skbuff.h 95.0 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
#include <linux/cache.h>
E
Eric Dumazet 已提交
23
#include <linux/rbtree.h>
24
#include <linux/socket.h>
L
Linus Torvalds 已提交
25

A
Arun Sharma 已提交
26
#include <linux/atomic.h>
L
Linus Torvalds 已提交
27 28 29
#include <asm/types.h>
#include <linux/spinlock.h>
#include <linux/net.h>
30
#include <linux/textsearch.h>
L
Linus Torvalds 已提交
31
#include <net/checksum.h>
32
#include <linux/rcupdate.h>
33
#include <linux/hrtimer.h>
34
#include <linux/dma-mapping.h>
35
#include <linux/netdev_features.h>
36
#include <linux/sched.h>
37
#include <net/flow_keys.h>
L
Linus Torvalds 已提交
38

39 40 41 42 43 44 45 46 47 48 49 50
/* A. Checksumming of received packets by device.
 *
 * CHECKSUM_NONE:
 *
 *   Device failed to checksum this packet e.g. due to lack of capabilities.
 *   The packet contains full (though not verified) checksum in packet but
 *   not in skb->csum. Thus, skb->csum is undefined in this case.
 *
 * CHECKSUM_UNNECESSARY:
 *
 *   The hardware you're dealing with doesn't calculate the full checksum
 *   (as in CHECKSUM_COMPLETE), but it does parse headers and verify checksums
51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73
 *   for specific protocols. For such packets it will set CHECKSUM_UNNECESSARY
 *   if their checksums are okay. skb->csum is still undefined in this case
 *   though. It is a bad option, but, unfortunately, nowadays most vendors do
 *   this. Apparently with the secret goal to sell you new devices, when you
 *   will add new protocol to your host, f.e. IPv6 8)
 *
 *   CHECKSUM_UNNECESSARY is applicable to following protocols:
 *     TCP: IPv6 and IPv4.
 *     UDP: IPv4 and IPv6. A device may apply CHECKSUM_UNNECESSARY to a
 *       zero UDP checksum for either IPv4 or IPv6, the networking stack
 *       may perform further validation in this case.
 *     GRE: only if the checksum is present in the header.
 *     SCTP: indicates the CRC in SCTP header has been validated.
 *
 *   skb->csum_level indicates the number of consecutive checksums found in
 *   the packet minus one that have been verified as CHECKSUM_UNNECESSARY.
 *   For instance if a device receives an IPv6->UDP->GRE->IPv4->TCP packet
 *   and a device is able to verify the checksums for UDP (possibly zero),
 *   GRE (checksum flag is set), and TCP-- skb->csum_level would be set to
 *   two. If the device were only able to verify the UDP checksum and not
 *   GRE, either because it doesn't support GRE checksum of because GRE
 *   checksum is bad, skb->csum_level would be set to zero (TCP checksum is
 *   not considered in this case).
74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127
 *
 * CHECKSUM_COMPLETE:
 *
 *   This is the most generic way. The device supplied checksum of the _whole_
 *   packet as seen by netif_rx() and fills out in skb->csum. Meaning, the
 *   hardware doesn't need to parse L3/L4 headers to implement this.
 *
 *   Note: Even if device supports only some protocols, but is able to produce
 *   skb->csum, it MUST use CHECKSUM_COMPLETE, not CHECKSUM_UNNECESSARY.
 *
 * CHECKSUM_PARTIAL:
 *
 *   This is identical to the case for output below. This may occur on a packet
 *   received directly from another Linux OS, e.g., a virtualized Linux kernel
 *   on the same host. The packet can be treated in the same way as
 *   CHECKSUM_UNNECESSARY, except that on output (i.e., forwarding) the
 *   checksum must be filled in by the OS or the hardware.
 *
 * B. Checksumming on output.
 *
 * CHECKSUM_NONE:
 *
 *   The skb was already checksummed by the protocol, or a checksum is not
 *   required.
 *
 * CHECKSUM_PARTIAL:
 *
 *   The device is required to checksum the packet as seen by hard_start_xmit()
 *   from skb->csum_start up to the end, and to record/write the checksum at
 *   offset skb->csum_start + skb->csum_offset.
 *
 *   The device must show its capabilities in dev->features, set up at device
 *   setup time, e.g. netdev_features.h:
 *
 *	NETIF_F_HW_CSUM	- It's a clever device, it's able to checksum everything.
 *	NETIF_F_IP_CSUM - Device is dumb, it's able to checksum only TCP/UDP over
 *			  IPv4. Sigh. Vendors like this way for an unknown reason.
 *			  Though, see comment above about CHECKSUM_UNNECESSARY. 8)
 *	NETIF_F_IPV6_CSUM - About as dumb as the last one but does IPv6 instead.
 *	NETIF_F_...     - Well, you get the picture.
 *
 * CHECKSUM_UNNECESSARY:
 *
 *   Normally, the device will do per protocol specific checksumming. Protocol
 *   implementations that do not want the NIC to perform the checksum
 *   calculation should use this flag in their outgoing skbs.
 *
 *	NETIF_F_FCOE_CRC - This indicates that the device can do FCoE FC CRC
 *			   offload. Correspondingly, the FCoE protocol driver
 *			   stack should use CHECKSUM_UNNECESSARY.
 *
 * Any questions? No questions, good.		--ANK
 */

128
/* Don't change this without changing skb_csum_unnecessary! */
129 130 131 132
#define CHECKSUM_NONE		0
#define CHECKSUM_UNNECESSARY	1
#define CHECKSUM_COMPLETE	2
#define CHECKSUM_PARTIAL	3
L
Linus Torvalds 已提交
133

134 135 136
/* Maximum value in skb->csum_level */
#define SKB_MAX_CSUM_LEVEL	3

137
#define SKB_DATA_ALIGN(X)	ALIGN(X, SMP_CACHE_BYTES)
138
#define SKB_WITH_OVERHEAD(X)	\
139
	((X) - SKB_DATA_ALIGN(sizeof(struct skb_shared_info)))
140 141
#define SKB_MAX_ORDER(X, ORDER) \
	SKB_WITH_OVERHEAD((PAGE_SIZE << (ORDER)) - (X))
L
Linus Torvalds 已提交
142 143 144
#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
145 146 147 148 149
/* 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
150
struct net_device;
151
struct scatterlist;
J
Jens Axboe 已提交
152
struct pipe_inode_info;
H
Herbert Xu 已提交
153
struct iov_iter;
L
Linus Torvalds 已提交
154

155
#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE)
L
Linus Torvalds 已提交
156 157 158
struct nf_conntrack {
	atomic_t use;
};
159
#endif
L
Linus Torvalds 已提交
160

161
#if IS_ENABLED(CONFIG_BRIDGE_NETFILTER)
L
Linus Torvalds 已提交
162
struct nf_bridge_info {
163 164 165 166 167
	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
168 169 170 171 172 173 174 175 176 177 178 179 180 181
};
#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;

182 183 184 185 186 187
/* 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.
188
 */
189
#if (65536/PAGE_SIZE + 1) < 16
190
#define MAX_SKB_FRAGS 16UL
191
#else
192
#define MAX_SKB_FRAGS (65536/PAGE_SIZE + 1)
193
#endif
L
Linus Torvalds 已提交
194 195 196 197

typedef struct skb_frag_struct skb_frag_t;

struct skb_frag_struct {
198 199 200
	struct {
		struct page *p;
	} page;
201
#if (BITS_PER_LONG > 32) || (PAGE_SIZE >= 65536)
202 203
	__u32 page_offset;
	__u32 size;
204 205 206 207
#else
	__u16 page_offset;
	__u16 size;
#endif
L
Linus Torvalds 已提交
208 209
};

E
Eric Dumazet 已提交
210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229
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;
}

230 231 232
#define HAVE_HW_TIME_STAMP

/**
233
 * struct skb_shared_hwtstamps - hardware time stamps
234 235 236 237
 * @hwtstamp:	hardware time stamp transformed into duration
 *		since arbitrary point in time
 *
 * Software time stamps generated by ktime_get_real() are stored in
238
 * skb->tstamp.
239 240 241 242 243 244 245 246 247 248 249
 *
 * 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;
};

250 251 252 253 254
/* Definitions for tx_flags in struct skb_shared_info */
enum {
	/* generate hardware time stamp */
	SKBTX_HW_TSTAMP = 1 << 0,

255
	/* generate software time stamp when queueing packet to NIC */
256 257 258 259 260
	SKBTX_SW_TSTAMP = 1 << 1,

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

261
	/* device driver supports TX zero-copy buffers */
E
Eric Dumazet 已提交
262
	SKBTX_DEV_ZEROCOPY = 1 << 3,
263 264

	/* generate wifi status information (where possible) */
E
Eric Dumazet 已提交
265
	SKBTX_WIFI_STATUS = 1 << 4,
266 267 268 269 270 271 272

	/* 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,
273 274 275

	/* generate software time stamp when entering packet scheduling */
	SKBTX_SCHED_TSTAMP = 1 << 6,
276 277 278

	/* generate software timestamp on peer data acknowledgment */
	SKBTX_ACK_TSTAMP = 1 << 7,
279 280
};

281 282 283
#define SKBTX_ANY_SW_TSTAMP	(SKBTX_SW_TSTAMP    | \
				 SKBTX_SCHED_TSTAMP | \
				 SKBTX_ACK_TSTAMP)
284 285
#define SKBTX_ANY_TSTAMP	(SKBTX_HW_TSTAMP | SKBTX_ANY_SW_TSTAMP)

286 287 288
/*
 * 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.
289 290
 * 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.
291 292
 * The ctx field is used to track device context.
 * The desc field is used to track userspace buffer index.
293 294
 */
struct ubuf_info {
295
	void (*callback)(struct ubuf_info *, bool zerocopy_success);
296
	void *ctx;
297
	unsigned long desc;
298 299
};

