skbuff.h 77.9 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
#include <linux/compiler.h>
#include <linux/time.h>
21
#include <linux/bug.h>
L
Linus Torvalds 已提交
22 23
#include <linux/cache.h>

A
Arun Sharma 已提交
24
#include <linux/atomic.h>
L
Linus Torvalds 已提交
25 26 27
#include <asm/types.h>
#include <linux/spinlock.h>
#include <linux/net.h>
28
#include <linux/textsearch.h>
L
Linus Torvalds 已提交
29
#include <net/checksum.h>
30
#include <linux/rcupdate.h>
31
#include <linux/dmaengine.h>
32
#include <linux/hrtimer.h>
33
#include <linux/dma-mapping.h>
34
#include <linux/netdev_features.h>
L
Linus Torvalds 已提交
35

36
/* Don't change this without changing skb_csum_unnecessary! */
L
Linus Torvalds 已提交
37
#define CHECKSUM_NONE 0
38 39 40
#define CHECKSUM_UNNECESSARY 1
#define CHECKSUM_COMPLETE 2
#define CHECKSUM_PARTIAL 3
L
Linus Torvalds 已提交
41 42 43

#define SKB_DATA_ALIGN(X)	(((X) + (SMP_CACHE_BYTES - 1)) & \
				 ~(SMP_CACHE_BYTES - 1))
44
#define SKB_WITH_OVERHEAD(X)	\
45
	((X) - SKB_DATA_ALIGN(sizeof(struct skb_shared_info)))
46 47
#define SKB_MAX_ORDER(X, ORDER) \
	SKB_WITH_OVERHEAD((PAGE_SIZE << (ORDER)) - (X))
L
Linus Torvalds 已提交
48 49 50
#define SKB_MAX_HEAD(X)		(SKB_MAX_ORDER((X), 0))
#define SKB_MAX_ALLOC		(SKB_MAX_ORDER(0, 2))

E
Eric Dumazet 已提交
51 52 53 54 55
/* return minimum truesize of one skb containing X bytes of data */
#define SKB_TRUESIZE(X) ((X) +						\
			 SKB_DATA_ALIGN(sizeof(struct sk_buff)) +	\
			 SKB_DATA_ALIGN(sizeof(struct skb_shared_info)))

L
Linus Torvalds 已提交
56 57 58 59 60 61 62 63 64 65 66
/* 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)
 *
67
 *	COMPLETE: the most generic way. Device supplied checksum of _all_
L
Linus Torvalds 已提交
68 69
 *	    the packet as seen by netif_rx in skb->csum.
 *	    NOTE: Even if device supports only some protocols, but
70
 *	    is able to produce some skb->csum, it MUST use COMPLETE,
L
Linus Torvalds 已提交
71 72
 *	    not UNNECESSARY.
 *
73 74 75 76 77 78 79
 *	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
80 81 82 83
 * B. Checksumming on output.
 *
 *	NONE: skb is checksummed by protocol or csum is not required.
 *
84
 *	PARTIAL: device is required to csum packet as seen by hard_start_xmit
85 86
 *	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
87 88 89 90 91 92 93 94 95
 *
 *	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_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)
96
 *	NETIF_F_IPV6_CSUM about as dumb as the last one but does IPv6 instead.
L
Linus Torvalds 已提交
97
 *
98 99 100 101 102 103 104
 *	UNNECESSARY: device will do per protocol specific csum. Protocol drivers
 *	that do not want net to perform the checksum calculation should use
 *	this flag in their outgoing skbs.
 *	NETIF_F_FCOE_CRC  this indicates the device can do FCoE FC CRC
 *			  offload. Correspondingly, the FCoE protocol driver
 *			  stack should use CHECKSUM_UNNECESSARY.
 *
L
Linus Torvalds 已提交
105 106 107 108
 *	Any questions? No questions, good. 		--ANK
 */

struct net_device;
109
struct scatterlist;
J
Jens Axboe 已提交
110
struct pipe_inode_info;
L
Linus Torvalds 已提交
111

112
#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE)
L
Linus Torvalds 已提交
113 114 115
struct nf_conntrack {
	atomic_t use;
};
116
#endif
L
Linus Torvalds 已提交
117 118 119

#ifdef CONFIG_BRIDGE_NETFILTER
struct nf_bridge_info {
120 121 122 123 124
	atomic_t		use;
	unsigned int		mask;
	struct net_device	*physindev;
	struct net_device	*physoutdev;
	unsigned long		data[32 / sizeof(unsigned long)];
L
Linus Torvalds 已提交
125 126 127 128 129 130 131 132 133 134 135 136 137 138
};
#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;

139 140 141 142 143 144
/* To allow 64K frame to be packed as single skb without frag_list we
 * require 64K/PAGE_SIZE pages plus 1 additional page to allow for
 * buffers which do not start on a page boundary.
 *
 * Since GRO uses frags we allocate at least 16 regardless of page
 * size.
145
 */
146
#if (65536/PAGE_SIZE + 1) < 16
147
#define MAX_SKB_FRAGS 16UL
148
#else
149
#define MAX_SKB_FRAGS (65536/PAGE_SIZE + 1)
150
#endif
L
Linus Torvalds 已提交
151 152 153 154

typedef struct skb_frag_struct skb_frag_t;

struct skb_frag_struct {
155 156 157
	struct {
		struct page *p;
	} page;
158
#if (BITS_PER_LONG > 32) || (PAGE_SIZE >= 65536)
159 160
	__u32 page_offset;
	__u32 size;
161 162 163 164
#else
	__u16 page_offset;
	__u16 size;
#endif
L
Linus Torvalds 已提交
165 166
};

E
Eric Dumazet 已提交
167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186
static inline unsigned int skb_frag_size(const skb_frag_t *frag)
{
	return frag->size;
}

static inline void skb_frag_size_set(skb_frag_t *frag, unsigned int size)
{
	frag->size = size;
}

static inline void skb_frag_size_add(skb_frag_t *frag, int delta)
{
	frag->size += delta;
}

static inline void skb_frag_size_sub(skb_frag_t *frag, int delta)
{
	frag->size -= delta;
}

187 188 189
#define HAVE_HW_TIME_STAMP

/**
190
 * struct skb_shared_hwtstamps - hardware time stamps
191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216
 * @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;
};

217 218 219 220 221 222 223 224 225 226 227
/* 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,

228
	/* device driver supports TX zero-copy buffers */
E
Eric Dumazet 已提交
229
	SKBTX_DEV_ZEROCOPY = 1 << 3,
230 231

	/* generate wifi status information (where possible) */
E
Eric Dumazet 已提交
232
	SKBTX_WIFI_STATUS = 1 << 4,
233 234 235 236 237 238 239

	/* This indicates at least one fragment might be overwritten
	 * (as in vmsplice(), sendfile() ...)
	 * If we need to compute a TX checksum, we'll need to copy
	 * all frags to avoid possible bad checksum
	 */
	SKBTX_SHARED_FRAG = 1 << 5,
240 241 242 243 244
};

/*
 * The callback notifies userspace to release buffers when skb DMA is done in
 * lower device, the skb last reference should be 0 when calling this.
245 246
 * The zerocopy_success argument is true if zero copy transmit occurred,
 * false on data copy or out of memory error caused by data copy attempt.
247 248
 * The ctx field is used to track device context.
 * The desc field is used to track userspace buffer index.
249 250
 */
struct ubuf_info {
251
	void (*callback)(struct ubuf_info *, bool zerocopy_success);
252
	void *ctx;
253
	unsigned long desc;
254 255
};

L
Linus Torvalds 已提交
256 257 258 259
/* This data is invariant across clones and lives at
 * the end of the header data, ie. at skb->end.
 */
struct skb_shared_info {
260 261
	unsigned char	nr_frags;
	__u8		tx_flags;
262 263 264 265
	unsigned short	gso_size;
	/* Warning: this field is not always filled in (UFO)! */
	unsigned short	gso_segs;
	unsigned short  gso_type;
L
Linus Torvalds 已提交
266
	struct sk_buff	*frag_list;
267
	struct skb_shared_hwtstamps hwtstamps;
268
	__be32          ip6_frag_id;
E
Eric Dumazet 已提交
269 270 271 272 273 274

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

J
Johann Baudy 已提交
275 276 277
	/* Intermediate layers must ensure that destructor_arg
	 * remains valid until skb destructor */
	void *		destructor_arg;
278

279 280
	/* must be last field, see pskb_expand_head() */
	skb_frag_t	frags[MAX_SKB_FRAGS];
L
Linus Torvalds 已提交
281 282 283 284
};

/* 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
285 286
 * 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
287 288 289 290 291 292 293 294 295 296
 *
 * 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)

297 298 299 300 301 302 303

enum {
	SKB_FCLONE_UNAVAILABLE,
	SKB_FCLONE_ORIG,
	SKB_FCLONE_CLONE,
};

304 305
enum {
	SKB_GSO_TCPV4 = 1 << 0,
H
Herbert Xu 已提交
306
	SKB_GSO_UDP = 1 << 1,
307 308 309

