891 892 893 894 895 896 897 898 899 900
extern void        skb_insert(struct sk_buff *old, struct sk_buff *newsk, struct sk_buff_head *list);
static inline void __skb_insert(struct sk_buff *newsk,
				struct sk_buff *prev, struct sk_buff *next,
				struct sk_buff_head *list)
{
	newsk->next = next;
	newsk->prev = prev;
	next->prev  = prev->next = newsk;
	list->qlen++;
}
L
Linus Torvalds 已提交
901

902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925
static inline void __skb_queue_splice(const struct sk_buff_head *list,
				      struct sk_buff *prev,
				      struct sk_buff *next)
{
	struct sk_buff *first = list->next;
	struct sk_buff *last = list->prev;

	first->prev = prev;
	prev->next = first;

	last->next = next;
	next->prev = last;
}

/**
 *	skb_queue_splice - join two skb lists, this is designed for stacks
 *	@list: the new list to add
 *	@head: the place to add it in the first list
 */
static inline void skb_queue_splice(const struct sk_buff_head *list,
				    struct sk_buff_head *head)
{
	if (!skb_queue_empty(list)) {
		__skb_queue_splice(list, (struct sk_buff *) head, head->next);
926
		head->qlen += list->qlen;
927 928 929 930 931 932 933 934 935 936 937 938 939 940 941
	}
}

/**
 *	skb_queue_splice - join two skb lists and reinitialise the emptied list
 *	@list: the new list to add
 *	@head: the place to add it in the first list
 *
 *	The list at @list is reinitialised
 */
static inline void skb_queue_splice_init(struct sk_buff_head *list,
					 struct sk_buff_head *head)
{
	if (!skb_queue_empty(list)) {
		__skb_queue_splice(list, (struct sk_buff *) head, head->next);
942
		head->qlen += list->qlen;
943 944 945 946 947 948 949 950 951 952 953 954 955 956
		__skb_queue_head_init(list);
	}
}

/**
 *	skb_queue_splice_tail - join two skb lists, each list being a queue
 *	@list: the new list to add
 *	@head: the place to add it in the first list
 */
static inline void skb_queue_splice_tail(const struct sk_buff_head *list,
					 struct sk_buff_head *head)
{
	if (!skb_queue_empty(list)) {
		__skb_queue_splice(list, head->prev, (struct sk_buff *) head);
957
		head->qlen += list->qlen;
958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973
	}
}

/**
 *	skb_queue_splice_tail - join two skb lists and reinitialise the emptied list
 *	@list: the new list to add
 *	@head: the place to add it in the first list
 *
 *	Each of the lists is a queue.
 *	The list at @list is reinitialised
 */
static inline void skb_queue_splice_tail_init(struct sk_buff_head *list,
					      struct sk_buff_head *head)
{
	if (!skb_queue_empty(list)) {
		__skb_queue_splice(list, head->prev, (struct sk_buff *) head);
974
		head->qlen += list->qlen;
975 976 977 978
		__skb_queue_head_init(list);
	}
}

L
Linus Torvalds 已提交
979
/**
980
 *	__skb_queue_after - queue a buffer at the list head
L
Linus Torvalds 已提交
981
 *	@list: list to use
982
 *	@prev: place after this buffer
L
Linus Torvalds 已提交
983 984
 *	@newsk: buffer to queue
 *
985
 *	Queue a buffer int the middle of a list. This function takes no locks
L
Linus Torvalds 已提交
986 987 988 989
 *	and you must therefore hold required locks before calling it.
 *
 *	A buffer cannot be placed on two lists at the same time.
 */
990 991 992
static inline void __skb_queue_after(struct sk_buff_head *list,
				     struct sk_buff *prev,
				     struct sk_buff *newsk)
L
Linus Torvalds 已提交
993
{
994
	__skb_insert(newsk, prev, prev->next, list);
L
Linus Torvalds 已提交
995 996
}

997 998 999
extern void skb_append(struct sk_buff *old, struct sk_buff *newsk,
		       struct sk_buff_head *list);

1000 1001 1002 1003 1004 1005 1006
static inline void __skb_queue_before(struct sk_buff_head *list,
				      struct sk_buff *next,
				      struct sk_buff *newsk)
{
	__skb_insert(newsk, next->prev, next, list);
}

1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023
/**
 *	__skb_queue_head - queue a buffer at the list head
 *	@list: list to use
 *	@newsk: buffer to queue
 *
 *	Queue a buffer at the start of a list. This function takes no locks
 *	and you must therefore hold required locks before calling it.
 *
 *	A buffer cannot be placed on two lists at the same time.
 */
extern void skb_queue_head(struct sk_buff_head *list, struct sk_buff *newsk);
static inline void __skb_queue_head(struct sk_buff_head *list,
				    struct sk_buff *newsk)
{
	__skb_queue_after(list, (struct sk_buff *)list, newsk);
}

L
Linus Torvalds 已提交
1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037
/**
 *	__skb_queue_tail - queue a buffer at the list tail
 *	@list: list to use
 *	@newsk: buffer to queue
 *
 *	Queue a buffer at the end of a list. This function takes no locks
 *	and you must therefore hold required locks before calling it.
 *
 *	A buffer cannot be placed on two lists at the same time.
 */
extern void skb_queue_tail(struct sk_buff_head *list, struct sk_buff *newsk);
static inline void __skb_queue_tail(struct sk_buff_head *list,
				   struct sk_buff *newsk)
{
1038
	__skb_queue_before(list, (struct sk_buff *)list, newsk);
L
Linus Torvalds 已提交
1039 1040 1041 1042 1043 1044
}

/*
 * remove sk_buff from list. _Must_ be called atomically, and with
 * the list known..
 */
D
David S. Miller 已提交
1045
extern void	   skb_unlink(struct sk_buff *skb, struct sk_buff_head *list);
L
Linus Torvalds 已提交
1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057
static inline void __skb_unlink(struct sk_buff *skb, struct sk_buff_head *list)
{
	struct sk_buff *next, *prev;

	list->qlen--;
	next	   = skb->next;
	prev	   = skb->prev;
	skb->next  = skb->prev = NULL;
	next->prev = prev;
	prev->next = next;
}

1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073
/**
 *	__skb_dequeue - remove from the head of the queue
 *	@list: list to dequeue from
 *
 *	Remove the head of the list. This function does not take any locks
 *	so must be used with appropriate locks held only. The head item is
 *	returned or %NULL if the list is empty.
 */
extern struct sk_buff *skb_dequeue(struct sk_buff_head *list);
static inline struct sk_buff *__skb_dequeue(struct sk_buff_head *list)
{
	struct sk_buff *skb = skb_peek(list);
	if (skb)
		__skb_unlink(skb, list);
	return skb;
}
L
Linus Torvalds 已提交
1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 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 1119 1120 1121 1122