L
Linus Torvalds 已提交
300 301 302 303
/* This data is invariant across clones and lives at
 * the end of the header data, ie. at skb->end.
 */
struct skb_shared_info {
304 305
	unsigned char	nr_frags;
	__u8		tx_flags;
306 307 308 309
	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
310
	struct sk_buff	*frag_list;
311
	struct skb_shared_hwtstamps hwtstamps;
312
	u32		tskey;
313
	__be32          ip6_frag_id;
E
Eric Dumazet 已提交
314 315 316 317 318 319

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

J
Johann Baudy 已提交
320 321 322
	/* Intermediate layers must ensure that destructor_arg
	 * remains valid until skb destructor */
	void *		destructor_arg;
323

324 325
	/* must be last field, see pskb_expand_head() */
	skb_frag_t	frags[MAX_SKB_FRAGS];
L
Linus Torvalds 已提交
326 327 328 329
};

/* 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
330 331
 * 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
332 333 334 335 336 337 338 339 340 341
 *
 * 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)

342 343

enum {
344 345 346 347
	SKB_FCLONE_UNAVAILABLE,	/* skb has no fclone (from head_cache) */
	SKB_FCLONE_ORIG,	/* orig skb (from fclone_cache) */
	SKB_FCLONE_CLONE,	/* companion fclone skb (from fclone_cache) */
	SKB_FCLONE_FREE,	/* this companion fclone skb is available */
348 349
};

350 351
enum {
	SKB_GSO_TCPV4 = 1 << 0,
H
Herbert Xu 已提交
352
	SKB_GSO_UDP = 1 << 1,
353 354 355

	/* This indicates the skb is from an untrusted source. */
	SKB_GSO_DODGY = 1 << 2,
M
Michael Chan 已提交
356 357

	/* This indicates the tcp segment has CWR set. */
H
Herbert Xu 已提交
358 359 360
	SKB_GSO_TCP_ECN = 1 << 3,

	SKB_GSO_TCPV6 = 1 << 4,
361 362

	SKB_GSO_FCOE = 1 << 5,
363 364

	SKB_GSO_GRE = 1 << 6,
365

366
	SKB_GSO_GRE_CSUM = 1 << 7,
S
Simon Horman 已提交
367

368
	SKB_GSO_IPIP = 1 << 8,
E
Eric Dumazet 已提交
369

370
	SKB_GSO_SIT = 1 << 9,
E
Eric Dumazet 已提交
371

372
	SKB_GSO_UDP_TUNNEL = 1 << 10,
373 374

	SKB_GSO_UDP_TUNNEL_CSUM = 1 << 11,
T
Tom Herbert 已提交
375

P
Pravin B Shelar 已提交
376
	SKB_GSO_TUNNEL_REMCSUM = 1 << 12,
377 378
};

379 380 381 382 383 384 385 386 387 388
#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

389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439
/**
 * struct skb_mstamp - multi resolution time stamps
 * @stamp_us: timestamp in us resolution
 * @stamp_jiffies: timestamp in jiffies
 */
struct skb_mstamp {
	union {
		u64		v64;
		struct {
			u32	stamp_us;
			u32	stamp_jiffies;
		};
	};
};

/**
 * skb_mstamp_get - get current timestamp
 * @cl: place to store timestamps
 */
static inline void skb_mstamp_get(struct skb_mstamp *cl)
{
	u64 val = local_clock();

	do_div(val, NSEC_PER_USEC);
	cl->stamp_us = (u32)val;
	cl->stamp_jiffies = (u32)jiffies;
}

/**
 * skb_mstamp_delta - compute the difference in usec between two skb_mstamp
 * @t1: pointer to newest sample
 * @t0: pointer to oldest sample
 */
static inline u32 skb_mstamp_us_delta(const struct skb_mstamp *t1,
				      const struct skb_mstamp *t0)
{
	s32 delta_us = t1->stamp_us - t0->stamp_us;
	u32 delta_jiffies = t1->stamp_jiffies - t0->stamp_jiffies;

	/* If delta_us is negative, this might be because interval is too big,
	 * or local_clock() drift is too big : fallback using jiffies.
	 */
	if (delta_us <= 0 ||
	    delta_jiffies >= (INT_MAX / (USEC_PER_SEC / HZ)))

		delta_us = jiffies_to_usecs(delta_jiffies);

	return delta_us;
}


L
Linus Torvalds 已提交
440 441 442 443
/** 
 *	struct sk_buff - socket buffer
 *	@next: Next buffer in list
 *	@prev: Previous buffer in list
444
 *	@tstamp: Time we arrived/left
E
Eric Dumazet 已提交
445
 *	@rbnode: RB tree node, alternative to next/prev for netem/tcp
446
 *	@sk: Socket we are owned by
L
Linus Torvalds 已提交
447
 *	@dev: Device we arrived on/are leaving by
448
 *	@cb: Control buffer. Free for use by every layer. Put private vars here
E
Eric Dumazet 已提交
449
 *	@_skb_refdst: destination entry (with norefcount bit)
450
 *	@sp: the security path, used for xfrm
L
Linus Torvalds 已提交
451 452 453
 *	@len: Length of actual data
 *	@data_len: Data length
 *	@mac_len: Length of link layer header
454
 *	@hdr_len: writable header length of cloned skb
455 456 457
 *	@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
458
 *	@priority: Packet queueing priority
W
WANG Cong 已提交
459
 *	@ignore_df: allow local fragmentation
L
Linus Torvalds 已提交
460
 *	@cloned: Head may be cloned (check refcnt to be sure)
461
 *	@ip_summed: Driver fed us an IP checksum
L
Linus Torvalds 已提交
462
 *	@nohdr: Payload reference only, must not modify header
463
 *	@nfctinfo: Relationship of this skb to the connection
L
Linus Torvalds 已提交
464
 *	@pkt_type: Packet class
465 466
 *	@fclone: skbuff clone status
 *	@ipvs_property: skbuff is owned by ipvs
467 468
 *	@peeked: this packet has been seen already, so stats have been
 *		done for it, don't do them again
469
 *	@nf_trace: netfilter packet trace flag
470 471 472
 *	@protocol: Packet protocol from driver
 *	@destructor: Destruct function
 *	@nfct: Associated connection, if any
L
Linus Torvalds 已提交
473
 *	@nf_bridge: Saved data about a bridged frame - see br_netfilter.c
474
 *	@skb_iif: ifindex of device we arrived on
L
Linus Torvalds 已提交
475 476
 *	@tc_index: Traffic control index
 *	@tc_verd: traffic control verdict
477
 *	@hash: the packet hash
478
 *	@queue_mapping: Queue mapping for multiqueue devices
479
 *	@xmit_more: More SKBs are pending for this queue
480
 *	@ndisc_nodetype: router type (from link layer)
481
 *	@ooo_okay: allow the mapping of a socket to a queue to be changed
482
 *	@l4_hash: indicate hash is a canonical 4-tuple hash over transport
483
 *		ports.
484
 *	@sw_hash: indicates hash was computed in software stack
485 486
 *	@wifi_acked_valid: wifi_acked was set
 *	@wifi_acked: whether frame was acked on wifi or not
487
 *	@no_fcs:  Request NIC to treat last 4 bytes as Ethernet FCS
E
Eliezer Tamir 已提交
488
  *	@napi_id: id of the NAPI struct this skb came from
489
 *	@secmark: security marking
490 491
 *	@mark: Generic packet mark
 *	@dropcount: total number of sk_receive_queue overflows
492
 *	@vlan_proto: vlan encapsulation protocol
493
 *	@vlan_tci: vlan tag control information
S
Simon Horman 已提交
494
 *	@inner_protocol: Protocol (encapsulation)
495 496
 *	@inner_transport_header: Inner transport layer header (encapsulation)
 *	@inner_network_header: Network layer header (encapsulation)
497
 *	@inner_mac_header: Link layer header (encapsulation)
498 499 500 501 502 503 504 505 506
 *	@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
507 508 509
 */

struct sk_buff {
510
	union {
E
Eric Dumazet 已提交
511 512 513 514 515 516 517 518 519 520 521
		struct {
			/* These two members must be first. */
			struct sk_buff		*next;
			struct sk_buff		*prev;

			union {
				ktime_t		tstamp;
				struct skb_mstamp skb_mstamp;
			};
		};
		struct rb_node	rbnode; /* used in netem & tcp stack */
522
	};
523
	struct sock		*sk;
L
Linus Torvalds 已提交
524 525 526 527 528 529 530 531
	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.
	 */
532
	char			cb[48] __aligned(8);
L
Linus Torvalds 已提交
533

E
Eric Dumazet 已提交
534
	unsigned long		_skb_refdst;
535
	void			(*destructor)(struct sk_buff *skb);
536 537
#ifdef CONFIG_XFRM
	struct	sec_path	*sp;
538 539 540 541
#endif
#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE)
	struct nf_conntrack	*nfct;
#endif
542
#if IS_ENABLED(CONFIG_BRIDGE_NETFILTER)
543
	struct nf_bridge_info	*nf_bridge;
544
#endif
L
Linus Torvalds 已提交
545
	unsigned int		len,
546 547 548
				data_len;
	__u16			mac_len,
				hdr_len;
549 550 551 552

	/* Following fields are _not_ copied in __copy_skb_header()
	 * Note that queue_mapping is here mostly to fill a hole.
	 */
553
	kmemcheck_bitfield_begin(flags1);
554 555
	__u16			queue_mapping;
	__u8			cloned:1,
556
				nohdr:1,
557
				fclone:2,
558
				peeked:1,
559 560 561
				head_frag:1,
				xmit_more:1;
	/* one bit hole */
562
	kmemcheck_bitfield_end(flags1);
563

564 565 566
	/* fields enclosed in headers_start/headers_end are copied
	 * using a single memcpy() in __copy_skb_header()
	 */
567
	/* private: */
568
	__u32			headers_start[0];
569
	/* public: */
570