	/* This indicates the skb is from an untrusted source. */
	SKB_GSO_DODGY = 1 << 2,
M
Michael Chan 已提交
310 311

	/* This indicates the tcp segment has CWR set. */
H
Herbert Xu 已提交
312 313 314
	SKB_GSO_TCP_ECN = 1 << 3,

	SKB_GSO_TCPV6 = 1 << 4,
315 316

	SKB_GSO_FCOE = 1 << 5,
317 318

	SKB_GSO_GRE = 1 << 6,
319 320
};

321 322 323 324 325 326 327 328 329 330
#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

331 332 333 334 335
#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
336 337 338 339
/** 
 *	struct sk_buff - socket buffer
 *	@next: Next buffer in list
 *	@prev: Previous buffer in list
H
Herbert Xu 已提交
340
 *	@tstamp: Time we arrived
341
 *	@sk: Socket we are owned by
L
Linus Torvalds 已提交
342
 *	@dev: Device we arrived on/are leaving by
343
 *	@cb: Control buffer. Free for use by every layer. Put private vars here
E
Eric Dumazet 已提交
344
 *	@_skb_refdst: destination entry (with norefcount bit)
345
 *	@sp: the security path, used for xfrm
L
Linus Torvalds 已提交
346 347 348
 *	@len: Length of actual data
 *	@data_len: Data length
 *	@mac_len: Length of link layer header
349
 *	@hdr_len: writable header length of cloned skb
350 351 352
 *	@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
353
 *	@priority: Packet queueing priority
354
 *	@local_df: allow local fragmentation
L
Linus Torvalds 已提交
355
 *	@cloned: Head may be cloned (check refcnt to be sure)
356
 *	@ip_summed: Driver fed us an IP checksum
L
Linus Torvalds 已提交
357
 *	@nohdr: Payload reference only, must not modify header
358
 *	@nfctinfo: Relationship of this skb to the connection
L
Linus Torvalds 已提交
359
 *	@pkt_type: Packet class
360 361
 *	@fclone: skbuff clone status
 *	@ipvs_property: skbuff is owned by ipvs
362 363
 *	@peeked: this packet has been seen already, so stats have been
 *		done for it, don't do them again
364
 *	@nf_trace: netfilter packet trace flag
365 366 367
 *	@protocol: Packet protocol from driver
 *	@destructor: Destruct function
 *	@nfct: Associated connection, if any
R
Randy Dunlap 已提交
368
 *	@nfct_reasm: netfilter conntrack re-assembly pointer
L
Linus Torvalds 已提交
369
 *	@nf_bridge: Saved data about a bridged frame - see br_netfilter.c
370
 *	@skb_iif: ifindex of device we arrived on
L
Linus Torvalds 已提交
371 372
 *	@tc_index: Traffic control index
 *	@tc_verd: traffic control verdict
373 374
 *	@rxhash: the packet hash computed on receive
 *	@queue_mapping: Queue mapping for multiqueue devices
375
 *	@ndisc_nodetype: router type (from link layer)
376
 *	@ooo_okay: allow the mapping of a socket to a queue to be changed
377 378
 *	@l4_rxhash: indicate rxhash is a canonical 4-tuple hash over transport
 *		ports.
379 380
 *	@wifi_acked_valid: wifi_acked was set
 *	@wifi_acked: whether frame was acked on wifi or not
381
 *	@no_fcs:  Request NIC to treat last 4 bytes as Ethernet FCS
R
Randy Dunlap 已提交
382 383
 *	@dma_cookie: a cookie to one of several possible DMA operations
 *		done by skb DMA functions
384
 *	@secmark: security marking
385 386
 *	@mark: Generic packet mark
 *	@dropcount: total number of sk_receive_queue overflows
387
 *	@vlan_tci: vlan tag control information
388 389
 *	@inner_transport_header: Inner transport layer header (encapsulation)
 *	@inner_network_header: Network layer header (encapsulation)
390 391 392 393 394 395 396 397 398
 *	@transport_header: Transport layer header
 *	@network_header: Network layer header
 *	@mac_header: Link layer header
 *	@tail: Tail pointer
 *	@end: End pointer
 *	@head: Head of buffer
 *	@data: Data head pointer
 *	@truesize: Buffer size
 *	@users: User count - see {datagram,tcp}.c
L
Linus Torvalds 已提交
399 400 401 402 403 404 405
 */

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

406
	ktime_t			tstamp;
407 408

	struct sock		*sk;
L
Linus Torvalds 已提交
409 410 411 412 413 414 415 416
	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.
	 */
417
	char			cb[48] __aligned(8);
L
Linus Torvalds 已提交
418

E
Eric Dumazet 已提交
419
	unsigned long		_skb_refdst;
420 421 422
#ifdef CONFIG_XFRM
	struct	sec_path	*sp;
#endif
L
Linus Torvalds 已提交
423
	unsigned int		len,
424 425 426
				data_len;
	__u16			mac_len,
				hdr_len;
A
Al Viro 已提交
427 428
	union {
		__wsum		csum;
429 430 431 432
		struct {
			__u16	csum_start;
			__u16	csum_offset;
		};
A
Al Viro 已提交
433
	};
L
Linus Torvalds 已提交
434
	__u32			priority;
435
	kmemcheck_bitfield_begin(flags1);
436 437 438
	__u8			local_df:1,
				cloned:1,
				ip_summed:2,
439 440
				nohdr:1,
				nfctinfo:3;
441
	__u8			pkt_type:3,
442
				fclone:2,
443
				ipvs_property:1,
444
				peeked:1,
445
				nf_trace:1;
446
	kmemcheck_bitfield_end(flags1);
E
Eric Dumazet 已提交
447
	__be16			protocol;
L
Linus Torvalds 已提交
448 449

	void			(*destructor)(struct sk_buff *skb);
450
#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE)
451
	struct nf_conntrack	*nfct;
452 453
#endif
#ifdef NET_SKBUFF_NF_DEFRAG_NEEDED
454 455
	struct sk_buff		*nfct_reasm;
#endif
L
Linus Torvalds 已提交
456 457 458
#ifdef CONFIG_BRIDGE_NETFILTER
	struct nf_bridge_info	*nf_bridge;
#endif
459

460
	int			skb_iif;
461 462 463 464 465

	__u32			rxhash;

	__u16			vlan_tci;

L
Linus Torvalds 已提交
466
#ifdef CONFIG_NET_SCHED
467
	__u16			tc_index;	/* traffic control index */
L
Linus Torvalds 已提交
468
#ifdef CONFIG_NET_CLS_ACT
469
	__u16			tc_verd;	/* traffic control verdict */
L
Linus Torvalds 已提交
470 471
#endif
#endif
472

473
	__u16			queue_mapping;
474
	kmemcheck_bitfield_begin(flags2);
475
#ifdef CONFIG_IPV6_NDISC_NODETYPE
476
	__u8			ndisc_nodetype:2;
477
#endif
478
	__u8			pfmemalloc:1;
479
	__u8			ooo_okay:1;
480
	__u8			l4_rxhash:1;
481 482
	__u8			wifi_acked_valid:1;
	__u8			wifi_acked:1;
483
	__u8			no_fcs:1;
484
	__u8			head_frag:1;
485 486 487 488 489 490 491
	/* Encapsulation protocol and NIC drivers should use
	 * this flag to indicate to each other if the skb contains
	 * encapsulated packet or not and maybe use the inner packet
	 * headers if needed
	 */
	__u8			encapsulation:1;
	/* 7/9 bit hole (depending on ndisc_nodetype presence) */
492 493
	kmemcheck_bitfield_end(flags2);

494 495 496
#ifdef CONFIG_NET_DMA
	dma_cookie_t		dma_cookie;
#endif
497 498 499
#ifdef CONFIG_NETWORK_SECMARK
	__u32			secmark;
#endif
500 501 502
	union {
		__u32		mark;
		__u32		dropcount;
E
Eric Dumazet 已提交
503
		__u32		reserved_tailroom;
504
	};
L
Linus Torvalds 已提交
505

506 507
	sk_buff_data_t		inner_transport_header;
	sk_buff_data_t		inner_network_header;
508 509 510
	sk_buff_data_t		transport_header;
	sk_buff_data_t		network_header;
	sk_buff_data_t		mac_header;
L
Linus Torvalds 已提交
511
	/* These elements must be at the end, see alloc_skb() for details.  */
512
	sk_buff_data_t		tail;
513
	sk_buff_data_t		end;
L
Linus Torvalds 已提交
514
	unsigned char		*head,
515
				*data;
516 517
	unsigned int		truesize;
	atomic_t		users;
L
Linus Torvalds 已提交
518 519 520 521 522 523 524 525 526
};

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


527 528 529 530 531 532 533 534 535
#define SKB_ALLOC_FCLONE	0x01
#define SKB_ALLOC_RX		0x02