/**
 *	__skb_dequeue_tail - remove from the tail of the queue
 *	@list: list to dequeue from
 *
 *	Remove the tail of the list. This function does not take any locks
 *	so must be used with appropriate locks held only. The tail item is
 *	returned or %NULL if the list is empty.
 */
extern struct sk_buff *skb_dequeue_tail(struct sk_buff_head *list);
static inline struct sk_buff *__skb_dequeue_tail(struct sk_buff_head *list)
{
	struct sk_buff *skb = skb_peek_tail(list);
	if (skb)
		__skb_unlink(skb, list);
	return skb;
}


static inline int skb_is_nonlinear(const struct sk_buff *skb)
{
	return skb->data_len;
}

static inline unsigned int skb_headlen(const struct sk_buff *skb)
{
	return skb->len - skb->data_len;
}

static inline int skb_pagelen(const struct sk_buff *skb)
{
	int i, len = 0;

	for (i = (int)skb_shinfo(skb)->nr_frags - 1; i >= 0; i--)
		len += skb_shinfo(skb)->frags[i].size;
	return len + skb_headlen(skb);
}

static inline void skb_fill_page_desc(struct sk_buff *skb, int i,
				      struct page *page, int off, int size)
{
	skb_frag_t *frag = &skb_shinfo(skb)->frags[i];

	frag->page		  = page;
	frag->page_offset	  = off;
	frag->size		  = size;
	skb_shinfo(skb)->nr_frags = i + 1;
}

P
Peter Zijlstra 已提交
1123 1124 1125
extern void skb_add_rx_frag(struct sk_buff *skb, int i, struct page *page,
			    int off, int size);

L
Linus Torvalds 已提交
1126
#define SKB_PAGE_ASSERT(skb) 	BUG_ON(skb_shinfo(skb)->nr_frags)
1127
#define SKB_FRAG_ASSERT(skb) 	BUG_ON(skb_has_frag_list(skb))
L
Linus Torvalds 已提交
1128 1129
#define SKB_LINEAR_ASSERT(skb)  BUG_ON(skb_is_nonlinear(skb))

1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160
#ifdef NET_SKBUFF_DATA_USES_OFFSET
static inline unsigned char *skb_tail_pointer(const struct sk_buff *skb)
{
	return skb->head + skb->tail;
}

static inline void skb_reset_tail_pointer(struct sk_buff *skb)
{
	skb->tail = skb->data - skb->head;
}

static inline void skb_set_tail_pointer(struct sk_buff *skb, const int offset)
{
	skb_reset_tail_pointer(skb);
	skb->tail += offset;
}
#else /* NET_SKBUFF_DATA_USES_OFFSET */
static inline unsigned char *skb_tail_pointer(const struct sk_buff *skb)
{
	return skb->tail;
}

static inline void skb_reset_tail_pointer(struct sk_buff *skb)
{
	skb->tail = skb->data;
}

static inline void skb_set_tail_pointer(struct sk_buff *skb, const int offset)
{
	skb->tail = skb->data + offset;
}
1161

1162 1163
#endif /* NET_SKBUFF_DATA_USES_OFFSET */

L
Linus Torvalds 已提交
1164 1165 1166
/*
 *	Add data to an sk_buff
 */
1167
extern unsigned char *skb_put(struct sk_buff *skb, unsigned int len);
L
Linus Torvalds 已提交
1168 1169
static inline unsigned char *__skb_put(struct sk_buff *skb, unsigned int len)
{
1170
	unsigned char *tmp = skb_tail_pointer(skb);
L
Linus Torvalds 已提交
1171 1172 1173 1174 1175 1176
	SKB_LINEAR_ASSERT(skb);
	skb->tail += len;
	skb->len  += len;
	return tmp;
}

1177
extern unsigned char *skb_push(struct sk_buff *skb, unsigned int len);
L
Linus Torvalds 已提交
1178 1179 1180 1181 1182 1183 1184
static inline unsigned char *__skb_push(struct sk_buff *skb, unsigned int len)
{
	skb->data -= len;
	skb->len  += len;
	return skb->data;
}

1185
extern unsigned char *skb_pull(struct sk_buff *skb, unsigned int len);
L
Linus Torvalds 已提交
1186 1187 1188 1189 1190 1191 1192
static inline unsigned char *__skb_pull(struct sk_buff *skb, unsigned int len)
{
	skb->len -= len;
	BUG_ON(skb->len < skb->data_len);
	return skb->data += len;
}

1193 1194 1195 1196 1197
static inline unsigned char *skb_pull_inline(struct sk_buff *skb, unsigned int len)
{
	return unlikely(len > skb->len) ? NULL : __skb_pull(skb, len);
}

L
Linus Torvalds 已提交
1198 1199 1200 1201 1202
extern unsigned char *__pskb_pull_tail(struct sk_buff *skb, int delta);

static inline unsigned char *__pskb_pull(struct sk_buff *skb, unsigned int len)
{
	if (len > skb_headlen(skb) &&
G
Gerrit Renker 已提交
1203
	    !__pskb_pull_tail(skb, len - skb_headlen(skb)))
L
Linus Torvalds 已提交
1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219
		return NULL;
	skb->len -= len;
	return skb->data += len;
}

static inline unsigned char *pskb_pull(struct sk_buff *skb, unsigned int len)
{
	return unlikely(len > skb->len) ? NULL : __pskb_pull(skb, len);
}

static inline int pskb_may_pull(struct sk_buff *skb, unsigned int len)
{
	if (likely(len <= skb_headlen(skb)))
		return 1;
	if (unlikely(len > skb->len))
		return 0;
G
Gerrit Renker 已提交
1220
	return __pskb_pull_tail(skb, len - skb_headlen(skb)) != NULL;
L
Linus Torvalds 已提交
1221 1222 1223 1224 1225 1226 1227 1228
}

/**
 *	skb_headroom - bytes at buffer head
 *	@skb: buffer to check
 *
 *	Return the number of bytes of free space at the head of an &sk_buff.
 */
1229
static inline unsigned int skb_headroom(const struct sk_buff *skb)
L
Linus Torvalds 已提交
1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241
{
	return skb->data - skb->head;
}

/**
 *	skb_tailroom - bytes at buffer end
 *	@skb: buffer to check
 *
 *	Return the number of bytes of free space at the tail of an sk_buff
 */
static inline int skb_tailroom(const struct sk_buff *skb)
{
1242
	return skb_is_nonlinear(skb) ? 0 : skb->end - skb->tail;
L
Linus Torvalds 已提交
1243 1244 1245 1246 1247 1248 1249 1250 1251 1252
}