571 572 573 574 575
/* if you move pkt_type around you also must adapt those constants */
#ifdef __BIG_ENDIAN_BITFIELD
#define PKT_TYPE_MAX	(7 << 5)
#else
#define PKT_TYPE_MAX	7
L
Linus Torvalds 已提交
576
#endif
577
#define PKT_TYPE_OFFSET()	offsetof(struct sk_buff, __pkt_type_offset)
578

579
	__u8			__pkt_type_offset[0];
580
	__u8			pkt_type:3;
581
	__u8			pfmemalloc:1;
582 583 584 585 586
	__u8			ignore_df:1;
	__u8			nfctinfo:3;

	__u8			nf_trace:1;
	__u8			ip_summed:2;
587
	__u8			ooo_okay:1;
588
	__u8			l4_hash:1;
589
	__u8			sw_hash:1;
590 591
	__u8			wifi_acked_valid:1;
	__u8			wifi_acked:1;
592

593
	__u8			no_fcs:1;
594
	/* Indicates the inner headers are valid in the skbuff. */
595
	__u8			encapsulation:1;
596
	__u8			encap_hdr_csum:1;
597
	__u8			csum_valid:1;
598
	__u8			csum_complete_sw:1;
599 600
	__u8			csum_level:2;
	__u8			csum_bad:1;
601

602 603 604 605
#ifdef CONFIG_IPV6_NDISC_NODETYPE
	__u8			ndisc_nodetype:2;
#endif
	__u8			ipvs_property:1;
T
Tom Herbert 已提交
606
	__u8			inner_protocol_type:1;
607 608
	__u8			remcsum_offload:1;
	/* 3 or 5 bit hole */
609 610 611 612 613 614 615

#ifdef CONFIG_NET_SCHED
	__u16			tc_index;	/* traffic control index */
#ifdef CONFIG_NET_CLS_ACT
	__u16			tc_verd;	/* traffic control verdict */
#endif
#endif
616

617 618 619 620 621 622 623 624 625 626 627 628
	union {
		__wsum		csum;
		struct {
			__u16	csum_start;
			__u16	csum_offset;
		};
	};
	__u32			priority;
	int			skb_iif;
	__u32			hash;
	__be16			vlan_proto;
	__u16			vlan_tci;
D
Dan Williams 已提交
629 630
#ifdef CONFIG_NET_RX_BUSY_POLL
	unsigned int	napi_id;
631
#endif
632 633 634
#ifdef CONFIG_NETWORK_SECMARK
	__u32			secmark;
#endif
635 636 637
	union {
		__u32		mark;
		__u32		dropcount;
E
Eric Dumazet 已提交
638
		__u32		reserved_tailroom;
639
	};
L
Linus Torvalds 已提交
640

T
Tom Herbert 已提交
641 642 643 644 645
	union {
		__be16		inner_protocol;
		__u8		inner_ipproto;
	};

646 647 648
	__u16			inner_transport_header;
	__u16			inner_network_header;
	__u16			inner_mac_header;
649 650

	__be16			protocol;
651 652 653
	__u16			transport_header;
	__u16			network_header;
	__u16			mac_header;
654

655
	/* private: */
656
	__u32			headers_end[0];
657
	/* public: */
658

L
Linus Torvalds 已提交
659
	/* These elements must be at the end, see alloc_skb() for details.  */
660
	sk_buff_data_t		tail;
661
	sk_buff_data_t		end;
L
Linus Torvalds 已提交
662
	unsigned char		*head,
663
				*data;
664 665
	unsigned int		truesize;
	atomic_t		users;
L
Linus Torvalds 已提交
666 667 668 669 670 671 672 673 674
};

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


675 676 677 678 679 680 681 682 683
#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
684 685 686 687 688 689 690 691 692 693 694 695 696
/*
 * 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
697 698
static inline struct dst_entry *skb_dst(const struct sk_buff *skb)
{
E
Eric Dumazet 已提交
699 700 701 702 703 704 705
	/* 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
706 707
}

E
Eric Dumazet 已提交
708 709 710 711 712 713 714 715
/**
 * 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
716 717
static inline void skb_dst_set(struct sk_buff *skb, struct dst_entry *dst)
{
E
Eric Dumazet 已提交
718 719 720
	skb->_skb_refdst = (unsigned long)dst;
}

721 722
void __skb_dst_set_noref(struct sk_buff *skb, struct dst_entry *dst,
			 bool force);
723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754

/**
 * skb_dst_set_noref - sets skb dst, hopefully, without taking reference
 * @skb: buffer
 * @dst: dst entry
 *
 * Sets skb dst, assuming a reference was not taken on dst.
 * If dst entry is cached, we do not take reference and dst_release
 * will be avoided by refdst_drop. If dst entry is not cached, we take
 * reference, so that last dst_release can destroy the dst immediately.
 */
static inline void skb_dst_set_noref(struct sk_buff *skb, struct dst_entry *dst)
{
	__skb_dst_set_noref(skb, dst, false);
}

/**
 * skb_dst_set_noref_force - sets skb dst, without taking reference
 * @skb: buffer
 * @dst: dst entry
 *
 * Sets skb dst, assuming a reference was not taken on dst.
 * No reference is taken and no dst_release will be called. While for
 * cached dsts deferred reclaim is a basic feature, for entries that are
 * not cached it is caller's job to guarantee that last dst_release for
 * provided dst happens when nobody uses it, eg. after a RCU grace period.
 */
static inline void skb_dst_set_noref_force(struct sk_buff *skb,
					   struct dst_entry *dst)
{
	__skb_dst_set_noref(skb, dst, true);
}
E
Eric Dumazet 已提交
755 756

/**
L
Lucas De Marchi 已提交
757
 * skb_dst_is_noref - Test if skb dst isn't refcounted
E
Eric Dumazet 已提交
758 759 760 761 762
 * @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
763 764
}

E
Eric Dumazet 已提交
765 766
static inline struct rtable *skb_rtable(const struct sk_buff *skb)
{
E
Eric Dumazet 已提交
767
	return (struct rtable *)skb_dst(skb);
E
Eric Dumazet 已提交
768 769
}

770 771 772 773 774
void kfree_skb(struct sk_buff *skb);
void kfree_skb_list(struct sk_buff *segs);
void skb_tx_error(struct sk_buff *skb);
void consume_skb(struct sk_buff *skb);
void  __kfree_skb(struct sk_buff *skb);
775
extern struct kmem_cache *skbuff_head_cache;
E
Eric Dumazet 已提交
776

777 778 779
void kfree_skb_partial(struct sk_buff *skb, bool head_stolen);
bool skb_try_coalesce(struct sk_buff *to, struct sk_buff *from,
		      bool *fragstolen, int *delta_truesize);
E
Eric Dumazet 已提交
780

781 782 783
struct sk_buff *__alloc_skb(unsigned int size, gfp_t priority, int flags,
			    int node);
struct sk_buff *build_skb(void *data, unsigned int frag_size);
784
static inline struct sk_buff *alloc_skb(unsigned int size,
A
Al Viro 已提交
785
					gfp_t priority)
786
{
E
Eric Dumazet 已提交
787
	return __alloc_skb(size, priority, 0, NUMA_NO_NODE);
788 789
}

790 791 792 793 794 795
struct sk_buff *alloc_skb_with_frags(unsigned long header_len,
				     unsigned long data_len,
				     int max_page_order,
				     int *errcode,
				     gfp_t gfp_mask);

796 797 798 799 800 801 802 803 804 805 806 807 808 809
/* Layout of fast clones : [skb1][skb2][fclone_ref] */
struct sk_buff_fclones {
	struct sk_buff	skb1;

	struct sk_buff	skb2;

	atomic_t	fclone_ref;
};

/**
 *	skb_fclone_busy - check if fclone is busy
 *	@skb: buffer
 *
 * Returns true is skb is a fast clone, and its clone is not freed.
810 811
 * Some drivers call skb_orphan() in their ndo_start_xmit(),
 * so we also check that this didnt happen.
812
 */
813 814
static inline bool skb_fclone_busy(const struct sock *sk,
				   const struct sk_buff *skb)
815 816 817 818 819 820
{
	const struct sk_buff_fclones *fclones;

	fclones = container_of(skb, struct sk_buff_fclones, skb1);

	return skb->fclone == SKB_FCLONE_ORIG &&
821 822
	       fclones->skb2.fclone == SKB_FCLONE_CLONE &&
	       fclones->skb2.sk == sk;
823 824
}

825
static inline struct sk_buff *alloc_skb_fclone(unsigned int size,
A
Al Viro 已提交
826
					       gfp_t priority)
827
{
828
	return __alloc_skb(size, priority, SKB_ALLOC_FCLONE, NUMA_NO_NODE);
829 830
}

831
struct sk_buff *__alloc_skb_head(gfp_t priority, int node);
832 833 834 835 836
static inline struct sk_buff *alloc_skb_head(gfp_t priority)
{
	return __alloc_skb_head(priority, -1);
}

837 838 839 840
struct sk_buff *skb_morph(struct sk_buff *dst, struct sk_buff *src);
int skb_copy_ubufs(struct sk_buff *skb, gfp_t gfp_mask);
struct sk_buff *skb_clone(struct sk_buff *skb, gfp_t priority);
struct sk_buff *skb_copy(const struct sk_buff *skb, gfp_t priority);
841 842 843 844 845 846 847
struct sk_buff *__pskb_copy_fclone(struct sk_buff *skb, int headroom,
				   gfp_t gfp_mask, bool fclone);
static inline struct sk_buff *__pskb_copy(struct sk_buff *skb, int headroom,
					  gfp_t gfp_mask)
{
	return __pskb_copy_fclone(skb, headroom, gfp_mask, false);
}
848 849 850 851 852 853

int pskb_expand_head(struct sk_buff *skb, int nhead, int ntail, gfp_t gfp_mask);
struct sk_buff *skb_realloc_headroom(struct sk_buff *skb,
				     unsigned int headroom);
struct sk_buff *skb_copy_expand(const struct sk_buff *skb, int newheadroom,
				int newtailroom, gfp_t priority);
854 855
int skb_to_sgvec_nomark(struct sk_buff *skb, struct scatterlist *sg,
			int offset, int len);
856 857 858 859
int skb_to_sgvec(struct sk_buff *skb, struct scatterlist *sg, int offset,
		 int len);
int skb_cow_data(struct sk_buff *skb, int tailbits, struct sk_buff **trailer);
int skb_pad(struct sk_buff *skb, int pad);
860
#define dev_kfree_skb(a)	consume_skb(a)
L
Linus Torvalds 已提交
861