/* Returns true if the skb was allocated from PFMEMALLOC reserves */
static inline bool skb_pfmemalloc(const struct sk_buff *skb)
{
	return unlikely(skb->pfmemalloc);
}

E
Eric Dumazet 已提交
536 537 538 539 540 541 542 543 544 545 546 547 548
/*
 * 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
549 550
static inline struct dst_entry *skb_dst(const struct sk_buff *skb)
{
E
Eric Dumazet 已提交
551 552 553 554 555 556 557
	/* 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
558 559
}

E
Eric Dumazet 已提交
560 561 562 563 564 565 566 567
/**
 * 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
568 569
static inline void skb_dst_set(struct sk_buff *skb, struct dst_entry *dst)
{
E
Eric Dumazet 已提交
570 571 572
	skb->_skb_refdst = (unsigned long)dst;
}

E
Eric Dumazet 已提交
573
extern void skb_dst_set_noref(struct sk_buff *skb, struct dst_entry *dst);
E
Eric Dumazet 已提交
574 575

/**
L
Lucas De Marchi 已提交
576
 * skb_dst_is_noref - Test if skb dst isn't refcounted
E
Eric Dumazet 已提交
577 578 579 580 581
 * @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
582 583
}

E
Eric Dumazet 已提交
584 585
static inline struct rtable *skb_rtable(const struct sk_buff *skb)
{
E
Eric Dumazet 已提交
586
	return (struct rtable *)skb_dst(skb);
E
Eric Dumazet 已提交
587 588
}

589
extern void kfree_skb(struct sk_buff *skb);
590
extern void skb_tx_error(struct sk_buff *skb);
591
extern void consume_skb(struct sk_buff *skb);
L
Linus Torvalds 已提交
592
extern void	       __kfree_skb(struct sk_buff *skb);
593
extern struct kmem_cache *skbuff_head_cache;
E
Eric Dumazet 已提交
594 595 596 597 598

extern void kfree_skb_partial(struct sk_buff *skb, bool head_stolen);
extern bool skb_try_coalesce(struct sk_buff *to, struct sk_buff *from,
			     bool *fragstolen, int *delta_truesize);

599
extern struct sk_buff *__alloc_skb(unsigned int size,
600
				   gfp_t priority, int flags, int node);
601
extern struct sk_buff *build_skb(void *data, unsigned int frag_size);
602
static inline struct sk_buff *alloc_skb(unsigned int size,
A
Al Viro 已提交
603
					gfp_t priority)
604
{
E
Eric Dumazet 已提交
605
	return __alloc_skb(size, priority, 0, NUMA_NO_NODE);
606 607 608
}

static inline struct sk_buff *alloc_skb_fclone(unsigned int size,
A
Al Viro 已提交
609
					       gfp_t priority)
610
{
611
	return __alloc_skb(size, priority, SKB_ALLOC_FCLONE, NUMA_NO_NODE);
612 613
}

H
Herbert Xu 已提交
614
extern struct sk_buff *skb_morph(struct sk_buff *dst, struct sk_buff *src);
615
extern int skb_copy_ubufs(struct sk_buff *skb, gfp_t gfp_mask);
V
Victor Fusco 已提交
616
extern struct sk_buff *skb_clone(struct sk_buff *skb,
A
Al Viro 已提交
617
				 gfp_t priority);
V
Victor Fusco 已提交
618
extern struct sk_buff *skb_copy(const struct sk_buff *skb,
A
Al Viro 已提交
619
				gfp_t priority);
E
Eric Dumazet 已提交
620 621 622
extern struct sk_buff *__pskb_copy(struct sk_buff *skb,
				 int headroom, gfp_t gfp_mask);

L
Linus Torvalds 已提交
623
extern int	       pskb_expand_head(struct sk_buff *skb,
V
Victor Fusco 已提交
624
					int nhead, int ntail,
A
Al Viro 已提交
625
					gfp_t gfp_mask);
L
Linus Torvalds 已提交
626 627 628 629
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
630
				       gfp_t priority);
631 632 633 634 635
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);
636
extern int	       skb_pad(struct sk_buff *skb, int pad);
637
#define dev_kfree_skb(a)	consume_skb(a)
L
Linus Torvalds 已提交
638

639 640 641 642 643
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
644
struct skb_seq_state {
645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660
	__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);

661 662 663 664
extern unsigned int   skb_find_text(struct sk_buff *skb, unsigned int from,
				    unsigned int to, struct ts_config *config,
				    struct ts_state *state);

665
extern void __skb_get_rxhash(struct sk_buff *skb);
666 667
static inline __u32 skb_get_rxhash(struct sk_buff *skb)
{
668
	if (!skb->l4_rxhash)
669
		__skb_get_rxhash(skb);
670 671 672 673

	return skb->rxhash;
}

674 675 676 677 678
#ifdef NET_SKBUFF_DATA_USES_OFFSET
static inline unsigned char *skb_end_pointer(const struct sk_buff *skb)
{
	return skb->head + skb->end;
}
679 680 681 682 683

static inline unsigned int skb_end_offset(const struct sk_buff *skb)
{
	return skb->end;
}
684 685 686 687 688
#else
static inline unsigned char *skb_end_pointer(const struct sk_buff *skb)
{
	return skb->end;
}
689 690 691 692 693

static inline unsigned int skb_end_offset(const struct sk_buff *skb)
{
	return skb->end - skb->head;
}
694 695
#endif

L
Linus Torvalds 已提交
696
/* Internal */
697
#define skb_shinfo(SKB)	((struct skb_shared_info *)(skb_end_pointer(SKB)))
L
Linus Torvalds 已提交
698

699 700 701 702 703
static inline struct skb_shared_hwtstamps *skb_hwtstamps(struct sk_buff *skb)
{
	return &skb_shinfo(skb)->hwtstamps;
}

L
Linus Torvalds 已提交
704 705 706 707 708 709 710 711 712 713 714
/**
 *	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
715 716 717 718 719 720 721 722 723 724
/**
 *	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
725
	return skb->next == (struct sk_buff *)list;
D
David S. Miller 已提交
726 727
}

728 729 730 731 732 733 734 735 736 737
/**
 *	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
738
	return skb->prev == (struct sk_buff *)list;
739 740
}

D
David S. Miller 已提交
741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758
/**
 *	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;
}

759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776
/**
 *	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
777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808
/**
 *	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;
}

809 810 811 812 813 814 815 816 817 818
static inline int skb_unclone(struct sk_buff *skb, gfp_t pri)
{
	might_sleep_if(pri & __GFP_WAIT);

	if (skb_cloned(skb))
		return pskb_expand_head(skb, 0, 0, pri);

	return 0;
}

L
Linus Torvalds 已提交
819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877
/**
 *	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.
 */
878
static inline struct sk_buff *skb_share_check(struct sk_buff *skb, gfp_t pri)
L
Linus Torvalds 已提交
879 880 881 882
{
	might_sleep_if(pri & __GFP_WAIT);
	if (skb_shared(skb)) {
		struct sk_buff *nskb = skb_clone(skb, pri);
883 884 885 886 887

		if (likely(nskb))
			consume_skb(skb);
		else
			kfree_skb(skb);
L
Linus Torvalds 已提交
888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912
		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.
 */
913
static inline struct sk_buff *skb_unshare(struct sk_buff *skb,
A
Al Viro 已提交
914
					  gfp_t pri)
L
Linus Torvalds 已提交
915 916 917 918 919 920 921 922 923 924 925
{
	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;
}

/**
926
 *	skb_peek - peek at the head of an &sk_buff_head
L
Linus Torvalds 已提交
927 928 929 930 931 932 933 934 935 936 937
 *	@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.
 */
938
static inline struct sk_buff *skb_peek(const struct sk_buff_head *list_)
L
Linus Torvalds 已提交
939
{
940 941 942 943 944
	struct sk_buff *skb = list_->next;

	if (skb == (struct sk_buff *)list_)
		skb = NULL;
	return skb;
L
Linus Torvalds 已提交
945 946
}

P
Pavel Emelyanov 已提交
947 948 949 950 951 952 953 954 955 956 957 958 959
/**
 *	skb_peek_next - peek skb following the given one from a queue
 *	@skb: skb to start from
 *	@list_: list to peek at
 *
 *	Returns %NULL when the end of the list is met or a pointer to the
 *	next element. The reference count is not incremented and the
 *	reference is therefore volatile. Use with caution.
 */
static inline struct sk_buff *skb_peek_next(struct sk_buff *skb,
		const struct sk_buff_head *list_)
{
	struct sk_buff *next = skb->next;
960

P
Pavel Emelyanov 已提交
961 962 963 964 965
	if (next == (struct sk_buff *)list_)
		next = NULL;
	return next;
}

L
Linus Torvalds 已提交
966
/**
967
 *	skb_peek_tail - peek at the tail of an &sk_buff_head
L
Linus Torvalds 已提交
968 969 970 971 972 973 974 975 976 977 978
 *	@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.
 */
979
static inline struct sk_buff *skb_peek_tail(const struct sk_buff_head *list_)
L
Linus Torvalds 已提交
980
{
981 982 983 984 985 986
	struct sk_buff *skb = list_->prev;

	if (skb == (struct sk_buff *)list_)
		skb = NULL;
	return skb;