/**
 *	skb_reserve - adjust headroom
 *	@skb: buffer to alter
 *	@len: bytes to move
 *
 *	Increase the headroom of an empty &sk_buff by reducing the tail
 *	room. This is only allowed for an empty buffer.
 */
1253
static inline void skb_reserve(struct sk_buff *skb, int len)
L
Linus Torvalds 已提交
1254 1255 1256 1257 1258
{
	skb->data += len;
	skb->tail += len;
}

1259
#ifdef NET_SKBUFF_DATA_USES_OFFSET
1260 1261
static inline unsigned char *skb_transport_header(const struct sk_buff *skb)
{
1262
	return skb->head + skb->transport_header;
1263 1264
}

1265 1266
static inline void skb_reset_transport_header(struct sk_buff *skb)
{
1267
	skb->transport_header = skb->data - skb->head;
1268 1269
}

1270 1271 1272
static inline void skb_set_transport_header(struct sk_buff *skb,
					    const int offset)
{
1273 1274
	skb_reset_transport_header(skb);
	skb->transport_header += offset;
1275 1276
}

1277 1278
static inline unsigned char *skb_network_header(const struct sk_buff *skb)
{
1279
	return skb->head + skb->network_header;
1280 1281
}

1282 1283
static inline void skb_reset_network_header(struct sk_buff *skb)
{
1284
	skb->network_header = skb->data - skb->head;
1285 1286
}

1287 1288
static inline void skb_set_network_header(struct sk_buff *skb, const int offset)
{
1289 1290
	skb_reset_network_header(skb);
	skb->network_header += offset;
1291 1292
}

1293
static inline unsigned char *skb_mac_header(const struct sk_buff *skb)
1294
{
1295
	return skb->head + skb->mac_header;
1296 1297
}

1298
static inline int skb_mac_header_was_set(const struct sk_buff *skb)
1299
{
1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344
	return skb->mac_header != ~0U;
}

static inline void skb_reset_mac_header(struct sk_buff *skb)
{
	skb->mac_header = skb->data - skb->head;
}

static inline void skb_set_mac_header(struct sk_buff *skb, const int offset)
{
	skb_reset_mac_header(skb);
	skb->mac_header += offset;
}

#else /* NET_SKBUFF_DATA_USES_OFFSET */

static inline unsigned char *skb_transport_header(const struct sk_buff *skb)
{
	return skb->transport_header;
}

static inline void skb_reset_transport_header(struct sk_buff *skb)
{
	skb->transport_header = skb->data;
}

static inline void skb_set_transport_header(struct sk_buff *skb,
					    const int offset)
{
	skb->transport_header = skb->data + offset;
}

static inline unsigned char *skb_network_header(const struct sk_buff *skb)
{
	return skb->network_header;
}

static inline void skb_reset_network_header(struct sk_buff *skb)
{
	skb->network_header = skb->data;
}

static inline void skb_set_network_header(struct sk_buff *skb, const int offset)
{
	skb->network_header = skb->data + offset;
1345 1346
}

1347 1348
static inline unsigned char *skb_mac_header(const struct sk_buff *skb)
{
1349
	return skb->mac_header;
1350 1351 1352 1353
}

static inline int skb_mac_header_was_set(const struct sk_buff *skb)
{
1354
	return skb->mac_header != NULL;
1355 1356
}

1357 1358
static inline void skb_reset_mac_header(struct sk_buff *skb)
{
1359
	skb->mac_header = skb->data;
1360 1361
}

1362 1363
static inline void skb_set_mac_header(struct sk_buff *skb, const int offset)
{
1364
	skb->mac_header = skb->data + offset;
1365
}
1366 1367
#endif /* NET_SKBUFF_DATA_USES_OFFSET */

1368 1369 1370 1371 1372
static inline int skb_checksum_start_offset(const struct sk_buff *skb)
{
	return skb->csum_start - skb_headroom(skb);
}

1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386
static inline int skb_transport_offset(const struct sk_buff *skb)
{
	return skb_transport_header(skb) - skb->data;
}

static inline u32 skb_network_header_len(const struct sk_buff *skb)
{
	return skb->transport_header - skb->network_header;
}

static inline int skb_network_offset(const struct sk_buff *skb)
{
	return skb_network_header(skb) - skb->data;
}
1387

1388 1389 1390 1391 1392
static inline int pskb_network_may_pull(struct sk_buff *skb, unsigned int len)
{
	return pskb_may_pull(skb, skb_network_offset(skb) + len);
}

L
Linus Torvalds 已提交
1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403
/*
 * CPUs often take a performance hit when accessing unaligned memory
 * locations. The actual performance hit varies, it can be small if the
 * hardware handles it or large if we have to take an exception and fix it
 * in software.
 *
 * Since an ethernet header is 14 bytes network drivers often end up with
 * the IP header at an unaligned offset. The IP header can be aligned by
 * shifting the start of the packet by 2 bytes. Drivers should do this
 * with:
 *
1404
 * skb_reserve(skb, NET_IP_ALIGN);
L
Linus Torvalds 已提交
1405 1406 1407 1408
 *
 * The downside to this alignment of the IP header is that the DMA is now
 * unaligned. On some architectures the cost of an unaligned DMA is high
 * and this cost outweighs the gains made by aligning the IP header.
1409
 *
L
Linus Torvalds 已提交
1410 1411 1412 1413 1414 1415 1416
 * Since this trade off varies between architectures, we allow NET_IP_ALIGN
 * to be overridden.
 */
#ifndef NET_IP_ALIGN
#define NET_IP_ALIGN	2
#endif

1417 1418 1419 1420
/*
 * The networking layer reserves some headroom in skb data (via
 * dev_alloc_skb). This is used to avoid having to reallocate skb data when
 * the header has to grow. In the default case, if the header has to grow
1421
 * 32 bytes or less we avoid the reallocation.
1422 1423 1424 1425 1426 1427 1428
 *
 * Unfortunately this headroom changes the DMA alignment of the resulting
 * network packet. As for NET_IP_ALIGN, this unaligned DMA is expensive
 * on some architectures. An architecture can override this value,
 * perhaps setting it to a cacheline in size (since that will maintain
 * cacheline alignment of the DMA). It must be a power of 2.
 *
1429
 * Various parts of the networking layer expect at least 32 bytes of
1430
 * headroom, you should not reduce this.
1431 1432 1433 1434
 *
 * Using max(32, L1_CACHE_BYTES) makes sense (especially with RPS)
 * to reduce average number of cache lines per packet.
 * get_rps_cpus() for example only access one 64 bytes aligned block :
1435
 * NET_IP_ALIGN(2) + ethernet_header(14) + IP_header(20/40) + ports(8)
1436 1437
 */
#ifndef NET_SKB_PAD
1438
#define NET_SKB_PAD	max(32, L1_CACHE_BYTES)
1439 1440
#endif