862 863 864 865
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);
866

E
Eric Dumazet 已提交
867
struct skb_seq_state {
868 869 870 871 872 873 874 875 876
	__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;
};

877 878 879 880 881
void skb_prepare_seq_read(struct sk_buff *skb, unsigned int from,
			  unsigned int to, struct skb_seq_state *st);
unsigned int skb_seq_read(unsigned int consumed, const u8 **data,
			  struct skb_seq_state *st);
void skb_abort_seq_read(struct skb_seq_state *st);
882

883 884 885
unsigned int skb_find_text(struct sk_buff *skb, unsigned int from,
			   unsigned int to, struct ts_config *config,
			   struct ts_state *state);
886

T
Tom Herbert 已提交
887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922
/*
 * Packet hash types specify the type of hash in skb_set_hash.
 *
 * Hash types refer to the protocol layer addresses which are used to
 * construct a packet's hash. The hashes are used to differentiate or identify
 * flows of the protocol layer for the hash type. Hash types are either
 * layer-2 (L2), layer-3 (L3), or layer-4 (L4).
 *
 * Properties of hashes:
 *
 * 1) Two packets in different flows have different hash values
 * 2) Two packets in the same flow should have the same hash value
 *
 * A hash at a higher layer is considered to be more specific. A driver should
 * set the most specific hash possible.
 *
 * A driver cannot indicate a more specific hash than the layer at which a hash
 * was computed. For instance an L3 hash cannot be set as an L4 hash.
 *
 * A driver may indicate a hash level which is less specific than the
 * actual layer the hash was computed on. For instance, a hash computed
 * at L4 may be considered an L3 hash. This should only be done if the
 * driver can't unambiguously determine that the HW computed the hash at
 * the higher layer. Note that the "should" in the second property above
 * permits this.
 */
enum pkt_hash_types {
	PKT_HASH_TYPE_NONE,	/* Undefined type */
	PKT_HASH_TYPE_L2,	/* Input: src_MAC, dest_MAC */
	PKT_HASH_TYPE_L3,	/* Input: src_IP, dst_IP */
	PKT_HASH_TYPE_L4,	/* Input: src_IP, dst_IP, src_port, dst_port */
};

static inline void
skb_set_hash(struct sk_buff *skb, __u32 hash, enum pkt_hash_types type)
{
923
	skb->l4_hash = (type == PKT_HASH_TYPE_L4);
924
	skb->sw_hash = 0;
925
	skb->hash = hash;
T
Tom Herbert 已提交
926 927
}

928 929
void __skb_get_hash(struct sk_buff *skb);
static inline __u32 skb_get_hash(struct sk_buff *skb)
930
{
931
	if (!skb->l4_hash && !skb->sw_hash)
932
		__skb_get_hash(skb);
933

934
	return skb->hash;
935 936
}

T
Tom Herbert 已提交
937 938
static inline __u32 skb_get_hash_raw(const struct sk_buff *skb)
{
939
	return skb->hash;
T
Tom Herbert 已提交
940 941
}

942 943
static inline void skb_clear_hash(struct sk_buff *skb)
{
944
	skb->hash = 0;
945
	skb->sw_hash = 0;
946
	skb->l4_hash = 0;
947 948 949 950
}

static inline void skb_clear_hash_if_not_l4(struct sk_buff *skb)
{
951
	if (!skb->l4_hash)
952 953 954
		skb_clear_hash(skb);
}

955 956
static inline void skb_copy_hash(struct sk_buff *to, const struct sk_buff *from)
{
957
	to->hash = from->hash;
958
	to->sw_hash = from->sw_hash;
959
	to->l4_hash = from->l4_hash;
960 961
};

962 963 964 965 966
#ifdef NET_SKBUFF_DATA_USES_OFFSET
static inline unsigned char *skb_end_pointer(const struct sk_buff *skb)
{
	return skb->head + skb->end;
}
967 968 969 970 971

static inline unsigned int skb_end_offset(const struct sk_buff *skb)
{
	return skb->end;
}
972 973 974 975 976
#else
static inline unsigned char *skb_end_pointer(const struct sk_buff *skb)
{
	return skb->end;
}
977 978 979 980 981

static inline unsigned int skb_end_offset(const struct sk_buff *skb)
{
	return skb->end - skb->head;
}
982 983
#endif

L
Linus Torvalds 已提交
984
/* Internal */
985
#define skb_shinfo(SKB)	((struct skb_shared_info *)(skb_end_pointer(SKB)))
L
Linus Torvalds 已提交
986

987 988 989 990 991
static inline struct skb_shared_hwtstamps *skb_hwtstamps(struct sk_buff *skb)
{
	return &skb_shinfo(skb)->hwtstamps;
}

L
Linus Torvalds 已提交
992 993 994 995 996 997 998 999
/**
 *	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)
{
1000
	return list->next == (const struct sk_buff *) list;
L
Linus Torvalds 已提交
1001 1002
}

D
David S. Miller 已提交
1003 1004 1005 1006 1007 1008 1009 1010 1011 1012
/**
 *	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)
{
1013
	return skb->next == (const struct sk_buff *) list;
D
David S. Miller 已提交
1014 1015
}

1016 1017 1018 1019 1020 1021 1022 1023 1024 1025
/**
 *	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)
{
1026
	return skb->prev == (const struct sk_buff *) list;
1027 1028
}

D
David S. Miller 已提交
1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046
/**
 *	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;
}

1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064
/**
 *	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
1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096
/**
 *	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;
}

1097 1098 1099 1100 1101 1102 1103 1104 1105 1106
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
1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132
/**
 *	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.
1133
 *	Note : Check if you can use __skb_header_release() instead.
L
Linus Torvalds 已提交
1134 1135 1136 1137 1138 1139 1140 1141
 */
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);
}

1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155
/**
 *	__skb_header_release - release reference to header
 *	@skb: buffer to operate on
 *
 *	Variant of skb_header_release() assuming skb is private to caller.
 *	We can avoid one atomic operation.
 */
static inline void __skb_header_release(struct sk_buff *skb)
{
	skb->nohdr = 1;
	atomic_set(&skb_shinfo(skb)->dataref, 1 + (1 << SKB_DATAREF_SHIFT));
}


L
Linus Torvalds 已提交
1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180
/**
 *	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.
 */
1181
static inline struct sk_buff *skb_share_check(struct sk_buff *skb, gfp_t pri)
L
Linus Torvalds 已提交
1182 1183 1184 1185
{
	might_sleep_if(pri & __GFP_WAIT);
	if (skb_shared(skb)) {
		struct sk_buff *nskb = skb_clone(skb, pri);
1186 1187 1188 1189 1190

		if (likely(nskb))
			consume_skb(skb);
		else
			kfree_skb(skb);
L
Linus Torvalds 已提交
1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215
		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.
 */
1216
static inline struct sk_buff *skb_unshare(struct sk_buff *skb,
A
Al Viro 已提交
1217
					  gfp_t pri)
L
Linus Torvalds 已提交
1218 1219 1220 1221
{
	might_sleep_if(pri & __GFP_WAIT);
	if (skb_cloned(skb)) {
		struct sk_buff *nskb = skb_copy(skb, pri);
1222 1223 1224 1225 1226 1227

		/* Free our shared copy */
		if (likely(nskb))
			consume_skb(skb);
		else
			kfree_skb(skb);
L
Linus Torvalds 已提交
1228 1229 1230 1231 1232 1233
		skb = nskb;
	}
	return skb;
}

/**
1234
 *	skb_peek - peek at the head of an &sk_buff_head
L
Linus Torvalds 已提交
1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245
 *	@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.
 */
1246
static inline struct sk_buff *skb_peek(const struct sk_buff_head *list_)
L
Linus Torvalds 已提交
1247
{
1248 1249 1250 1251 1252
	struct sk_buff *skb = list_->next;

	if (skb == (struct sk_buff *)list_)
		skb = NULL;
	return skb;
L
Linus Torvalds 已提交
1253 1254
}

P
Pavel Emelyanov 已提交
1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267
/**
 *	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;
1268

P
Pavel Emelyanov 已提交
1269 1270 1271 1272 1273
	if (next == (struct sk_buff *)list_)
		next = NULL;
	return next;
}

L
Linus Torvalds 已提交
1274
/**
1275
 *	skb_peek_tail - peek at the tail of an &sk_buff_head
L
Linus Torvalds 已提交
1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286
 *	@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.
 */
1287
static inline struct sk_buff *skb_peek_tail(const struct sk_buff_head *list_)
L
Linus Torvalds 已提交
1288
{
1289 1290 1291 1292 1293 1294
	struct sk_buff *skb = list_->prev;

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

L
Linus Torvalds 已提交
1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307
}

/**
 *	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;
}

1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323
/**
 *	__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;
}

1324 1325 1326 1327 1328 1329 1330 1331
/*
 * 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
1332 1333 1334
static inline void skb_queue_head_init(struct sk_buff_head *list)
{
	spin_lock_init(&list->lock);
1335
	__skb_queue_head_init(list);
L
Linus Torvalds 已提交
1336 1337
}

1338 1339 1340 1341 1342 1343 1344
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
1345
/*
1346
 *	Insert an sk_buff on a list.
L
Linus Torvalds 已提交
1347 1348 1349 1350
 *
 *	The "__skb_xxxx()" functions are the non-atomic ones that
 *	can only be called with interrupts disabled.
 */
1351 1352
void skb_insert(struct sk_buff *old, struct sk_buff *newsk,
		struct sk_buff_head *list);
1353 1354 1355 1356 1357 1358 1359 1360 1361
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
1362