L
Linus Torvalds 已提交
987 988 989 990 991 992 993 994 995 996 997 998 999
}

/**
 *	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;
}

1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015
/**
 *	__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;
}

1016 1017 1018 1019 1020 1021 1022 1023
/*
 * 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
1024 1025 1026
static inline void skb_queue_head_init(struct sk_buff_head *list)
{
	spin_lock_init(&list->lock);
1027
	__skb_queue_head_init(list);
L
Linus Torvalds 已提交
1028 1029
}

1030 1031 1032 1033 1034 1035 1036
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
1037
/*
1038
 *	Insert an sk_buff on a list.
L
Linus Torvalds 已提交
1039 1040 1041 1042
 *
 *	The "__skb_xxxx()" functions are the non-atomic ones that
 *	can only be called with interrupts disabled.
 */
1043 1044 1045 1046 1047 1048 1049 1050 1051 1052
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
1053

1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077
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);
1078
		head->qlen += list->qlen;
1079 1080 1081 1082
	}
}

/**
E
Eric Dumazet 已提交
1083
 *	skb_queue_splice_init - join two skb lists and reinitialise the emptied list
1084 1085 1086 1087 1088 1089 1090 1091 1092 1093
 *	@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);
1094
		head->qlen += list->qlen;
1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108
		__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);
1109
		head->qlen += list->qlen;
1110 1111 1112 1113
	}
}

/**
E
Eric Dumazet 已提交
1114
 *	skb_queue_splice_tail_init - join two skb lists and reinitialise the emptied list
1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125
 *	@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);
1126
		head->qlen += list->qlen;
1127 1128 1129 1130
		__skb_queue_head_init(list);
	}
}

L
Linus Torvalds 已提交
1131
/**
1132
 *	__skb_queue_after - queue a buffer at the list head
L
Linus Torvalds 已提交
1133
 *	@list: list to use
1134
 *	@prev: place after this buffer
L
Linus Torvalds 已提交
1135 1136
 *	@newsk: buffer to queue
 *
1137
 *	Queue a buffer int the middle of a list. This function takes no locks
L
Linus Torvalds 已提交
1138 1139 1140 1141
 *	and you must therefore hold required locks before calling it.
 *
 *	A buffer cannot be placed on two lists at the same time.
 */
1142 1143 1144
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
1145
{
1146
	__skb_insert(newsk, prev, prev->next, list);
L
Linus Torvalds 已提交
1147 1148
}

1149 1150 1151
extern void skb_append(struct sk_buff *old, struct sk_buff *newsk,
		       struct sk_buff_head *list);

1152 1153 1154 1155 1156 1157 1158
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);
}

1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175
/**
 *	__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
1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189
/**
 *	__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)
{
1190
	__skb_queue_before(list, (struct sk_buff *)list, newsk);
L
Linus Torvalds 已提交
1191 1192 1193 1194 1195 1196
}

/*
 * remove sk_buff from list. _Must_ be called atomically, and with
 * the list known..
 */
D
David S. Miller 已提交
1197
extern void	   skb_unlink(struct sk_buff *skb, struct sk_buff_head *list);
L
Linus Torvalds 已提交
1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209
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;
}

1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225
/**
 *	__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
1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244

/**
 *	__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;
}


1245
static inline bool skb_is_nonlinear(const struct sk_buff *skb)
L
Linus Torvalds 已提交
1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259
{
	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--)
E
Eric Dumazet 已提交
1260
		len += skb_frag_size(&skb_shinfo(skb)->frags[i]);
L
Linus Torvalds 已提交
1261 1262 1263
	return len + skb_headlen(skb);
}

1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278
/**
 * __skb_fill_page_desc - initialise a paged fragment in an skb
 * @skb: buffer containing fragment to be initialised
 * @i: paged fragment index to initialise
 * @page: the page to use for this fragment
 * @off: the offset to the data with @page
 * @size: the length of the data
 *
 * Initialises the @i'th fragment of @skb to point to &size bytes at
 * offset @off within @page.
 *
 * Does not take any additional reference on the fragment.
 */
static inline void __skb_fill_page_desc(struct sk_buff *skb, int i,
					struct page *page, int off, int size)
L
Linus Torvalds 已提交
1279 1280 1281
{
	skb_frag_t *frag = &skb_shinfo(skb)->frags[i];

1282 1283 1284 1285 1286 1287 1288 1289 1290
	/*
	 * Propagate page->pfmemalloc to the skb if we can. The problem is
	 * that not all callers have unique ownership of the page. If
	 * pfmemalloc is set, we check the mapping as a mapping implies
	 * page->index is set (index and pfmemalloc share space).
	 * If it's a valid mapping, we cannot use page->pfmemalloc but we
	 * do not lose pfmemalloc information as the pages would not be
	 * allocated using __GFP_MEMALLOC.
	 */
1291
	frag->page.p		  = page;
L
Linus Torvalds 已提交
1292
	frag->page_offset	  = off;
E
Eric Dumazet 已提交
1293
	skb_frag_size_set(frag, size);
1294 1295 1296 1297

	page = compound_head(page);
	if (page->pfmemalloc && !page->mapping)
		skb->pfmemalloc	= true;
1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317
}

/**
 * skb_fill_page_desc - initialise a paged fragment in an skb
 * @skb: buffer containing fragment to be initialised
 * @i: paged fragment index to initialise
 * @page: the page to use for this fragment
 * @off: the offset to the data with @page
 * @size: the length of the data
 *
 * As per __skb_fill_page_desc() -- initialises the @i'th fragment of
 * @skb to point to &size bytes at offset @off within @page. In
 * addition updates @skb such that @i is the last fragment.
 *
 * Does not take any additional reference on the fragment.
 */
static inline void skb_fill_page_desc(struct sk_buff *skb, int i,
				      struct page *page, int off, int size)
{
	__skb_fill_page_desc(skb, i, page, off, size);
L
Linus Torvalds 已提交
1318 1319 1320
	skb_shinfo(skb)->nr_frags = i + 1;
}

P
Peter Zijlstra 已提交
1321
extern void skb_add_rx_frag(struct sk_buff *skb, int i, struct page *page,
1322
			    int off, int size, unsigned int truesize);
P
Peter Zijlstra 已提交
1323

L
Linus Torvalds 已提交
1324
#define SKB_PAGE_ASSERT(skb) 	BUG_ON(skb_shinfo(skb)->nr_frags)
1325
#define SKB_FRAG_ASSERT(skb) 	BUG_ON(skb_has_frag_list(skb))
L
Linus Torvalds 已提交
1326 1327
#define SKB_LINEAR_ASSERT(skb)  BUG_ON(skb_is_nonlinear(skb))

1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358
#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;
}
1359

1360 1361
#endif /* NET_SKBUFF_DATA_USES_OFFSET */

L
Linus Torvalds 已提交
1362 1363 1364
/*
 *	Add data to an sk_buff
 */
1365
extern unsigned char *skb_put(struct sk_buff *skb, unsigned int len);
L
Linus Torvalds 已提交
1366 1367
static inline unsigned char *__skb_put(struct sk_buff *skb, unsigned int len)
{
1368
	unsigned char *tmp = skb_tail_pointer(skb);
L
Linus Torvalds 已提交
1369 1370 1371 1372 1373 1374
	SKB_LINEAR_ASSERT(skb);
	skb->tail += len;
	skb->len  += len;
	return tmp;
}

1375
extern unsigned char *skb_push(struct sk_buff *skb, unsigned int len);
L
Linus Torvalds 已提交
1376 1377 1378 1379 1380 1381 1382
static inline unsigned char *__skb_push(struct sk_buff *skb, unsigned int len)
{
	skb->data -= len;
	skb->len  += len;
	return skb->data;
}

1383
extern unsigned char *skb_pull(struct sk_buff *skb, unsigned int len);
L
Linus Torvalds 已提交
1384 1385 1386 1387 1388 1389 1390
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;
}

1391 1392 1393 1394 1395
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
1396 1397 1398 1399 1400
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
1401
	    !__pskb_pull_tail(skb, len - skb_headlen(skb)))
L
Linus Torvalds 已提交
1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417
		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
1418
	return __pskb_pull_tail(skb, len - skb_headlen(skb)) != NULL;
L
Linus Torvalds 已提交
1419 1420 1421 1422 1423 1424 1425 1426
}

/**
 *	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.
 */
1427
static inline unsigned int skb_headroom(const struct sk_buff *skb)
L
Linus Torvalds 已提交
1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439
{
	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)
{
1440
	return skb_is_nonlinear(skb) ? 0 : skb->end - skb->tail;
L
Linus Torvalds 已提交
1441 1442
}