1441
extern int ___pskb_trim(struct sk_buff *skb, unsigned int len);
L
Linus Torvalds 已提交
1442 1443 1444

static inline void __skb_trim(struct sk_buff *skb, unsigned int len)
{
1445 1446 1447 1448
	if (unlikely(skb->data_len)) {
		WARN_ON(1);
		return;
	}
1449 1450
	skb->len = len;
	skb_set_tail_pointer(skb, len);
L
Linus Torvalds 已提交
1451 1452
}

1453
extern void skb_trim(struct sk_buff *skb, unsigned int len);
L
Linus Torvalds 已提交
1454 1455 1456

static inline int __pskb_trim(struct sk_buff *skb, unsigned int len)
{
1457 1458 1459 1460
	if (skb->data_len)
		return ___pskb_trim(skb, len);
	__skb_trim(skb, len);
	return 0;
L
Linus Torvalds 已提交
1461 1462 1463 1464 1465 1466 1467
}

static inline int pskb_trim(struct sk_buff *skb, unsigned int len)
{
	return (len < skb->len) ? __pskb_trim(skb, len) : 0;
}

1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482
/**
 *	pskb_trim_unique - remove end from a paged unique (not cloned) buffer
 *	@skb: buffer to alter
 *	@len: new length
 *
 *	This is identical to pskb_trim except that the caller knows that
 *	the skb is not cloned so we should never get an error due to out-
 *	of-memory.
 */
static inline void pskb_trim_unique(struct sk_buff *skb, unsigned int len)
{
	int err = pskb_trim(skb, len);
	BUG_ON(err);
}

L
Linus Torvalds 已提交
1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515
/**
 *	skb_orphan - orphan a buffer
 *	@skb: buffer to orphan
 *
 *	If a buffer currently has an owner then we call the owner's
 *	destructor function and make the @skb unowned. The buffer continues
 *	to exist but is no longer charged to its former owner.
 */
static inline void skb_orphan(struct sk_buff *skb)
{
	if (skb->destructor)
		skb->destructor(skb);
	skb->destructor = NULL;
	skb->sk		= NULL;
}

/**
 *	__skb_queue_purge - empty a list
 *	@list: list to empty
 *
 *	Delete all buffers on an &sk_buff list. Each buffer is removed from
 *	the list and one reference dropped. This function does not take the
 *	list lock and the caller must hold the relevant locks to use it.
 */
extern void skb_queue_purge(struct sk_buff_head *list);
static inline void __skb_queue_purge(struct sk_buff_head *list)
{
	struct sk_buff *skb;
	while ((skb = __skb_dequeue(list)) != NULL)
		kfree_skb(skb);
}

/**
1516
 *	__dev_alloc_skb - allocate an skbuff for receiving
L
Linus Torvalds 已提交
1517 1518 1519 1520 1521 1522 1523 1524
 *	@length: length to allocate
 *	@gfp_mask: get_free_pages mask, passed to alloc_skb
 *
 *	Allocate a new &sk_buff and assign it a usage count of one. The
 *	buffer has unspecified headroom built in. Users should allocate
 *	the headroom they think they need without accounting for the
 *	built in space. The built in space is used for optimisations.
 *
1525
 *	%NULL is returned if there is no free memory.
L
Linus Torvalds 已提交
1526 1527
 */
static inline struct sk_buff *__dev_alloc_skb(unsigned int length,
A
Al Viro 已提交
1528
					      gfp_t gfp_mask)
L
Linus Torvalds 已提交
1529
{
1530
	struct sk_buff *skb = alloc_skb(length + NET_SKB_PAD, gfp_mask);
L
Linus Torvalds 已提交
1531
	if (likely(skb))
1532
		skb_reserve(skb, NET_SKB_PAD);
L
Linus Torvalds 已提交
1533 1534 1535
	return skb;
}

1536
extern struct sk_buff *dev_alloc_skb(unsigned int length);
L
Linus Torvalds 已提交
1537

1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559
extern struct sk_buff *__netdev_alloc_skb(struct net_device *dev,
		unsigned int length, gfp_t gfp_mask);

/**
 *	netdev_alloc_skb - allocate an skbuff for rx on a specific device
 *	@dev: network device to receive on
 *	@length: length to allocate
 *
 *	Allocate a new &sk_buff and assign it a usage count of one. The
 *	buffer has unspecified headroom built in. Users should allocate
 *	the headroom they think they need without accounting for the
 *	built in space. The built in space is used for optimisations.
 *
 *	%NULL is returned if there is no free memory. Although this function
 *	allocates memory it can be called from an interrupt.
 */
static inline struct sk_buff *netdev_alloc_skb(struct net_device *dev,
		unsigned int length)
{
	return __netdev_alloc_skb(dev, length, GFP_ATOMIC);
}

1560 1561 1562 1563 1564 1565 1566 1567 1568 1569
static inline struct sk_buff *netdev_alloc_skb_ip_align(struct net_device *dev,
		unsigned int length)
{
	struct sk_buff *skb = netdev_alloc_skb(dev, length + NET_IP_ALIGN);

	if (NET_IP_ALIGN && skb)
		skb_reserve(skb, NET_IP_ALIGN);
	return skb;
}

E
Eric Dumazet 已提交
1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582
/**
 *	__netdev_alloc_page - allocate a page for ps-rx on a specific device
 *	@dev: network device to receive on
 *	@gfp_mask: alloc_pages_node mask
 *
 * 	Allocate a new page. dev currently unused.
 *
 * 	%NULL is returned if there is no free memory.
 */
static inline struct page *__netdev_alloc_page(struct net_device *dev, gfp_t gfp_mask)
{
	return alloc_pages_node(NUMA_NO_NODE, gfp_mask, 0);
}
P
Peter Zijlstra 已提交
1583 1584 1585 1586 1587

/**
 *	netdev_alloc_page - allocate a page for ps-rx on a specific device
 *	@dev: network device to receive on
 *
E
Eric Dumazet 已提交
1588
 * 	Allocate a new page. dev currently unused.
P
Peter Zijlstra 已提交
1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601
 *
 * 	%NULL is returned if there is no free memory.
 */
static inline struct page *netdev_alloc_page(struct net_device *dev)
{
	return __netdev_alloc_page(dev, GFP_ATOMIC);
}

static inline void netdev_free_page(struct net_device *dev, struct page *page)
{
	__free_page(page);
}

1602 1603 1604 1605 1606 1607 1608 1609
/**
 *	skb_clone_writable - is the header of a clone writable
 *	@skb: buffer to check
 *	@len: length up to which to write
 *
 *	Returns true if modifying the header part of the cloned buffer
 *	does not requires the data to be copied.
 */
1610
static inline int skb_clone_writable(struct sk_buff *skb, unsigned int len)
1611 1612 1613 1614 1615
{
	return !skb_header_cloned(skb) &&
	       skb_headroom(skb) + len <= skb->hdr_len;
}