1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386
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);
1387
		head->qlen += list->qlen;
1388 1389 1390 1391
	}
}

/**
E
Eric Dumazet 已提交
1392
 *	skb_queue_splice_init - join two skb lists and reinitialise the emptied list
1393 1394 1395 1396 1397 1398 1399 1400 1401 1402
 *	@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);
1403
		head->qlen += list->qlen;
1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417
		__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);
1418
		head->qlen += list->qlen;
1419 1420 1421 1422
	}
}

/**
E
Eric Dumazet 已提交
1423
 *	skb_queue_splice_tail_init - join two skb lists and reinitialise the emptied list
1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434
 *	@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);
1435
		head->qlen += list->qlen;
1436 1437 1438 1439
		__skb_queue_head_init(list);
	}
}

L
Linus Torvalds 已提交
1440
/**
1441
 *	__skb_queue_after - queue a buffer at the list head
L
Linus Torvalds 已提交
1442
 *	@list: list to use
1443
 *	@prev: place after this buffer
L
Linus Torvalds 已提交
1444 1445
 *	@newsk: buffer to queue
 *
1446
 *	Queue a buffer int the middle of a list. This function takes no locks
L
Linus Torvalds 已提交
1447 1448 1449 1450
 *	and you must therefore hold required locks before calling it.
 *
 *	A buffer cannot be placed on two lists at the same time.
 */
1451 1452 1453
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
1454
{
1455
	__skb_insert(newsk, prev, prev->next, list);
L
Linus Torvalds 已提交
1456 1457
}

1458 1459
void skb_append(struct sk_buff *old, struct sk_buff *newsk,
		struct sk_buff_head *list);
1460

1461 1462 1463 1464 1465 1466 1467
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);
}

1468 1469 1470 1471 1472 1473 1474 1475 1476 1477
/**
 *	__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.
 */
1478
void skb_queue_head(struct sk_buff_head *list, struct sk_buff *newsk);
1479 1480 1481 1482 1483 1484
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
1485 1486 1487 1488 1489 1490 1491 1492 1493 1494
/**
 *	__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.
 */
1495
void skb_queue_tail(struct sk_buff_head *list, struct sk_buff *newsk);
L
Linus Torvalds 已提交
1496 1497 1498
static inline void __skb_queue_tail(struct sk_buff_head *list,
				   struct sk_buff *newsk)
{
1499
	__skb_queue_before(list, (struct sk_buff *)list, newsk);
L
Linus Torvalds 已提交
1500 1501 1502 1503 1504 1505
}

/*
 * remove sk_buff from list. _Must_ be called atomically, and with
 * the list known..
 */
1506
void skb_unlink(struct sk_buff *skb, struct sk_buff_head *list);
L
Linus Torvalds 已提交
1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518
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;
}

1519 1520 1521 1522 1523 1524 1525 1526
/**
 *	__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.
 */
1527
struct sk_buff *skb_dequeue(struct sk_buff_head *list);
1528 1529 1530 1531 1532 1533 1534
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
1535 1536 1537 1538 1539 1540 1541 1542 1543

/**
 *	__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.
 */
1544
struct sk_buff *skb_dequeue_tail(struct sk_buff_head *list);
L
Linus Torvalds 已提交
1545 1546 1547 1548 1549 1550 1551 1552 1553
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;
}


1554
static inline bool skb_is_nonlinear(const struct sk_buff *skb)
L
Linus Torvalds 已提交
1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568
{
	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
1569
		len += skb_frag_size(&skb_shinfo(skb)->frags[i]);
L
Linus Torvalds 已提交
1570 1571 1572
	return len + skb_headlen(skb);
}

1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587
/**
 * __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
1588 1589 1590
{
	skb_frag_t *frag = &skb_shinfo(skb)->frags[i];

1591 1592 1593 1594 1595 1596 1597 1598 1599
	/*
	 * 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.
	 */
1600
	frag->page.p		  = page;
L
Linus Torvalds 已提交
1601
	frag->page_offset	  = off;
E
Eric Dumazet 已提交
1602
	skb_frag_size_set(frag, size);
1603 1604 1605 1606

	page = compound_head(page);
	if (page->pfmemalloc && !page->mapping)
		skb->pfmemalloc	= true;
1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617
}

/**
 * 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
M
Mathias Krause 已提交
1618
 * @skb to point to @size bytes at offset @off within @page. In
1619 1620 1621 1622 1623 1624 1625 1626
 * 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
1627 1628 1629
	skb_shinfo(skb)->nr_frags = i + 1;
}

1630 1631
void skb_add_rx_frag(struct sk_buff *skb, int i, struct page *page, int off,
		     int size, unsigned int truesize);
P
Peter Zijlstra 已提交
1632

J
Jason Wang 已提交
1633 1634 1635
void skb_coalesce_rx_frag(struct sk_buff *skb, int i, int size,
			  unsigned int truesize);

L
Linus Torvalds 已提交
1636
#define SKB_PAGE_ASSERT(skb) 	BUG_ON(skb_shinfo(skb)->nr_frags)
1637
#define SKB_FRAG_ASSERT(skb) 	BUG_ON(skb_has_frag_list(skb))
L
Linus Torvalds 已提交
1638 1639
#define SKB_LINEAR_ASSERT(skb)  BUG_ON(skb_is_nonlinear(skb))

1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655
#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;
}
1656

1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671
#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;
}
1672

1673 1674
#endif /* NET_SKBUFF_DATA_USES_OFFSET */

L
Linus Torvalds 已提交
1675 1676 1677
/*
 *	Add data to an sk_buff
 */
M
Mathias Krause 已提交
1678
unsigned char *pskb_put(struct sk_buff *skb, struct sk_buff *tail, int len);
1679
unsigned char *skb_put(struct sk_buff *skb, unsigned int len);
L
Linus Torvalds 已提交
1680 1681
static inline unsigned char *__skb_put(struct sk_buff *skb, unsigned int len)
{
1682
	unsigned char *tmp = skb_tail_pointer(skb);
L
Linus Torvalds 已提交
1683 1684 1685 1686 1687 1688
	SKB_LINEAR_ASSERT(skb);
	skb->tail += len;
	skb->len  += len;
	return tmp;
}

1689
unsigned char *skb_push(struct sk_buff *skb, unsigned int len);
L
Linus Torvalds 已提交
1690 1691 1692 1693 1694 1695 1696
static inline unsigned char *__skb_push(struct sk_buff *skb, unsigned int len)
{
	skb->data -= len;
	skb->len  += len;
	return skb->data;
}

1697
unsigned char *skb_pull(struct sk_buff *skb, unsigned int len);
L
Linus Torvalds 已提交
1698 1699 1700 1701 1702 1703 1704
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;
}

1705 1706 1707 1708 1709
static inline unsigned char *skb_pull_inline(struct sk_buff *skb, unsigned int len)
{
	return unlikely(len > skb->len) ? NULL : __skb_pull(skb, len);
}

1710
unsigned char *__pskb_pull_tail(struct sk_buff *skb, int delta);
L
Linus Torvalds 已提交
1711 1712 1713 1714

static inline unsigned char *__pskb_pull(struct sk_buff *skb, unsigned int len)
{
	if (len > skb_headlen(skb) &&
G
Gerrit Renker 已提交
1715
	    !__pskb_pull_tail(skb, len - skb_headlen(skb)))
L
Linus Torvalds 已提交
1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731
		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
1732
	return __pskb_pull_tail(skb, len - skb_headlen(skb)) != NULL;
L
Linus Torvalds 已提交
1733 1734 1735 1736 1737 1738 1739 1740
}

/**
 *	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.
 */
1741
static inline unsigned int skb_headroom(const struct sk_buff *skb)
L
Linus Torvalds 已提交
1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753
{
	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)
{
1754
	return skb_is_nonlinear(skb) ? 0 : skb->end - skb->tail;
L
Linus Torvalds 已提交
1755 1756
}

1757 1758 1759 1760 1761 1762 1763 1764 1765
/**
 *	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
1766 1767 1768 1769
	if (skb_is_nonlinear(skb))
		return 0;

	return skb->end - skb->tail - skb->reserved_tailroom;
1770 1771
}

L
Linus Torvalds 已提交
1772 1773 1774 1775 1776 1777 1778 1779
/**
 *	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.
 */
1780
static inline void skb_reserve(struct sk_buff *skb, int len)
L
Linus Torvalds 已提交
1781 1782 1783 1784 1785
{
	skb->data += len;
	skb->tail += len;
}

T
Tom Herbert 已提交
1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802
#define ENCAP_TYPE_ETHER	0
#define ENCAP_TYPE_IPPROTO	1

static inline void skb_set_inner_protocol(struct sk_buff *skb,
					  __be16 protocol)
{
	skb->inner_protocol = protocol;
	skb->inner_protocol_type = ENCAP_TYPE_ETHER;
}

static inline void skb_set_inner_ipproto(struct sk_buff *skb,
					 __u8 ipproto)
{
	skb->inner_ipproto = ipproto;
	skb->inner_protocol_type = ENCAP_TYPE_IPPROTO;
}

1803 1804
static inline void skb_reset_inner_headers(struct sk_buff *skb)
{
1805
	skb->inner_mac_header = skb->mac_header;
1806 1807 1808 1809
	skb->inner_network_header = skb->network_header;
	skb->inner_transport_header = skb->transport_header;
}

1810 1811 1812 1813 1814
static inline void skb_reset_mac_len(struct sk_buff *skb)
{
	skb->mac_len = skb->network_header - skb->mac_header;
}

1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849
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;
}

1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865
static inline unsigned char *skb_inner_mac_header(const struct sk_buff *skb)
{
	return skb->head + skb->inner_mac_header;
}