1443 1444 1445 1446 1447 1448 1449 1450 1451
/**
 *	skb_availroom - bytes at buffer end
 *	@skb: buffer to check
 *
 *	Return the number of bytes of free space at the tail of an sk_buff
 *	allocated by sk_stream_alloc()
 */
static inline int skb_availroom(const struct sk_buff *skb)
{
E
Eric Dumazet 已提交
1452 1453 1454 1455
	if (skb_is_nonlinear(skb))
		return 0;

	return skb->end - skb->tail - skb->reserved_tailroom;
1456 1457
}

L
Linus Torvalds 已提交
1458 1459 1460 1461 1462 1463 1464 1465
/**
 *	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.
 */
1466
static inline void skb_reserve(struct sk_buff *skb, int len)
L
Linus Torvalds 已提交
1467 1468 1469 1470 1471
{
	skb->data += len;
	skb->tail += len;
}

1472 1473 1474 1475 1476 1477
static inline void skb_reset_inner_headers(struct sk_buff *skb)
{
	skb->inner_network_header = skb->network_header;
	skb->inner_transport_header = skb->transport_header;
}

1478 1479 1480 1481 1482
static inline void skb_reset_mac_len(struct sk_buff *skb)
{
	skb->mac_len = skb->network_header - skb->mac_header;
}

1483
#ifdef NET_SKBUFF_DATA_USES_OFFSET
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 1516 1517 1518
static inline unsigned char *skb_inner_transport_header(const struct sk_buff
							*skb)
{
	return skb->head + skb->inner_transport_header;
}

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

static inline void skb_set_inner_transport_header(struct sk_buff *skb,
						   const int offset)
{
	skb_reset_inner_transport_header(skb);
	skb->inner_transport_header += offset;
}

static inline unsigned char *skb_inner_network_header(const struct sk_buff *skb)
{
	return skb->head + skb->inner_network_header;
}

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

static inline void skb_set_inner_network_header(struct sk_buff *skb,
						const int offset)
{
	skb_reset_inner_network_header(skb);
	skb->inner_network_header += offset;
}

1519 1520 1521 1522 1523
static inline bool skb_transport_header_was_set(const struct sk_buff *skb)
{
	return skb->transport_header != ~0U;
}

1524 1525
static inline unsigned char *skb_transport_header(const struct sk_buff *skb)
{
1526
	return skb->head + skb->transport_header;
1527 1528
}

1529 1530
static inline void skb_reset_transport_header(struct sk_buff *skb)
{
1531
	skb->transport_header = skb->data - skb->head;
1532 1533
}

1534 1535 1536
static inline void skb_set_transport_header(struct sk_buff *skb,
					    const int offset)
{
1537 1538
	skb_reset_transport_header(skb);
	skb->transport_header += offset;
1539 1540
}

1541 1542
static inline unsigned char *skb_network_header(const struct sk_buff *skb)
{
1543
	return skb->head + skb->network_header;
1544 1545
}

1546 1547
static inline void skb_reset_network_header(struct sk_buff *skb)
{
1548
	skb->network_header = skb->data - skb->head;
1549 1550
}

1551 1552
static inline void skb_set_network_header(struct sk_buff *skb, const int offset)
{
1553 1554
	skb_reset_network_header(skb);
	skb->network_header += offset;
1555 1556
}

1557
static inline unsigned char *skb_mac_header(const struct sk_buff *skb)
1558
{
1559
	return skb->head + skb->mac_header;
1560 1561
}

1562
static inline int skb_mac_header_was_set(const struct sk_buff *skb)
1563
{
1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578
	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 */
1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610
static inline unsigned char *skb_inner_transport_header(const struct sk_buff
							*skb)
{
	return skb->inner_transport_header;
}

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

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

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

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

static inline void skb_set_inner_network_header(struct sk_buff *skb,
						const int offset)
{
	skb->inner_network_header = skb->data + offset;
}
1611

1612 1613 1614 1615 1616
static inline bool skb_transport_header_was_set(const struct sk_buff *skb)
{
	return skb->transport_header != NULL;
}

1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645
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;
1646 1647
}

1648 1649
static inline unsigned char *skb_mac_header(const struct sk_buff *skb)
{
1650
	return skb->mac_header;
1651 1652 1653 1654
}

static inline int skb_mac_header_was_set(const struct sk_buff *skb)
{
1655
	return skb->mac_header != NULL;
1656 1657
}

1658 1659
static inline void skb_reset_mac_header(struct sk_buff *skb)
{
1660
	skb->mac_header = skb->data;
1661 1662
}

1663 1664
static inline void skb_set_mac_header(struct sk_buff *skb, const int offset)
{
1665
	skb->mac_header = skb->data + offset;
1666
}
1667 1668
#endif /* NET_SKBUFF_DATA_USES_OFFSET */

1669 1670 1671 1672 1673 1674 1675 1676 1677 1678
static inline void skb_mac_header_rebuild(struct sk_buff *skb)
{
	if (skb_mac_header_was_set(skb)) {
		const unsigned char *old_mac = skb_mac_header(skb);

		skb_set_mac_header(skb, -skb->mac_len);
		memmove(skb_mac_header(skb), old_mac, skb->mac_len);
	}
}

1679 1680 1681 1682 1683
static inline int skb_checksum_start_offset(const struct sk_buff *skb)
{
	return skb->csum_start - skb_headroom(skb);
}

1684 1685 1686 1687 1688 1689 1690 1691 1692 1693
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;
}

1694 1695 1696 1697 1698
static inline u32 skb_inner_network_header_len(const struct sk_buff *skb)
{
	return skb->inner_transport_header - skb->inner_network_header;
}

1699 1700 1701 1702
static inline int skb_network_offset(const struct sk_buff *skb)
{
	return skb_network_header(skb) - skb->data;
}
1703

1704 1705 1706 1707 1708
static inline int skb_inner_network_offset(const struct sk_buff *skb)
{
	return skb_inner_network_header(skb) - skb->data;
}

1709 1710 1711 1712 1713
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
1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724
/*
 * 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:
 *
1725
 * skb_reserve(skb, NET_IP_ALIGN);
L
Linus Torvalds 已提交
1726 1727 1728 1729
 *
 * 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.
1730
 *
L
Linus Torvalds 已提交
1731 1732 1733 1734 1735 1736 1737
 * 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

1738 1739 1740 1741
/*
 * 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
1742
 * 32 bytes or less we avoid the reallocation.
1743 1744 1745 1746 1747 1748 1749
 *
 * 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.
 *
1750
 * Various parts of the networking layer expect at least 32 bytes of
1751
 * headroom, you should not reduce this.
1752 1753 1754 1755
 *
 * 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 :
1756
 * NET_IP_ALIGN(2) + ethernet_header(14) + IP_header(20/40) + ports(8)
1757 1758
 */
#ifndef NET_SKB_PAD
1759
#define NET_SKB_PAD	max(32, L1_CACHE_BYTES)
1760 1761
#endif

1762
extern int ___pskb_trim(struct sk_buff *skb, unsigned int len);
L
Linus Torvalds 已提交
1763 1764 1765

static inline void __skb_trim(struct sk_buff *skb, unsigned int len)
{
1766
	if (unlikely(skb_is_nonlinear(skb))) {
1767 1768 1769
		WARN_ON(1);
		return;
	}
1770 1771
	skb->len = len;
	skb_set_tail_pointer(skb, len);
L
Linus Torvalds 已提交
1772 1773
}

1774
extern void skb_trim(struct sk_buff *skb, unsigned int len);
L
Linus Torvalds 已提交
1775 1776 1777

static inline int __pskb_trim(struct sk_buff *skb, unsigned int len)
{
1778 1779 1780 1781
	if (skb->data_len)
		return ___pskb_trim(skb, len);
	__skb_trim(skb, len);
	return 0;
L
Linus Torvalds 已提交
1782 1783 1784 1785 1786 1787 1788
}

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

1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803
/**
 *	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
1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819
/**
 *	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;
}

1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835
/**
 *	skb_orphan_frags - orphan the frags contained in a buffer
 *	@skb: buffer to orphan frags from
 *	@gfp_mask: allocation mask for replacement pages
 *
 *	For each frag in the SKB which needs a destructor (i.e. has an
 *	owner) create a copy of that frag and release the original
 *	page by calling the destructor.
 */
static inline int skb_orphan_frags(struct sk_buff *skb, gfp_t gfp_mask)
{
	if (likely(!(skb_shinfo(skb)->tx_flags & SKBTX_DEV_ZEROCOPY)))
		return 0;
	return skb_copy_ubufs(skb, gfp_mask);
}

L
Linus Torvalds 已提交
1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851
/**
 *	__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);
}

1852 1853 1854 1855
#define NETDEV_FRAG_PAGE_MAX_ORDER get_order(32768)
#define NETDEV_FRAG_PAGE_MAX_SIZE  (PAGE_SIZE << NETDEV_FRAG_PAGE_MAX_ORDER)
#define NETDEV_PAGECNT_MAX_BIAS	   NETDEV_FRAG_PAGE_MAX_SIZE