H
Herbert Xu 已提交
1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631
static inline int __skb_cow(struct sk_buff *skb, unsigned int headroom,
			    int cloned)
{
	int delta = 0;

	if (headroom < NET_SKB_PAD)
		headroom = NET_SKB_PAD;
	if (headroom > skb_headroom(skb))
		delta = headroom - skb_headroom(skb);

	if (delta || cloned)
		return pskb_expand_head(skb, ALIGN(delta, NET_SKB_PAD), 0,
					GFP_ATOMIC);
	return 0;
}

L
Linus Torvalds 已提交
1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645
/**
 *	skb_cow - copy header of skb when it is required
 *	@skb: buffer to cow
 *	@headroom: needed headroom
 *
 *	If the skb passed lacks sufficient headroom or its data part
 *	is shared, data is reallocated. If reallocation fails, an error
 *	is returned and original skb is not changed.
 *
 *	The result is skb with writable area skb->head...skb->tail
 *	and at least @headroom of space at head.
 */
static inline int skb_cow(struct sk_buff *skb, unsigned int headroom)
{
H
Herbert Xu 已提交
1646 1647
	return __skb_cow(skb, headroom, skb_cloned(skb));
}
L
Linus Torvalds 已提交
1648

H
Herbert Xu 已提交
1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661
/**
 *	skb_cow_head - skb_cow but only making the head writable
 *	@skb: buffer to cow
 *	@headroom: needed headroom
 *
 *	This function is identical to skb_cow except that we replace the
 *	skb_cloned check by skb_header_cloned.  It should be used when
 *	you only need to push on some header and do not need to modify
 *	the data.
 */
static inline int skb_cow_head(struct sk_buff *skb, unsigned int headroom)
{
	return __skb_cow(skb, headroom, skb_header_cloned(skb));
L
Linus Torvalds 已提交
1662 1663 1664 1665 1666 1667 1668 1669 1670
}

/**
 *	skb_padto	- pad an skbuff up to a minimal size
 *	@skb: buffer to pad
 *	@len: minimal length
 *
 *	Pads up a buffer to ensure the trailing bytes exist and are
 *	blanked. If the buffer already contains sufficient data it
1671 1672
 *	is untouched. Otherwise it is extended. Returns zero on
 *	success. The skb is freed on error.
L
Linus Torvalds 已提交
1673 1674
 */
 
1675
static inline int skb_padto(struct sk_buff *skb, unsigned int len)
L
Linus Torvalds 已提交
1676 1677 1678
{
	unsigned int size = skb->len;
	if (likely(size >= len))
1679
		return 0;
G
Gerrit Renker 已提交
1680
	return skb_pad(skb, len - size);
L
Linus Torvalds 已提交
1681 1682 1683 1684 1685 1686 1687 1688 1689
}

static inline int skb_add_data(struct sk_buff *skb,
			       char __user *from, int copy)
{
	const int off = skb->len;

	if (skb->ip_summed == CHECKSUM_NONE) {
		int err = 0;
1690
		__wsum csum = csum_and_copy_from_user(from, skb_put(skb, copy),
L
Linus Torvalds 已提交
1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714
							    copy, 0, &err);
		if (!err) {
			skb->csum = csum_block_add(skb->csum, csum, off);
			return 0;
		}
	} else if (!copy_from_user(skb_put(skb, copy), from, copy))
		return 0;

	__skb_trim(skb, off);
	return -EFAULT;
}

static inline int skb_can_coalesce(struct sk_buff *skb, int i,
				   struct page *page, int off)
{
	if (i) {
		struct skb_frag_struct *frag = &skb_shinfo(skb)->frags[i - 1];

		return page == frag->page &&
		       off == frag->page_offset + frag->size;
	}
	return 0;
}

H
Herbert Xu 已提交
1715 1716 1717 1718 1719
static inline int __skb_linearize(struct sk_buff *skb)
{
	return __pskb_pull_tail(skb, skb->data_len) ? 0 : -ENOMEM;
}

L
Linus Torvalds 已提交
1720 1721 1722 1723 1724 1725 1726
/**
 *	skb_linearize - convert paged skb to linear one
 *	@skb: buffer to linarize
 *
 *	If there is no free memory -ENOMEM is returned, otherwise zero
 *	is returned and the old skb data released.
 */
H
Herbert Xu 已提交
1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739
static inline int skb_linearize(struct sk_buff *skb)
{
	return skb_is_nonlinear(skb) ? __skb_linearize(skb) : 0;
}

/**
 *	skb_linearize_cow - make sure skb is linear and writable
 *	@skb: buffer to process
 *
 *	If there is no free memory -ENOMEM is returned, otherwise zero
 *	is returned and the old skb data released.
 */
static inline int skb_linearize_cow(struct sk_buff *skb)
L
Linus Torvalds 已提交
1740
{
H
Herbert Xu 已提交
1741 1742
	return skb_is_nonlinear(skb) || skb_cloned(skb) ?
	       __skb_linearize(skb) : 0;
L
Linus Torvalds 已提交
1743 1744 1745 1746 1747 1748 1749 1750 1751
}

/**
 *	skb_postpull_rcsum - update checksum for received skb after pull
 *	@skb: buffer to update
 *	@start: start of data before pull
 *	@len: length of data pulled
 *
 *	After doing a pull on a received packet, you need to call this to
1752 1753
 *	update the CHECKSUM_COMPLETE checksum, or set ip_summed to
 *	CHECKSUM_NONE so that it can be recomputed from scratch.
L
Linus Torvalds 已提交
1754 1755 1756
 */

static inline void skb_postpull_rcsum(struct sk_buff *skb,
1757
				      const void *start, unsigned int len)
L
Linus Torvalds 已提交
1758
{
1759
	if (skb->ip_summed == CHECKSUM_COMPLETE)
L
Linus Torvalds 已提交
1760 1761 1762
		skb->csum = csum_sub(skb->csum, csum_partial(start, len, 0));
}

1763 1764
unsigned char *skb_pull_rcsum(struct sk_buff *skb, unsigned int len);

L
Linus Torvalds 已提交
1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775
/**
 *	pskb_trim_rcsum - trim received skb and update checksum
 *	@skb: buffer to trim
 *	@len: new length
 *
 *	This is exactly the same as pskb_trim except that it ensures the
 *	checksum of received packets are still valid after the operation.
 */

static inline int pskb_trim_rcsum(struct sk_buff *skb, unsigned int len)
{
1776
	if (likely(len >= skb->len))
L
Linus Torvalds 已提交
1777
		return 0;
1778
	if (skb->ip_summed == CHECKSUM_COMPLETE)
L
Linus Torvalds 已提交
1779 1780 1781 1782 1783 1784
		skb->ip_summed = CHECKSUM_NONE;
	return __pskb_trim(skb, len);
}