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

static inline void skb_set_inner_mac_header(struct sk_buff *skb,
					    const int offset)
{
	skb_reset_inner_mac_header(skb);
	skb->inner_mac_header += offset;
}
1866 1867
static inline bool skb_transport_header_was_set(const struct sk_buff *skb)
{
C
Cong Wang 已提交
1868
	return skb->transport_header != (typeof(skb->transport_header))~0U;
1869 1870
}

1871 1872
static inline unsigned char *skb_transport_header(const struct sk_buff *skb)
{
1873
	return skb->head + skb->transport_header;
1874 1875
}

1876 1877
static inline void skb_reset_transport_header(struct sk_buff *skb)
{
1878
	skb->transport_header = skb->data - skb->head;
1879 1880
}

1881 1882 1883
static inline void skb_set_transport_header(struct sk_buff *skb,
					    const int offset)
{
1884 1885
	skb_reset_transport_header(skb);
	skb->transport_header += offset;
1886 1887
}

1888 1889
static inline unsigned char *skb_network_header(const struct sk_buff *skb)
{
1890
	return skb->head + skb->network_header;
1891 1892
}

1893 1894
static inline void skb_reset_network_header(struct sk_buff *skb)
{
1895
	skb->network_header = skb->data - skb->head;
1896 1897
}

1898 1899
static inline void skb_set_network_header(struct sk_buff *skb, const int offset)
{
1900 1901
	skb_reset_network_header(skb);
	skb->network_header += offset;
1902 1903
}

1904
static inline unsigned char *skb_mac_header(const struct sk_buff *skb)
1905
{
1906
	return skb->head + skb->mac_header;
1907 1908
}

1909
static inline int skb_mac_header_was_set(const struct sk_buff *skb)
1910
{
C
Cong Wang 已提交
1911
	return skb->mac_header != (typeof(skb->mac_header))~0U;
1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924
}

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;
}

1925 1926 1927 1928 1929
static inline void skb_pop_mac_header(struct sk_buff *skb)
{
	skb->mac_header = skb->network_header;
}

1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942
static inline void skb_probe_transport_header(struct sk_buff *skb,
					      const int offset_hint)
{
	struct flow_keys keys;

	if (skb_transport_header_was_set(skb))
		return;
	else if (skb_flow_dissect(skb, &keys))
		skb_set_transport_header(skb, keys.thoff);
	else
		skb_set_transport_header(skb, offset_hint);
}

1943 1944 1945 1946 1947 1948 1949 1950 1951 1952
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);
	}
}

1953 1954 1955 1956 1957
static inline int skb_checksum_start_offset(const struct sk_buff *skb)
{
	return skb->csum_start - skb_headroom(skb);
}

1958 1959 1960 1961 1962 1963 1964 1965 1966 1967
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;
}

1968 1969 1970 1971 1972
static inline u32 skb_inner_network_header_len(const struct sk_buff *skb)
{
	return skb->inner_transport_header - skb->inner_network_header;
}

1973 1974 1975 1976
static inline int skb_network_offset(const struct sk_buff *skb)
{
	return skb_network_header(skb) - skb->data;
}
1977

1978 1979 1980 1981 1982
static inline int skb_inner_network_offset(const struct sk_buff *skb)
{
	return skb_inner_network_header(skb) - skb->data;
}

1983 1984 1985 1986 1987
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
1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998
/*
 * 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:
 *
1999
 * skb_reserve(skb, NET_IP_ALIGN);
L
Linus Torvalds 已提交
2000 2001 2002 2003
 *
 * 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.
2004
 *
L
Linus Torvalds 已提交
2005 2006 2007 2008 2009 2010 2011
 * 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

2012 2013 2014 2015
/*
 * 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
2016
 * 32 bytes or less we avoid the reallocation.
2017 2018 2019 2020 2021 2022 2023
 *
 * 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.
 *
2024
 * Various parts of the networking layer expect at least 32 bytes of
2025
 * headroom, you should not reduce this.
2026 2027 2028 2029
 *
 * 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 :
2030
 * NET_IP_ALIGN(2) + ethernet_header(14) + IP_header(20/40) + ports(8)
2031 2032
 */
#ifndef NET_SKB_PAD
2033
#define NET_SKB_PAD	max(32, L1_CACHE_BYTES)
2034 2035
#endif

2036
int ___pskb_trim(struct sk_buff *skb, unsigned int len);
L
Linus Torvalds 已提交
2037 2038 2039

static inline void __skb_trim(struct sk_buff *skb, unsigned int len)
{
2040
	if (unlikely(skb_is_nonlinear(skb))) {
2041 2042 2043
		WARN_ON(1);
		return;
	}
2044 2045
	skb->len = len;
	skb_set_tail_pointer(skb, len);
L
Linus Torvalds 已提交
2046 2047
}

2048
void skb_trim(struct sk_buff *skb, unsigned int len);
L
Linus Torvalds 已提交
2049 2050 2051

static inline int __pskb_trim(struct sk_buff *skb, unsigned int len)
{
2052 2053 2054 2055
	if (skb->data_len)
		return ___pskb_trim(skb, len);
	__skb_trim(skb, len);
	return 0;
L
Linus Torvalds 已提交
2056 2057 2058 2059 2060 2061 2062
}

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

2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077
/**
 *	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
2078 2079 2080 2081 2082 2083 2084 2085 2086 2087
/**
 *	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)
{
E
Eric Dumazet 已提交
2088
	if (skb->destructor) {
L
Linus Torvalds 已提交
2089
		skb->destructor(skb);
E
Eric Dumazet 已提交
2090 2091
		skb->destructor = NULL;
		skb->sk		= NULL;
2092 2093
	} else {
		BUG_ON(skb->sk);
E
Eric Dumazet 已提交
2094
	}
L
Linus Torvalds 已提交
2095 2096
}

2097 2098 2099 2100 2101 2102 2103 2104 2105 2106 2107 2108 2109 2110 2111 2112
/**
 *	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
2113 2114 2115 2116 2117 2118 2119 2120
/**
 *	__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.
 */
2121
void skb_queue_purge(struct sk_buff_head *list);
L
Linus Torvalds 已提交
2122 2123 2124 2125 2126 2127 2128
static inline void __skb_queue_purge(struct sk_buff_head *list)
{
	struct sk_buff *skb;
	while ((skb = __skb_dequeue(list)) != NULL)
		kfree_skb(skb);
}

2129 2130 2131 2132
#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

2133
void *netdev_alloc_frag(unsigned int fragsz);
L
Linus Torvalds 已提交
2134

2135 2136
struct sk_buff *__netdev_alloc_skb(struct net_device *dev, unsigned int length,
				   gfp_t gfp_mask);
2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148 2149 2150 2151

/**
 *	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,
2152
					       unsigned int length)
2153 2154 2155 2156
{
	return __netdev_alloc_skb(dev, length, GFP_ATOMIC);
}

2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170
/* 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);
}


2171 2172
static inline struct sk_buff *__netdev_alloc_skb_ip_align(struct net_device *dev,
		unsigned int length, gfp_t gfp)
2173
{
2174
	struct sk_buff *skb = __netdev_alloc_skb(dev, length + NET_IP_ALIGN, gfp);
2175 2176 2177 2178 2179 2180

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

2181 2182 2183 2184 2185 2186
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);
}

2187 2188 2189 2190 2191 2192 2193 2194 2195 2196 2197 2198 2199 2200 2201 2202 2203 2204 2205 2206 2207 2208 2209 2210 2211 2212 2213 2214 2215 2216 2217 2218 2219 2220 2221 2222 2223 2224 2225 2226 2227 2228 2229 2230 2231 2232 2233 2234
/**
 * __dev_alloc_pages - allocate page for network Rx
 * @gfp_mask: allocation priority. Set __GFP_NOMEMALLOC if not for network Rx
 * @order: size of the allocation
 *
 * Allocate a new page.
 *
 * %NULL is returned if there is no free memory.
*/
static inline struct page *__dev_alloc_pages(gfp_t gfp_mask,
					     unsigned int order)
{
	/* This piece of code contains several assumptions.
	 * 1.  This is for device Rx, therefor a cold page is preferred.
	 * 2.  The expectation is the user wants a compound page.
	 * 3.  If requesting a order 0 page it will not be compound
	 *     due to the check to see if order has a value in prep_new_page
	 * 4.  __GFP_MEMALLOC is ignored if __GFP_NOMEMALLOC is set due to
	 *     code in gfp_to_alloc_flags that should be enforcing this.
	 */
	gfp_mask |= __GFP_COLD | __GFP_COMP | __GFP_MEMALLOC;

	return alloc_pages_node(NUMA_NO_NODE, gfp_mask, order);
}

static inline struct page *dev_alloc_pages(unsigned int order)
{
	return __dev_alloc_pages(GFP_ATOMIC, order);
}

/**
 * __dev_alloc_page - allocate a page for network Rx
 * @gfp_mask: allocation priority. Set __GFP_NOMEMALLOC if not for network Rx
 *
 * Allocate a new page.
 *
 * %NULL is returned if there is no free memory.
 */
static inline struct page *__dev_alloc_page(gfp_t gfp_mask)
{
	return __dev_alloc_pages(gfp_mask, 0);
}

static inline struct page *dev_alloc_page(void)
{
	return __dev_alloc_page(GFP_ATOMIC);
}

2235 2236 2237 2238 2239 2240 2241 2242 2243 2244 2245 2246
/**
 *	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;
}

2247
/**
2248
 * skb_frag_page - retrieve the page referred to by a paged fragment
2249 2250 2251 2252 2253 2254
 * @frag: the paged fragment
 *
 * Returns the &struct page associated with @frag.
 */
static inline struct page *skb_frag_page(const skb_frag_t *frag)
{
2255
	return frag->page.p;
2256 2257 2258 2259 2260 2261 2262 2263 2264 2265 2266 2267 2268 2269 2270 2271 2272 2273 2274 2275 2276 2277 2278 2279 2280 2281 2282 2283 2284 2285 2286 2287 2288 2289 2290 2291 2292 2293 2294 2295 2296 2297 2298 2299 2300 2301 2302 2303 2304 2305 2306 2307 2308 2309 2310 2311 2312 2313 2314 2315 2316 2317 2318 2319 2320 2321 2322 2323 2324 2325 2326 2327 2328 2329 2330 2331 2332 2333 2334 2335 2336 2337 2338 2339 2340
}