1856
extern void *netdev_alloc_frag(unsigned int fragsz);
L
Linus Torvalds 已提交
1857

1858
extern struct sk_buff *__netdev_alloc_skb(struct net_device *dev,
1859 1860
					  unsigned int length,
					  gfp_t gfp_mask);
1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875

/**
 *	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,
1876
					       unsigned int length)
1877 1878 1879 1880
{
	return __netdev_alloc_skb(dev, length, GFP_ATOMIC);
}

1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894
/* legacy helper around __netdev_alloc_skb() */
static inline struct sk_buff *__dev_alloc_skb(unsigned int length,
					      gfp_t gfp_mask)
{
	return __netdev_alloc_skb(NULL, length, gfp_mask);
}

/* legacy helper around netdev_alloc_skb() */
static inline struct sk_buff *dev_alloc_skb(unsigned int length)
{
	return netdev_alloc_skb(NULL, length);
}


1895 1896
static inline struct sk_buff *__netdev_alloc_skb_ip_align(struct net_device *dev,
		unsigned int length, gfp_t gfp)
1897
{
1898
	struct sk_buff *skb = __netdev_alloc_skb(dev, length + NET_IP_ALIGN, gfp);
1899 1900 1901 1902 1903 1904

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

1905 1906 1907 1908 1909 1910
static inline struct sk_buff *netdev_alloc_skb_ip_align(struct net_device *dev,
		unsigned int length)
{
	return __netdev_alloc_skb_ip_align(dev, length, GFP_ATOMIC);
}

1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965
/*
 *	__skb_alloc_page - allocate pages for ps-rx on a skb and preserve pfmemalloc data
 *	@gfp_mask: alloc_pages_node mask. Set __GFP_NOMEMALLOC if not for network packet RX
 *	@skb: skb to set pfmemalloc on if __GFP_MEMALLOC is used
 *	@order: size of the allocation
 *
 * 	Allocate a new page.
 *
 * 	%NULL is returned if there is no free memory.
*/
static inline struct page *__skb_alloc_pages(gfp_t gfp_mask,
					      struct sk_buff *skb,
					      unsigned int order)
{
	struct page *page;

	gfp_mask |= __GFP_COLD;

	if (!(gfp_mask & __GFP_NOMEMALLOC))
		gfp_mask |= __GFP_MEMALLOC;

	page = alloc_pages_node(NUMA_NO_NODE, gfp_mask, order);
	if (skb && page && page->pfmemalloc)
		skb->pfmemalloc = true;

	return page;
}

/**
 *	__skb_alloc_page - allocate a page for ps-rx for a given skb and preserve pfmemalloc data
 *	@gfp_mask: alloc_pages_node mask. Set __GFP_NOMEMALLOC if not for network packet RX
 *	@skb: skb to set pfmemalloc on if __GFP_MEMALLOC is used
 *
 * 	Allocate a new page.
 *
 * 	%NULL is returned if there is no free memory.
 */
static inline struct page *__skb_alloc_page(gfp_t gfp_mask,
					     struct sk_buff *skb)
{
	return __skb_alloc_pages(gfp_mask, skb, 0);
}

/**
 *	skb_propagate_pfmemalloc - Propagate pfmemalloc if skb is allocated after RX page
 *	@page: The page that was allocated from skb_alloc_page
 *	@skb: The skb that may need pfmemalloc set
 */
static inline void skb_propagate_pfmemalloc(struct page *page,
					     struct sk_buff *skb)
{
	if (page && page->pfmemalloc)
		skb->pfmemalloc = true;
}

1966 1967 1968 1969 1970 1971 1972 1973
/**
 * skb_frag_page - retrieve the page refered to by a paged fragment
 * @frag: the paged fragment
 *
 * Returns the &struct page associated with @frag.
 */
static inline struct page *skb_frag_page(const skb_frag_t *frag)
{
1974
	return frag->page.p;
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 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059
}

/**
 * __skb_frag_ref - take an addition reference on a paged fragment.
 * @frag: the paged fragment
 *
 * Takes an additional reference on the paged fragment @frag.
 */
static inline void __skb_frag_ref(skb_frag_t *frag)
{
	get_page(skb_frag_page(frag));
}

/**
 * skb_frag_ref - take an addition reference on a paged fragment of an skb.
 * @skb: the buffer
 * @f: the fragment offset.
 *
 * Takes an additional reference on the @f'th paged fragment of @skb.
 */
static inline void skb_frag_ref(struct sk_buff *skb, int f)
{
	__skb_frag_ref(&skb_shinfo(skb)->frags[f]);
}

/**
 * __skb_frag_unref - release a reference on a paged fragment.
 * @frag: the paged fragment
 *
 * Releases a reference on the paged fragment @frag.
 */
static inline void __skb_frag_unref(skb_frag_t *frag)
{
	put_page(skb_frag_page(frag));
}

/**
 * skb_frag_unref - release a reference on a paged fragment of an skb.
 * @skb: the buffer
 * @f: the fragment offset
 *
 * Releases a reference on the @f'th paged fragment of @skb.
 */
static inline void skb_frag_unref(struct sk_buff *skb, int f)
{
	__skb_frag_unref(&skb_shinfo(skb)->frags[f]);
}

/**
 * skb_frag_address - gets the address of the data contained in a paged fragment
 * @frag: the paged fragment buffer
 *
 * Returns the address of the data within @frag. The page must already
 * be mapped.
 */
static inline void *skb_frag_address(const skb_frag_t *frag)
{
	return page_address(skb_frag_page(frag)) + frag->page_offset;
}

/**
 * skb_frag_address_safe - gets the address of the data contained in a paged fragment
 * @frag: the paged fragment buffer
 *
 * Returns the address of the data within @frag. Checks that the page
 * is mapped and returns %NULL otherwise.
 */
static inline void *skb_frag_address_safe(const skb_frag_t *frag)
{
	void *ptr = page_address(skb_frag_page(frag));
	if (unlikely(!ptr))
		return NULL;

	return ptr + frag->page_offset;
}

/**
 * __skb_frag_set_page - sets the page contained in a paged fragment
 * @frag: the paged fragment
 * @page: the page to set
 *
 * Sets the fragment @frag to contain @page.
 */
static inline void __skb_frag_set_page(skb_frag_t *frag, struct page *page)
{
2060
	frag->page.p = page;
2061 2062 2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077 2078
}

/**
 * skb_frag_set_page - sets the page contained in a paged fragment of an skb
 * @skb: the buffer
 * @f: the fragment offset
 * @page: the page to set
 *
 * Sets the @f'th fragment of @skb to contain @page.
 */
static inline void skb_frag_set_page(struct sk_buff *skb, int f,
				     struct page *page)
{
	__skb_frag_set_page(&skb_shinfo(skb)->frags[f], page);
}

/**
 * skb_frag_dma_map - maps a paged fragment via the DMA API
2079
 * @dev: the device to map the fragment to
2080 2081 2082 2083
 * @frag: the paged fragment to map
 * @offset: the offset within the fragment (starting at the
 *          fragment's own offset)
 * @size: the number of bytes to map
2084
 * @dir: the direction of the mapping (%PCI_DMA_*)
2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096
 *
 * Maps the page associated with @frag to @device.
 */
static inline dma_addr_t skb_frag_dma_map(struct device *dev,
					  const skb_frag_t *frag,
					  size_t offset, size_t size,
					  enum dma_data_direction dir)
{
	return dma_map_page(dev, skb_frag_page(frag),
			    frag->page_offset + offset, size, dir);
}

E
Eric Dumazet 已提交
2097 2098 2099 2100 2101 2102
static inline struct sk_buff *pskb_copy(struct sk_buff *skb,
					gfp_t gfp_mask)
{
	return __pskb_copy(skb, skb_headroom(skb), gfp_mask);
}

2103 2104 2105 2106 2107 2108 2109 2110
/**
 *	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.
 */
2111
static inline int skb_clone_writable(const struct sk_buff *skb, unsigned int len)
2112 2113 2114 2115 2116
{
	return !skb_header_cloned(skb) &&
	       skb_headroom(skb) + len <= skb->hdr_len;
}

H
Herbert Xu 已提交
2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130
static inline int __skb_cow(struct sk_buff *skb, unsigned int headroom,
			    int cloned)
{
	int delta = 0;

	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
2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144
/**
 *	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
2145 2146
	return __skb_cow(skb, headroom, skb_cloned(skb));
}
L
Linus Torvalds 已提交
2147

H
Herbert Xu 已提交
2148 2149 2150 2151 2152 2153 2154 2155 2156 2157 2158 2159 2160
/**
 *	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
2161 2162 2163 2164 2165 2166 2167 2168 2169
}

/**
 *	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
2170 2171
 *	is untouched. Otherwise it is extended. Returns zero on
 *	success. The skb is freed on error.
L
Linus Torvalds 已提交
2172 2173
 */
 