#define skb_queue_walk(queue, skb) \
		for (skb = (queue)->next;					\
1785
		     (skb != (struct sk_buff *)(queue));			\
L
Linus Torvalds 已提交
1786 1787
		     skb = skb->next)

1788 1789 1790 1791 1792
#define skb_queue_walk_safe(queue, skb, tmp)					\
		for (skb = (queue)->next, tmp = skb->next;			\
		     skb != (struct sk_buff *)(queue);				\
		     skb = tmp, tmp = skb->next)

1793
#define skb_queue_walk_from(queue, skb)						\
1794
		for (; (skb != (struct sk_buff *)(queue));			\
1795 1796 1797 1798 1799 1800 1801
		     skb = skb->next)

#define skb_queue_walk_from_safe(queue, skb, tmp)				\
		for (tmp = skb->next;						\
		     skb != (struct sk_buff *)(queue);				\
		     skb = tmp, tmp = skb->next)

1802 1803
#define skb_queue_reverse_walk(queue, skb) \
		for (skb = (queue)->prev;					\
1804
		     (skb != (struct sk_buff *)(queue));			\
1805 1806
		     skb = skb->prev)

1807 1808 1809 1810 1811 1812 1813 1814 1815
#define skb_queue_reverse_walk_safe(queue, skb, tmp)				\
		for (skb = (queue)->prev, tmp = skb->prev;			\
		     skb != (struct sk_buff *)(queue);				\
		     skb = tmp, tmp = skb->prev)

#define skb_queue_reverse_walk_from_safe(queue, skb, tmp)			\
		for (tmp = skb->prev;						\
		     skb != (struct sk_buff *)(queue);				\
		     skb = tmp, tmp = skb->prev)
L
Linus Torvalds 已提交
1816

1817
static inline bool skb_has_frag_list(const struct sk_buff *skb)
1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835
{
	return skb_shinfo(skb)->frag_list != NULL;
}

static inline void skb_frag_list_init(struct sk_buff *skb)
{
	skb_shinfo(skb)->frag_list = NULL;
}

static inline void skb_frag_add_head(struct sk_buff *skb, struct sk_buff *frag)
{
	frag->next = skb_shinfo(skb)->frag_list;
	skb_shinfo(skb)->frag_list = frag;
}

#define skb_walk_frags(skb, iter)	\
	for (iter = skb_shinfo(skb)->frag_list; iter; iter = iter->next)

1836 1837
extern struct sk_buff *__skb_recv_datagram(struct sock *sk, unsigned flags,
					   int *peeked, int *err);
L
Linus Torvalds 已提交
1838 1839 1840 1841 1842 1843 1844
extern struct sk_buff *skb_recv_datagram(struct sock *sk, unsigned flags,
					 int noblock, int *err);
extern unsigned int    datagram_poll(struct file *file, struct socket *sock,
				     struct poll_table_struct *wait);
extern int	       skb_copy_datagram_iovec(const struct sk_buff *from,
					       int offset, struct iovec *to,
					       int size);
1845
extern int	       skb_copy_and_csum_datagram_iovec(struct sk_buff *skb,
L
Linus Torvalds 已提交
1846 1847
							int hlen,
							struct iovec *iov);
1848 1849
extern int	       skb_copy_datagram_from_iovec(struct sk_buff *skb,
						    int offset,
1850 1851
						    const struct iovec *from,
						    int from_offset,
1852
						    int len);
1853 1854 1855 1856 1857
extern int	       skb_copy_datagram_const_iovec(const struct sk_buff *from,
						     int offset,
						     const struct iovec *to,
						     int to_offset,
						     int size);
L
Linus Torvalds 已提交
1858
extern void	       skb_free_datagram(struct sock *sk, struct sk_buff *skb);
1859 1860
extern void	       skb_free_datagram_locked(struct sock *sk,
						struct sk_buff *skb);
1861
extern int	       skb_kill_datagram(struct sock *sk, struct sk_buff *skb,
1862
					 unsigned int flags);
1863 1864
extern __wsum	       skb_checksum(const struct sk_buff *skb, int offset,
				    int len, __wsum csum);
L
Linus Torvalds 已提交
1865 1866
extern int	       skb_copy_bits(const struct sk_buff *skb, int offset,
				     void *to, int len);
1867 1868
extern int	       skb_store_bits(struct sk_buff *skb, int offset,
				      const void *from, int len);
1869
extern __wsum	       skb_copy_and_csum_bits(const struct sk_buff *skb,
L
Linus Torvalds 已提交
1870
					      int offset, u8 *to, int len,
1871
					      __wsum csum);
J
Jens Axboe 已提交
1872 1873 1874 1875 1876
extern int             skb_splice_bits(struct sk_buff *skb,
						unsigned int offset,
						struct pipe_inode_info *pipe,
						unsigned int len,
						unsigned int flags);
L
Linus Torvalds 已提交
1877 1878 1879
extern void	       skb_copy_and_csum_dev(const struct sk_buff *skb, u8 *to);
extern void	       skb_split(struct sk_buff *skb,
				 struct sk_buff *skb1, const u32 len);
1880 1881
extern int	       skb_shift(struct sk_buff *tgt, struct sk_buff *skb,
				 int shiftlen);
L
Linus Torvalds 已提交
1882

1883
extern struct sk_buff *skb_segment(struct sk_buff *skb, u32 features);
1884

L
Linus Torvalds 已提交
1885 1886 1887 1888 1889
static inline void *skb_header_pointer(const struct sk_buff *skb, int offset,
				       int len, void *buffer)
{
	int hlen = skb_headlen(skb);

1890
	if (hlen - offset >= len)
L
Linus Torvalds 已提交
1891 1892 1893 1894 1895 1896 1897 1898
		return skb->data + offset;

	if (skb_copy_bits(skb, offset, buffer, len) < 0)
		return NULL;

	return buffer;
}

1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912
static inline void skb_copy_from_linear_data(const struct sk_buff *skb,
					     void *to,
					     const unsigned int len)
{
	memcpy(to, skb->data, len);
}

static inline void skb_copy_from_linear_data_offset(const struct sk_buff *skb,
						    const int offset, void *to,
						    const unsigned int len)
{
	memcpy(to, skb->data + offset, len);
}

1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927
static inline void skb_copy_to_linear_data(struct sk_buff *skb,
					   const void *from,
					   const unsigned int len)
{
	memcpy(skb->data, from, len);
}

static inline void skb_copy_to_linear_data_offset(struct sk_buff *skb,
						  const int offset,
						  const void *from,
						  const unsigned int len)
{
	memcpy(skb->data + offset, from, len);
}