/**
 * __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)
{
2341
	frag->page.p = page;
2342 2343 2344 2345 2346 2347 2348 2349 2350 2351 2352 2353 2354 2355 2356 2357
}

/**
 * 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);
}

E
Eric Dumazet 已提交
2358 2359
bool skb_page_frag_refill(unsigned int sz, struct page_frag *pfrag, gfp_t prio);

2360 2361
/**
 * skb_frag_dma_map - maps a paged fragment via the DMA API
2362
 * @dev: the device to map the fragment to
2363 2364 2365 2366
 * @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
2367
 * @dir: the direction of the mapping (%PCI_DMA_*)
2368 2369 2370 2371 2372 2373 2374 2375 2376 2377 2378 2379
 *
 * 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
2380 2381 2382 2383 2384 2385
static inline struct sk_buff *pskb_copy(struct sk_buff *skb,
					gfp_t gfp_mask)
{
	return __pskb_copy(skb, skb_headroom(skb), gfp_mask);
}

2386 2387 2388 2389 2390 2391 2392 2393

static inline struct sk_buff *pskb_copy_for_clone(struct sk_buff *skb,
						  gfp_t gfp_mask)
{
	return __pskb_copy_fclone(skb, skb_headroom(skb), gfp_mask, true);
}


2394 2395 2396 2397 2398 2399 2400 2401
/**
 *	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.
 */
2402
static inline int skb_clone_writable(const struct sk_buff *skb, unsigned int len)
2403 2404 2405 2406 2407
{
	return !skb_header_cloned(skb) &&
	       skb_headroom(skb) + len <= skb->hdr_len;
}

H
Herbert Xu 已提交
2408 2409 2410 2411 2412 2413 2414 2415 2416 2417 2418 2419 2420 2421
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
2422 2423 2424 2425 2426 2427 2428 2429 2430 2431 2432 2433 2434 2435
/**
 *	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
2436 2437
	return __skb_cow(skb, headroom, skb_cloned(skb));
}
L
Linus Torvalds 已提交
2438

H
Herbert Xu 已提交
2439 2440 2441 2442 2443 2444 2445 2446 2447 2448 2449 2450 2451
/**
 *	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
2452 2453 2454 2455 2456 2457 2458 2459 2460
}

/**
 *	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
2461 2462
 *	is untouched. Otherwise it is extended. Returns zero on
 *	success. The skb is freed on error.
L
Linus Torvalds 已提交
2463 2464
 */
 
2465
static inline int skb_padto(struct sk_buff *skb, unsigned int len)
L
Linus Torvalds 已提交
2466 2467 2468
{
	unsigned int size = skb->len;
	if (likely(size >= len))
2469
		return 0;
G
Gerrit Renker 已提交
2470
	return skb_pad(skb, len - size);
L
Linus Torvalds 已提交
2471 2472 2473 2474 2475 2476 2477 2478 2479
}

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;
2480
		__wsum csum = csum_and_copy_from_user(from, skb_put(skb, copy),
L
Linus Torvalds 已提交
2481 2482 2483 2484 2485 2486 2487 2488 2489 2490 2491 2492
							    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;
}

2493 2494
static inline bool skb_can_coalesce(struct sk_buff *skb, int i,
				    const struct page *page, int off)
L
Linus Torvalds 已提交
2495 2496
{
	if (i) {
E
Eric Dumazet 已提交
2497
		const struct skb_frag_struct *frag = &skb_shinfo(skb)->frags[i - 1];
L
Linus Torvalds 已提交
2498

2499
		return page == skb_frag_page(frag) &&
E
Eric Dumazet 已提交
2500
		       off == frag->page_offset + skb_frag_size(frag);
L
Linus Torvalds 已提交
2501
	}
2502
	return false;
L
Linus Torvalds 已提交
2503 2504
}

H
Herbert Xu 已提交
2505 2506 2507 2508 2509
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
2510 2511 2512 2513 2514 2515 2516
/**
 *	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
2517 2518 2519 2520 2521
static inline int skb_linearize(struct sk_buff *skb)
{
	return skb_is_nonlinear(skb) ? __skb_linearize(skb) : 0;
}

2522 2523 2524 2525 2526 2527 2528 2529 2530
/**
 * 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)
{
2531 2532
	return skb_is_nonlinear(skb) &&
	       skb_shinfo(skb)->tx_flags & SKBTX_SHARED_FRAG;
2533 2534
}

H
Herbert Xu 已提交
2535 2536 2537 2538 2539 2540 2541 2542
/**
 *	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
2543
{
H
Herbert Xu 已提交
2544 2545
	return skb_is_nonlinear(skb) || skb_cloned(skb) ?
	       __skb_linearize(skb) : 0;
L
Linus Torvalds 已提交
2546 2547 2548 2549 2550 2551 2552 2553 2554
}

/**
 *	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
2555 2556
 *	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
2557 2558 2559
 */

static inline void skb_postpull_rcsum(struct sk_buff *skb,
2560
				      const void *start, unsigned int len)
L
Linus Torvalds 已提交
2561
{
2562
	if (skb->ip_summed == CHECKSUM_COMPLETE)
L
Linus Torvalds 已提交
2563 2564 2565
		skb->csum = csum_sub(skb->csum, csum_partial(start, len, 0));
}

2566 2567
unsigned char *skb_pull_rcsum(struct sk_buff *skb, unsigned int len);

2568 2569 2570 2571 2572 2573 2574 2575 2576 2577 2578 2579 2580 2581 2582 2583 2584 2585
/**
 *	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)
{
	if (likely(len >= skb->len))
		return 0;
	if (skb->ip_summed == CHECKSUM_COMPLETE)
		skb->ip_summed = CHECKSUM_NONE;
	return __pskb_trim(skb, len);
}

L
Linus Torvalds 已提交
2586 2587
#define skb_queue_walk(queue, skb) \
		for (skb = (queue)->next;					\
2588
		     skb != (struct sk_buff *)(queue);				\
L
Linus Torvalds 已提交
2589 2590
		     skb = skb->next)

2591 2592 2593 2594 2595
#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)

2596
#define skb_queue_walk_from(queue, skb)						\
2597
		for (; skb != (struct sk_buff *)(queue);			\
2598 2599 2600 2601 2602 2603 2604
		     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)

2605 2606
#define skb_queue_reverse_walk(queue, skb) \
		for (skb = (queue)->prev;					\
2607
		     skb != (struct sk_buff *)(queue);				\
2608 2609
		     skb = skb->prev)

2610 2611 2612 2613 2614 2615 2616 2617 2618
#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
2619

2620
static inline bool skb_has_frag_list(const struct sk_buff *skb)
2621 2622 2623 2624 2625 2626 2627 2628 2629 2630 2631 2632 2633 2634 2635 2636 2637 2638
{
	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)

2639 2640 2641 2642 2643 2644 2645 2646
struct sk_buff *__skb_recv_datagram(struct sock *sk, unsigned flags,
				    int *peeked, int *off, int *err);
struct sk_buff *skb_recv_datagram(struct sock *sk, unsigned flags, int noblock,
				  int *err);
unsigned int datagram_poll(struct file *file, struct socket *sock,
			   struct poll_table_struct *wait);
int skb_copy_datagram_iovec(const struct sk_buff *from, int offset,
			    struct iovec *to, int size);
2647 2648 2649 2650 2651
static inline int skb_copy_datagram_msg(const struct sk_buff *from, int offset,
					struct msghdr *msg, int size)
{
	return skb_copy_datagram_iovec(from, offset, msg->msg_iov, size);
}
2652 2653
int skb_copy_and_csum_datagram_iovec(struct sk_buff *skb, int hlen,
				     struct iovec *iov);
2654 2655 2656 2657 2658
static inline int skb_copy_and_csum_datagram_msg(struct sk_buff *skb, int hlen,
			    struct msghdr *msg)
{
	return skb_copy_and_csum_datagram_iovec(skb, hlen, msg->msg_iov);
}
2659 2660
int skb_copy_datagram_from_iter(struct sk_buff *skb, int offset,
				 struct iov_iter *from, int len);
H
Herbert Xu 已提交
2661 2662
int skb_copy_datagram_iter(const struct sk_buff *from, int offset,
			   struct iov_iter *to, int size);
2663
int zerocopy_sg_from_iter(struct sk_buff *skb, struct iov_iter *frm);
2664 2665 2666 2667 2668 2669 2670 2671 2672 2673 2674
void skb_free_datagram(struct sock *sk, struct sk_buff *skb);
void skb_free_datagram_locked(struct sock *sk, struct sk_buff *skb);
int skb_kill_datagram(struct sock *sk, struct sk_buff *skb, unsigned int flags);
int skb_copy_bits(const struct sk_buff *skb, int offset, void *to, int len);
int skb_store_bits(struct sk_buff *skb, int offset, const void *from, int len);
__wsum skb_copy_and_csum_bits(const struct sk_buff *skb, int offset, u8 *to,
			      int len, __wsum csum);
int skb_splice_bits(struct sk_buff *skb, unsigned int offset,
		    struct pipe_inode_info *pipe, unsigned int len,
		    unsigned int flags);
void skb_copy_and_csum_dev(const struct sk_buff *skb, u8 *to);
2675
unsigned int skb_zerocopy_headlen(const struct sk_buff *from);
2676 2677
int skb_zerocopy(struct sk_buff *to, struct sk_buff *from,
		 int len, int hlen);
2678 2679 2680
void skb_split(struct sk_buff *skb, struct sk_buff *skb1, const u32 len);
int skb_shift(struct sk_buff *tgt, struct sk_buff *skb, int shiftlen);
void skb_scrub_packet(struct sk_buff *skb, bool xnet);
2681
unsigned int skb_gso_transport_seglen(const struct sk_buff *skb);
2682
struct sk_buff *skb_segment(struct sk_buff *skb, netdev_features_t features);
2683
struct sk_buff *skb_vlan_untag(struct sk_buff *skb);
2684
int skb_ensure_writable(struct sk_buff *skb, int write_len);
2685 2686
int skb_vlan_pop(struct sk_buff *skb);
int skb_vlan_push(struct sk_buff *skb, __be16 vlan_proto, u16 vlan_tci);
2687