2174
static inline int skb_padto(struct sk_buff *skb, unsigned int len)
L
Linus Torvalds 已提交
2175 2176 2177
{
	unsigned int size = skb->len;
	if (likely(size >= len))
2178
		return 0;
G
Gerrit Renker 已提交
2179
	return skb_pad(skb, len - size);
L
Linus Torvalds 已提交
2180 2181 2182 2183 2184 2185 2186 2187 2188
}

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;
2189
		__wsum csum = csum_and_copy_from_user(from, skb_put(skb, copy),
L
Linus Torvalds 已提交
2190 2191 2192 2193 2194 2195 2196 2197 2198 2199 2200 2201
							    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;
}

2202 2203
static inline bool skb_can_coalesce(struct sk_buff *skb, int i,
				    const struct page *page, int off)
L
Linus Torvalds 已提交
2204 2205
{
	if (i) {
E
Eric Dumazet 已提交
2206
		const struct skb_frag_struct *frag = &skb_shinfo(skb)->frags[i - 1];
L
Linus Torvalds 已提交
2207

2208
		return page == skb_frag_page(frag) &&
E
Eric Dumazet 已提交
2209
		       off == frag->page_offset + skb_frag_size(frag);
L
Linus Torvalds 已提交
2210
	}
2211
	return false;
L
Linus Torvalds 已提交
2212 2213
}

H
Herbert Xu 已提交
2214 2215 2216 2217 2218
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
2219 2220 2221 2222 2223 2224 2225
/**
 *	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
2226 2227 2228 2229 2230
static inline int skb_linearize(struct sk_buff *skb)
{
	return skb_is_nonlinear(skb) ? __skb_linearize(skb) : 0;
}

2231 2232 2233 2234 2235 2236 2237 2238 2239
/**
 * skb_has_shared_frag - can any frag be overwritten
 * @skb: buffer to test
 *
 * Return true if the skb has at least one frag that might be modified
 * by an external entity (as in vmsplice()/sendfile())
 */
static inline bool skb_has_shared_frag(const struct sk_buff *skb)
{
2240 2241
	return skb_is_nonlinear(skb) &&
	       skb_shinfo(skb)->tx_flags & SKBTX_SHARED_FRAG;
2242 2243
}

H
Herbert Xu 已提交
2244 2245 2246 2247 2248 2249 2250 2251
/**
 *	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
2252
{
H
Herbert Xu 已提交
2253 2254
	return skb_is_nonlinear(skb) || skb_cloned(skb) ?
	       __skb_linearize(skb) : 0;
L
Linus Torvalds 已提交
2255 2256 2257 2258 2259 2260 2261 2262 2263
}

/**
 *	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
2264 2265
 *	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
2266 2267 2268
 */

static inline void skb_postpull_rcsum(struct sk_buff *skb,
2269
				      const void *start, unsigned int len)
L
Linus Torvalds 已提交
2270
{
2271
	if (skb->ip_summed == CHECKSUM_COMPLETE)
L
Linus Torvalds 已提交
2272 2273 2274
		skb->csum = csum_sub(skb->csum, csum_partial(start, len, 0));
}

2275 2276
unsigned char *skb_pull_rcsum(struct sk_buff *skb, unsigned int len);

L
Linus Torvalds 已提交
2277 2278 2279 2280 2281 2282 2283 2284 2285 2286 2287
/**
 *	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)
{
2288
	if (likely(len >= skb->len))
L
Linus Torvalds 已提交
2289
		return 0;
2290
	if (skb->ip_summed == CHECKSUM_COMPLETE)
L
Linus Torvalds 已提交
2291 2292 2293 2294 2295 2296
		skb->ip_summed = CHECKSUM_NONE;
	return __pskb_trim(skb, len);
}

#define skb_queue_walk(queue, skb) \
		for (skb = (queue)->next;					\
2297
		     skb != (struct sk_buff *)(queue);				\
L
Linus Torvalds 已提交
2298 2299
		     skb = skb->next)

2300 2301 2302 2303 2304
#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)

2305
#define skb_queue_walk_from(queue, skb)						\
2306
		for (; skb != (struct sk_buff *)(queue);			\
2307 2308 2309 2310 2311 2312 2313
		     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)

2314 2315
#define skb_queue_reverse_walk(queue, skb) \
		for (skb = (queue)->prev;					\
2316
		     skb != (struct sk_buff *)(queue);				\
2317 2318
		     skb = skb->prev)

2319 2320 2321 2322 2323 2324 2325 2326 2327
#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
2328

2329
static inline bool skb_has_frag_list(const struct sk_buff *skb)
2330 2331 2332 2333 2334 2335 2336 2337 2338 2339 2340 2341 2342 2343 2344 2345 2346 2347
{
	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)

2348
extern struct sk_buff *__skb_recv_datagram(struct sock *sk, unsigned flags,
2349
					   int *peeked, int *off, int *err);
L
Linus Torvalds 已提交
2350 2351 2352 2353 2354 2355 2356
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);
2357
extern int	       skb_copy_and_csum_datagram_iovec(struct sk_buff *skb,
L
Linus Torvalds 已提交
2358 2359
							int hlen,
							struct iovec *iov);
2360 2361
extern int	       skb_copy_datagram_from_iovec(struct sk_buff *skb,
						    int offset,
2362 2363
						    const struct iovec *from,
						    int from_offset,
2364
						    int len);
2365 2366 2367 2368 2369
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
2370
extern void	       skb_free_datagram(struct sock *sk, struct sk_buff *skb);
2371 2372
extern void	       skb_free_datagram_locked(struct sock *sk,
						struct sk_buff *skb);
2373
extern int	       skb_kill_datagram(struct sock *sk, struct sk_buff *skb,
2374
					 unsigned int flags);
2375 2376
extern __wsum	       skb_checksum(const struct sk_buff *skb, int offset,
				    int len, __wsum csum);
L
Linus Torvalds 已提交
2377 2378
extern int	       skb_copy_bits(const struct sk_buff *skb, int offset,
				     void *to, int len);
2379 2380
extern int	       skb_store_bits(struct sk_buff *skb, int offset,
				      const void *from, int len);
2381
extern __wsum	       skb_copy_and_csum_bits(const struct sk_buff *skb,
L
Linus Torvalds 已提交
2382
					      int offset, u8 *to, int len,
2383
					      __wsum csum);
J
Jens Axboe 已提交
2384 2385 2386 2387 2388
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
2389 2390 2391
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);
2392 2393
extern int	       skb_shift(struct sk_buff *tgt, struct sk_buff *skb,
				 int shiftlen);
L
Linus Torvalds 已提交
2394

2395 2396
extern struct sk_buff *skb_segment(struct sk_buff *skb,
				   netdev_features_t features);
2397

L
Linus Torvalds 已提交
2398 2399 2400 2401 2402
static inline void *skb_header_pointer(const struct sk_buff *skb, int offset,
				       int len, void *buffer)
{
	int hlen = skb_headlen(skb);

2403
	if (hlen - offset >= len)
L
Linus Torvalds 已提交
2404 2405 2406 2407 2408 2409 2410 2411
		return skb->data + offset;

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

	return buffer;
}

2412 2413 2414 2415 2416 2417 2418 2419 2420 2421 2422 2423 2424 2425
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);
}

2426 2427 2428 2429 2430 2431 2432 2433 2434 2435 2436 2437 2438 2439 2440
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
2441 2442
extern void skb_init(void);

2443 2444 2445 2446 2447
static inline ktime_t skb_get_ktime(const struct sk_buff *skb)
{
	return skb->tstamp;
}

2448 2449 2450 2451 2452 2453 2454 2455 2456
/**
 *	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.
 */
2457 2458
static inline void skb_get_timestamp(const struct sk_buff *skb,
				     struct timeval *stamp)
2459
{
2460
	*stamp = ktime_to_timeval(skb->tstamp);
2461 2462
}

2463 2464 2465 2466 2467 2468
static inline void skb_get_timestampns(const struct sk_buff *skb,
				       struct timespec *stamp)
{
	*stamp = ktime_to_timespec(skb->tstamp);
}

2469
static inline void __net_timestamp(struct sk_buff *skb)
2470
{
2471
	skb->tstamp = ktime_get_real();
2472 2473
}

2474 2475 2476 2477 2478
static inline ktime_t net_timedelta(ktime_t t)
{
	return ktime_sub(ktime_get_real(), t);
}

2479 2480 2481 2482
static inline ktime_t net_invalid_timestamp(void)
{
	return ktime_set(0, 0);
}
2483

2484 2485 2486 2487 2488 2489 2490 2491 2492 2493 2494 2495 2496 2497 2498 2499 2500 2501 2502 2503 2504 2505 2506
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
 *
2507 2508 2509 2510 2511
 * PHY drivers may accept clones of transmitted packets for
 * timestamping via their phy_driver.txtstamp method. These drivers
 * must call this function to return the skb back to the stack, with
 * or without a timestamp.
 *
2512
 * @skb: clone of the the original outgoing packet
2513
 * @hwtstamps: hardware time stamps, may be NULL if not available
2514 2515 2516 2517 2518
 *
 */
void skb_complete_tx_timestamp(struct sk_buff *skb,
			       struct skb_shared_hwtstamps *hwtstamps);