L
Linus Torvalds 已提交
1928 1929
extern void skb_init(void);

1930 1931 1932 1933 1934
static inline ktime_t skb_get_ktime(const struct sk_buff *skb)
{
	return skb->tstamp;
}

1935 1936 1937 1938 1939 1940 1941 1942 1943
/**
 *	skb_get_timestamp - get timestamp from a skb
 *	@skb: skb to get stamp from
 *	@stamp: pointer to struct timeval to store stamp in
 *
 *	Timestamps are stored in the skb as offsets to a base timestamp.
 *	This function converts the offset back to a struct timeval and stores
 *	it in stamp.
 */
1944 1945
static inline void skb_get_timestamp(const struct sk_buff *skb,
				     struct timeval *stamp)
1946
{
1947
	*stamp = ktime_to_timeval(skb->tstamp);
1948 1949
}

1950 1951 1952 1953 1954 1955
static inline void skb_get_timestampns(const struct sk_buff *skb,
				       struct timespec *stamp)
{
	*stamp = ktime_to_timespec(skb->tstamp);
}

1956
static inline void __net_timestamp(struct sk_buff *skb)
1957
{
1958
	skb->tstamp = ktime_get_real();
1959 1960
}

1961 1962 1963 1964 1965
static inline ktime_t net_timedelta(ktime_t t)
{
	return ktime_sub(ktime_get_real(), t);
}

1966 1967 1968 1969
static inline ktime_t net_invalid_timestamp(void)
{
	return ktime_set(0, 0);
}
1970

1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000
extern void skb_timestamping_init(void);

#ifdef CONFIG_NETWORK_PHY_TIMESTAMPING

extern void skb_clone_tx_timestamp(struct sk_buff *skb);
extern bool skb_defer_rx_timestamp(struct sk_buff *skb);

#else /* CONFIG_NETWORK_PHY_TIMESTAMPING */

static inline void skb_clone_tx_timestamp(struct sk_buff *skb)
{
}

static inline bool skb_defer_rx_timestamp(struct sk_buff *skb)
{
	return false;
}

#endif /* !CONFIG_NETWORK_PHY_TIMESTAMPING */

/**
 * skb_complete_tx_timestamp() - deliver cloned skb with tx timestamps
 *
 * @skb: clone of the the original outgoing packet
 * @hwtstamps: hardware time stamps
 *
 */
void skb_complete_tx_timestamp(struct sk_buff *skb,
			       struct skb_shared_hwtstamps *hwtstamps);

2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014
/**
 * skb_tstamp_tx - queue clone of skb with send time stamps
 * @orig_skb:	the original outgoing packet
 * @hwtstamps:	hardware time stamps, may be NULL if not available
 *
 * If the skb has a socket associated, then this function clones the
 * skb (thus sharing the actual data and optional structures), stores
 * the optional hardware time stamping information (if non NULL) or
 * generates a software time stamp (otherwise), then queues the clone
 * to the error queue of the socket.  Errors are silently ignored.
 */
extern void skb_tstamp_tx(struct sk_buff *orig_skb,
			struct skb_shared_hwtstamps *hwtstamps);

2015 2016
static inline void sw_tx_timestamp(struct sk_buff *skb)
{
2017 2018
	if (skb_shinfo(skb)->tx_flags & SKBTX_SW_TSTAMP &&
	    !(skb_shinfo(skb)->tx_flags & SKBTX_IN_PROGRESS))
2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032
		skb_tstamp_tx(skb, NULL);
}

/**
 * skb_tx_timestamp() - Driver hook for transmit timestamping
 *
 * Ethernet MAC Drivers should call this function in their hard_xmit()
 * function as soon as possible after giving the sk_buff to the MAC
 * hardware, but before freeing the sk_buff.
 *
 * @skb: A socket buffer.
 */
static inline void skb_tx_timestamp(struct sk_buff *skb)
{
2033
	skb_clone_tx_timestamp(skb);
2034 2035 2036
	sw_tx_timestamp(skb);
}

2037
extern __sum16 __skb_checksum_complete_head(struct sk_buff *skb, int len);
2038
extern __sum16 __skb_checksum_complete(struct sk_buff *skb);
2039

2040 2041 2042 2043 2044
static inline int skb_csum_unnecessary(const struct sk_buff *skb)
{
	return skb->ip_summed & CHECKSUM_UNNECESSARY;
}

2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060
/**
 *	skb_checksum_complete - Calculate checksum of an entire packet
 *	@skb: packet to process
 *
 *	This function calculates the checksum over the entire packet plus
 *	the value of skb->csum.  The latter can be used to supply the
 *	checksum of a pseudo header as used by TCP/UDP.  It returns the
 *	checksum.
 *
 *	For protocols that contain complete checksums such as ICMP/TCP/UDP,
 *	this function can be used to verify that checksum on received
 *	packets.  In that case the function should return zero if the
 *	checksum is correct.  In particular, this function will return zero
 *	if skb->ip_summed is CHECKSUM_UNNECESSARY which indicates that the
 *	hardware has already verified the correctness of the checksum.
 */
2061
static inline __sum16 skb_checksum_complete(struct sk_buff *skb)
2062
{
2063 2064
	return skb_csum_unnecessary(skb) ?
	       0 : __skb_checksum_complete(skb);
2065 2066
}

2067
#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE)
2068
extern void nf_conntrack_destroy(struct nf_conntrack *nfct);
L
Linus Torvalds 已提交
2069 2070 2071
static inline void nf_conntrack_put(struct nf_conntrack *nfct)
{
	if (nfct && atomic_dec_and_test(&nfct->use))
2072
		nf_conntrack_destroy(nfct);
L
Linus Torvalds 已提交
2073 2074 2075 2076 2077 2078
}
static inline void nf_conntrack_get(struct nf_conntrack *nfct)
{
	if (nfct)
		atomic_inc(&nfct->use);
}
2079 2080
#endif
#ifdef NET_SKBUFF_NF_DEFRAG_NEEDED
2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091
static inline void nf_conntrack_get_reasm(struct sk_buff *skb)
{
	if (skb)
		atomic_inc(&skb->users);
}
static inline void nf_conntrack_put_reasm(struct sk_buff *skb)
{
	if (skb)
		kfree_skb(skb);
}
#endif
L
Linus Torvalds 已提交
2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103
#ifdef CONFIG_BRIDGE_NETFILTER
static inline void nf_bridge_put(struct nf_bridge_info *nf_bridge)
{
	if (nf_bridge && atomic_dec_and_test(&nf_bridge->use))
		kfree(nf_bridge);
}
static inline void nf_bridge_get(struct nf_bridge_info *nf_bridge)
{
	if (nf_bridge)
		atomic_inc(&nf_bridge->use);
}
#endif /* CONFIG_BRIDGE_NETFILTER */
2104 2105
static inline void nf_reset(struct sk_buff *skb)
{
2106
#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE)
2107 2108
	nf_conntrack_put(skb->nfct);
	skb->nfct = NULL;
2109 2110
#endif
#ifdef NET_SKBUFF_NF_DEFRAG_NEEDED
2111 2112 2113 2114 2115 2116 2117 2118 2119
	nf_conntrack_put_reasm(skb->nfct_reasm);
	skb->nfct_reasm = NULL;
#endif
#ifdef CONFIG_BRIDGE_NETFILTER
	nf_bridge_put(skb->nf_bridge);
	skb->nf_bridge = NULL;
#endif
}