2519 2520 2521 2522 2523 2524 2525 2526 2527 2528 2529 2530 2531 2532
/**
 * 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);

2533 2534
static inline void sw_tx_timestamp(struct sk_buff *skb)
{
2535 2536
	if (skb_shinfo(skb)->tx_flags & SKBTX_SW_TSTAMP &&
	    !(skb_shinfo(skb)->tx_flags & SKBTX_IN_PROGRESS))
2537 2538 2539 2540 2541 2542 2543
		skb_tstamp_tx(skb, NULL);
}

/**
 * skb_tx_timestamp() - Driver hook for transmit timestamping
 *
 * Ethernet MAC Drivers should call this function in their hard_xmit()
2544
 * function immediately before giving the sk_buff to the MAC hardware.
2545 2546 2547 2548 2549
 *
 * @skb: A socket buffer.
 */
static inline void skb_tx_timestamp(struct sk_buff *skb)
{
2550
	skb_clone_tx_timestamp(skb);
2551 2552 2553
	sw_tx_timestamp(skb);
}

2554 2555 2556 2557 2558 2559 2560 2561 2562
/**
 * skb_complete_wifi_ack - deliver skb with wifi status
 *
 * @skb: the original outgoing packet
 * @acked: ack status
 *
 */
void skb_complete_wifi_ack(struct sk_buff *skb, bool acked);

2563
extern __sum16 __skb_checksum_complete_head(struct sk_buff *skb, int len);
2564
extern __sum16 __skb_checksum_complete(struct sk_buff *skb);
2565

2566 2567 2568 2569 2570
static inline int skb_csum_unnecessary(const struct sk_buff *skb)
{
	return skb->ip_summed & CHECKSUM_UNNECESSARY;
}

2571 2572 2573 2574 2575 2576 2577 2578 2579 2580 2581 2582 2583 2584 2585 2586
/**
 *	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.
 */
2587
static inline __sum16 skb_checksum_complete(struct sk_buff *skb)
2588
{
2589 2590
	return skb_csum_unnecessary(skb) ?
	       0 : __skb_checksum_complete(skb);
2591 2592
}

2593
#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE)
2594
extern void nf_conntrack_destroy(struct nf_conntrack *nfct);
L
Linus Torvalds 已提交
2595 2596 2597
static inline void nf_conntrack_put(struct nf_conntrack *nfct)
{
	if (nfct && atomic_dec_and_test(&nfct->use))
2598
		nf_conntrack_destroy(nfct);
L
Linus Torvalds 已提交
2599 2600 2601 2602 2603 2604
}
static inline void nf_conntrack_get(struct nf_conntrack *nfct)
{
	if (nfct)
		atomic_inc(&nfct->use);
}
2605 2606
#endif
#ifdef NET_SKBUFF_NF_DEFRAG_NEEDED
2607 2608 2609 2610 2611 2612 2613 2614 2615 2616 2617
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
2618 2619 2620 2621 2622 2623 2624 2625 2626 2627 2628 2629
#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 */
2630 2631
static inline void nf_reset(struct sk_buff *skb)
{
2632
#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE)
2633 2634
	nf_conntrack_put(skb->nfct);
	skb->nfct = NULL;
2635 2636
#endif
#ifdef NET_SKBUFF_NF_DEFRAG_NEEDED
2637 2638 2639 2640 2641 2642 2643
	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
G
Gao feng 已提交
2644 2645 2646
#if IS_ENABLED(CONFIG_NETFILTER_XT_TARGET_TRACE)
	skb->nf_trace = 0;
#endif
2647 2648
}

2649 2650 2651
/* 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)
{
2652
#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE)
2653 2654 2655
	dst->nfct = src->nfct;
	nf_conntrack_get(src->nfct);
	dst->nfctinfo = src->nfctinfo;
2656 2657
#endif
#ifdef NET_SKBUFF_NF_DEFRAG_NEEDED
2658 2659 2660 2661 2662 2663 2664 2665 2666
	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
}

2667 2668 2669
static inline void nf_copy(struct sk_buff *dst, const struct sk_buff *src)
{
#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE)
2670
	nf_conntrack_put(dst->nfct);
2671 2672
#endif
#ifdef NET_SKBUFF_NF_DEFRAG_NEEDED
2673 2674 2675 2676 2677 2678 2679 2680
	nf_conntrack_put_reasm(dst->nfct_reasm);
#endif
#ifdef CONFIG_BRIDGE_NETFILTER
	nf_bridge_put(dst->nf_bridge);
#endif
	__nf_copy(dst, src);
}

2681 2682 2683 2684 2685 2686 2687 2688 2689 2690 2691 2692 2693 2694 2695 2696 2697 2698
#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

2699 2700 2701 2702 2703
static inline void skb_set_queue_mapping(struct sk_buff *skb, u16 queue_mapping)
{
	skb->queue_mapping = queue_mapping;
}

2704
static inline u16 skb_get_queue_mapping(const struct sk_buff *skb)
2705 2706 2707 2708
{
	return skb->queue_mapping;
}

2709 2710 2711 2712 2713
static inline void skb_copy_queue_mapping(struct sk_buff *to, const struct sk_buff *from)
{
	to->queue_mapping = from->queue_mapping;
}

2714 2715 2716 2717 2718
static inline void skb_record_rx_queue(struct sk_buff *skb, u16 rx_queue)
{
	skb->queue_mapping = rx_queue + 1;
}

2719
static inline u16 skb_get_rx_queue(const struct sk_buff *skb)
2720 2721 2722 2723
{
	return skb->queue_mapping - 1;
}

2724
static inline bool skb_rx_queue_recorded(const struct sk_buff *skb)
2725
{
E
Eric Dumazet 已提交
2726
	return skb->queue_mapping != 0;
2727 2728
}

2729 2730 2731
extern u16 __skb_tx_hash(const struct net_device *dev,
			 const struct sk_buff *skb,
			 unsigned int num_tx_queues);
2732

2733 2734 2735 2736 2737 2738 2739 2740 2741 2742 2743 2744
#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

2745 2746 2747 2748 2749 2750 2751 2752 2753 2754 2755 2756 2757 2758 2759
/* Keeps track of mac header offset relative to skb->head.
 * It is useful for TSO of Tunneling protocol. e.g. GRE.
 * For non-tunnel skb it points to skb_mac_header() and for
 * tunnel skb it points to outer mac header. */
struct skb_gso_cb {
	int mac_offset;
};
#define SKB_GSO_CB(skb) ((struct skb_gso_cb *)(skb)->cb)

static inline int skb_tnl_header_len(const struct sk_buff *inner_skb)
{
	return (skb_mac_header(inner_skb) - inner_skb->head) -
		SKB_GSO_CB(inner_skb)->mac_offset;
}

2760
static inline bool skb_is_gso(const struct sk_buff *skb)
H
Herbert Xu 已提交
2761 2762 2763 2764
{
	return skb_shinfo(skb)->gso_size;
}

2765
static inline bool skb_is_gso_v6(const struct sk_buff *skb)
B
Brice Goglin 已提交
2766 2767 2768 2769
{
	return skb_shinfo(skb)->gso_type & SKB_GSO_TCPV6;
}

2770 2771 2772 2773 2774 2775
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. */
2776 2777
	const struct skb_shared_info *shinfo = skb_shinfo(skb);

2778 2779
	if (skb_is_nonlinear(skb) && shinfo->gso_size != 0 &&
	    unlikely(shinfo->gso_type == 0)) {
2780 2781 2782 2783 2784 2785
		__skb_warn_lro_forwarding(skb);
		return true;
	}
	return false;
}

2786 2787 2788 2789 2790 2791 2792
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;
}

2793 2794 2795 2796 2797 2798 2799 2800
/**
 * 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.
 */
2801
static inline void skb_checksum_none_assert(const struct sk_buff *skb)
2802 2803 2804 2805 2806 2807
{
#ifdef DEBUG
	BUG_ON(skb->ip_summed != CHECKSUM_NONE);
#endif
}

2808
bool skb_partial_csum_set(struct sk_buff *skb, u16 start, u16 off);
2809

2810 2811 2812 2813 2814 2815 2816 2817 2818 2819 2820 2821 2822
/**
 * skb_head_is_locked - Determine if the skb->head is locked down
 * @skb: skb to check
 *
 * The head on skbs build around a head frag can be removed if they are
 * not cloned.  This function returns true if the skb head is locked down
 * due to either being allocated via kmalloc, or by being a clone with
 * multiple references to the head.
 */
static inline bool skb_head_is_locked(const struct sk_buff *skb)
{
	return !skb->head_frag || skb_cloned(skb);
}
L
Linus Torvalds 已提交
2823 2824
#endif	/* __KERNEL__ */
#endif	/* _LINUX_SKBUFF_H */