2120 2121 2122
/* Note: This doesn't put any conntrack and bridge info in dst. */
static inline void __nf_copy(struct sk_buff *dst, const struct sk_buff *src)
{
2123
#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE)
2124 2125 2126
	dst->nfct = src->nfct;
	nf_conntrack_get(src->nfct);
	dst->nfctinfo = src->nfctinfo;
2127 2128
#endif
#ifdef NET_SKBUFF_NF_DEFRAG_NEEDED
2129 2130 2131 2132 2133 2134 2135 2136 2137
	dst->nfct_reasm = src->nfct_reasm;
	nf_conntrack_get_reasm(src->nfct_reasm);
#endif
#ifdef CONFIG_BRIDGE_NETFILTER
	dst->nf_bridge  = src->nf_bridge;
	nf_bridge_get(src->nf_bridge);
#endif
}

2138 2139 2140
static inline void nf_copy(struct sk_buff *dst, const struct sk_buff *src)
{
#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE)
2141
	nf_conntrack_put(dst->nfct);
2142 2143
#endif
#ifdef NET_SKBUFF_NF_DEFRAG_NEEDED
2144 2145 2146 2147 2148 2149 2150 2151
	nf_conntrack_put_reasm(dst->nfct_reasm);
#endif
#ifdef CONFIG_BRIDGE_NETFILTER
	nf_bridge_put(dst->nf_bridge);
#endif
	__nf_copy(dst, src);
}

2152 2153 2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169
#ifdef CONFIG_NETWORK_SECMARK
static inline void skb_copy_secmark(struct sk_buff *to, const struct sk_buff *from)
{
	to->secmark = from->secmark;
}

static inline void skb_init_secmark(struct sk_buff *skb)
{
	skb->secmark = 0;
}
#else
static inline void skb_copy_secmark(struct sk_buff *to, const struct sk_buff *from)
{ }

static inline void skb_init_secmark(struct sk_buff *skb)
{ }
#endif

2170 2171 2172 2173 2174
static inline void skb_set_queue_mapping(struct sk_buff *skb, u16 queue_mapping)
{
	skb->queue_mapping = queue_mapping;
}

2175
static inline u16 skb_get_queue_mapping(const struct sk_buff *skb)
2176 2177 2178 2179
{
	return skb->queue_mapping;
}

2180 2181 2182 2183 2184
static inline void skb_copy_queue_mapping(struct sk_buff *to, const struct sk_buff *from)
{
	to->queue_mapping = from->queue_mapping;
}

2185 2186 2187 2188 2189
static inline void skb_record_rx_queue(struct sk_buff *skb, u16 rx_queue)
{
	skb->queue_mapping = rx_queue + 1;
}

2190
static inline u16 skb_get_rx_queue(const struct sk_buff *skb)
2191 2192 2193 2194
{
	return skb->queue_mapping - 1;
}

2195
static inline bool skb_rx_queue_recorded(const struct sk_buff *skb)
2196
{
E
Eric Dumazet 已提交
2197
	return skb->queue_mapping != 0;
2198 2199
}

2200 2201 2202
extern u16 __skb_tx_hash(const struct net_device *dev,
			 const struct sk_buff *skb,
			 unsigned int num_tx_queues);
2203

2204 2205 2206 2207 2208 2209 2210 2211 2212 2213 2214 2215
#ifdef CONFIG_XFRM
static inline struct sec_path *skb_sec_path(struct sk_buff *skb)
{
	return skb->sp;
}
#else
static inline struct sec_path *skb_sec_path(struct sk_buff *skb)
{
	return NULL;
}
#endif

H
Herbert Xu 已提交
2216 2217 2218 2219 2220
static inline int skb_is_gso(const struct sk_buff *skb)
{
	return skb_shinfo(skb)->gso_size;
}

B
Brice Goglin 已提交
2221 2222 2223 2224 2225
static inline int skb_is_gso_v6(const struct sk_buff *skb)
{
	return skb_shinfo(skb)->gso_type & SKB_GSO_TCPV6;
}

2226 2227 2228 2229 2230 2231 2232
extern void __skb_warn_lro_forwarding(const struct sk_buff *skb);

static inline bool skb_warn_if_lro(const struct sk_buff *skb)
{
	/* LRO sets gso_size but not gso_type, whereas if GSO is really
	 * wanted then gso_type will be set. */
	struct skb_shared_info *shinfo = skb_shinfo(skb);
2233 2234
	if (skb_is_nonlinear(skb) && shinfo->gso_size != 0 &&
	    unlikely(shinfo->gso_type == 0)) {
2235 2236 2237 2238 2239 2240
		__skb_warn_lro_forwarding(skb);
		return true;
	}
	return false;
}

2241 2242 2243 2244 2245 2246 2247
static inline void skb_forward_csum(struct sk_buff *skb)
{
	/* Unfortunately we don't support this one.  Any brave souls? */
	if (skb->ip_summed == CHECKSUM_COMPLETE)
		skb->ip_summed = CHECKSUM_NONE;
}

2248 2249 2250 2251 2252 2253 2254 2255 2256 2257 2258 2259 2260 2261 2262
/**
 * skb_checksum_none_assert - make sure skb ip_summed is CHECKSUM_NONE
 * @skb: skb to check
 *
 * fresh skbs have their ip_summed set to CHECKSUM_NONE.
 * Instead of forcing ip_summed to CHECKSUM_NONE, we can
 * use this helper, to document places where we make this assertion.
 */
static inline void skb_checksum_none_assert(struct sk_buff *skb)
{
#ifdef DEBUG
	BUG_ON(skb->ip_summed != CHECKSUM_NONE);
#endif
}

2263
bool skb_partial_csum_set(struct sk_buff *skb, u16 start, u16 off);
L
Linus Torvalds 已提交
2264 2265
#endif	/* __KERNEL__ */
#endif	/* _LINUX_SKBUFF_H */