A
Al Viro 已提交
2688 2689 2690 2691 2692
static inline int memcpy_from_msg(void *data, struct msghdr *msg, int len)
{
	return memcpy_fromiovec(data, msg->msg_iov, len);
}

A
Al Viro 已提交
2693 2694 2695 2696 2697
static inline int memcpy_to_msg(struct msghdr *msg, void *data, int len)
{
	return memcpy_toiovec(msg->msg_iov, data, len);
}

2698 2699 2700 2701 2702 2703 2704 2705 2706 2707
struct skb_checksum_ops {
	__wsum (*update)(const void *mem, int len, __wsum wsum);
	__wsum (*combine)(__wsum csum, __wsum csum2, int offset, int len);
};

__wsum __skb_checksum(const struct sk_buff *skb, int offset, int len,
		      __wsum csum, const struct skb_checksum_ops *ops);
__wsum skb_checksum(const struct sk_buff *skb, int offset, int len,
		    __wsum csum);

2708 2709
static inline void *__skb_header_pointer(const struct sk_buff *skb, int offset,
					 int len, void *data, int hlen, void *buffer)
L
Linus Torvalds 已提交
2710
{
2711
	if (hlen - offset >= len)
2712
		return data + offset;
L
Linus Torvalds 已提交
2713

2714 2715
	if (!skb ||
	    skb_copy_bits(skb, offset, buffer, len) < 0)
L
Linus Torvalds 已提交
2716 2717 2718 2719 2720
		return NULL;

	return buffer;
}

2721 2722 2723 2724 2725 2726 2727
static inline void *skb_header_pointer(const struct sk_buff *skb, int offset,
				       int len, void *buffer)
{
	return __skb_header_pointer(skb, offset, len, skb->data,
				    skb_headlen(skb), buffer);
}

2728 2729 2730 2731 2732 2733 2734 2735 2736 2737 2738 2739 2740 2741 2742 2743 2744 2745
/**
 *	skb_needs_linearize - check if we need to linearize a given skb
 *			      depending on the given device features.
 *	@skb: socket buffer to check
 *	@features: net device features
 *
 *	Returns true if either:
 *	1. skb has frag_list and the device doesn't support FRAGLIST, or
 *	2. skb is fragmented and the device does not support SG.
 */
static inline bool skb_needs_linearize(struct sk_buff *skb,
				       netdev_features_t features)
{
	return skb_is_nonlinear(skb) &&
	       ((skb_has_frag_list(skb) && !(features & NETIF_F_FRAGLIST)) ||
		(skb_shinfo(skb)->nr_frags && !(features & NETIF_F_SG)));
}

2746 2747 2748 2749 2750 2751 2752 2753 2754 2755 2756 2757 2758 2759
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);
}

2760 2761 2762 2763 2764 2765 2766 2767 2768 2769 2770 2771 2772 2773 2774
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);
}

2775
void skb_init(void);
L
Linus Torvalds 已提交
2776

2777 2778 2779 2780 2781
static inline ktime_t skb_get_ktime(const struct sk_buff *skb)
{
	return skb->tstamp;
}

2782 2783 2784 2785 2786 2787 2788 2789 2790
/**
 *	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.
 */
2791 2792
static inline void skb_get_timestamp(const struct sk_buff *skb,
				     struct timeval *stamp)
2793
{
2794
	*stamp = ktime_to_timeval(skb->tstamp);
2795 2796
}

2797 2798 2799 2800 2801 2802
static inline void skb_get_timestampns(const struct sk_buff *skb,
				       struct timespec *stamp)
{
	*stamp = ktime_to_timespec(skb->tstamp);
}

2803
static inline void __net_timestamp(struct sk_buff *skb)
2804
{
2805
	skb->tstamp = ktime_get_real();
2806 2807
}

2808 2809 2810 2811 2812
static inline ktime_t net_timedelta(ktime_t t)
{
	return ktime_sub(ktime_get_real(), t);
}

2813 2814 2815 2816
static inline ktime_t net_invalid_timestamp(void)
{
	return ktime_set(0, 0);
}
2817

2818 2819
struct sk_buff *skb_clone_sk(struct sk_buff *skb);

2820 2821
#ifdef CONFIG_NETWORK_PHY_TIMESTAMPING

2822 2823
void skb_clone_tx_timestamp(struct sk_buff *skb);
bool skb_defer_rx_timestamp(struct sk_buff *skb);
2824 2825 2826 2827 2828 2829 2830 2831 2832 2833 2834 2835 2836 2837 2838 2839 2840

#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
 *
2841 2842 2843 2844 2845
 * 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.
 *
2846
 * @skb: clone of the the original outgoing packet
2847
 * @hwtstamps: hardware time stamps, may be NULL if not available
2848 2849 2850 2851 2852
 *
 */
void skb_complete_tx_timestamp(struct sk_buff *skb,
			       struct skb_shared_hwtstamps *hwtstamps);

2853 2854 2855 2856
void __skb_tstamp_tx(struct sk_buff *orig_skb,
		     struct skb_shared_hwtstamps *hwtstamps,
		     struct sock *sk, int tstype);

2857 2858 2859 2860 2861 2862 2863 2864 2865 2866 2867
/**
 * 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.
 */
2868 2869
void skb_tstamp_tx(struct sk_buff *orig_skb,
		   struct skb_shared_hwtstamps *hwtstamps);
2870

2871 2872
static inline void sw_tx_timestamp(struct sk_buff *skb)
{
2873 2874
	if (skb_shinfo(skb)->tx_flags & SKBTX_SW_TSTAMP &&
	    !(skb_shinfo(skb)->tx_flags & SKBTX_IN_PROGRESS))
2875 2876 2877 2878 2879 2880 2881
		skb_tstamp_tx(skb, NULL);
}

/**
 * skb_tx_timestamp() - Driver hook for transmit timestamping
 *
 * Ethernet MAC Drivers should call this function in their hard_xmit()
2882
 * function immediately before giving the sk_buff to the MAC hardware.
2883
 *
2884 2885 2886 2887
 * Specifically, one should make absolutely sure that this function is
 * called before TX completion of this packet can trigger.  Otherwise
 * the packet could potentially already be freed.
 *
2888 2889 2890 2891
 * @skb: A socket buffer.
 */
static inline void skb_tx_timestamp(struct sk_buff *skb)
{
2892
	skb_clone_tx_timestamp(skb);
2893 2894 2895
	sw_tx_timestamp(skb);
}

2896 2897 2898 2899 2900 2901 2902 2903 2904
/**
 * 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);

2905 2906
__sum16 __skb_checksum_complete_head(struct sk_buff *skb, int len);
__sum16 __skb_checksum_complete(struct sk_buff *skb);
2907

2908 2909
static inline int skb_csum_unnecessary(const struct sk_buff *skb)
{
2910
	return ((skb->ip_summed & CHECKSUM_UNNECESSARY) || skb->csum_valid);
2911 2912
}

2913 2914 2915 2916 2917 2918 2919 2920 2921 2922 2923 2924 2925 2926 2927 2928
/**
 *	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.
 */
2929
static inline __sum16 skb_checksum_complete(struct sk_buff *skb)
2930
{
2931 2932
	return skb_csum_unnecessary(skb) ?
	       0 : __skb_checksum_complete(skb);
2933 2934
}

2935 2936 2937 2938 2939 2940 2941 2942 2943 2944 2945 2946 2947 2948 2949 2950 2951 2952 2953 2954 2955
static inline void __skb_decr_checksum_unnecessary(struct sk_buff *skb)
{
	if (skb->ip_summed == CHECKSUM_UNNECESSARY) {
		if (skb->csum_level == 0)
			skb->ip_summed = CHECKSUM_NONE;
		else
			skb->csum_level--;
	}
}

static inline void __skb_incr_checksum_unnecessary(struct sk_buff *skb)
{
	if (skb->ip_summed == CHECKSUM_UNNECESSARY) {
		if (skb->csum_level < SKB_MAX_CSUM_LEVEL)
			skb->csum_level++;
	} else if (skb->ip_summed == CHECKSUM_NONE) {
		skb->ip_summed = CHECKSUM_UNNECESSARY;
		skb->csum_level = 0;
	}
}

2956 2957 2958 2959 2960 2961 2962 2963 2964 2965 2966 2967 2968 2969 2970
static inline void __skb_mark_checksum_bad(struct sk_buff *skb)
{
	/* Mark current checksum as bad (typically called from GRO
	 * path). In the case that ip_summed is CHECKSUM_NONE
	 * this must be the first checksum encountered in the packet.
	 * When ip_summed is CHECKSUM_UNNECESSARY, this is the first
	 * checksum after the last one validated. For UDP, a zero
	 * checksum can not be marked as bad.
	 */

	if (skb->ip_summed == CHECKSUM_NONE ||
	    skb->ip_summed == CHECKSUM_UNNECESSARY)
		skb->csum_bad = 1;
}

2971 2972 2973 2974 2975 2976 2977 2978 2979
/* Check if we need to perform checksum complete validation.
 *
 * Returns true if checksum complete is needed, false otherwise
 * (either checksum is unnecessary or zero checksum is allowed).
 */
static inline bool __skb_checksum_validate_needed(struct sk_buff *skb,
						  bool zero_okay,
						  __sum16 check)
{
2980 2981
	if (skb_csum_unnecessary(skb) || (zero_okay && !check)) {
		skb->csum_valid = 1;
2982
		__skb_decr_checksum_unnecessary(skb);
2983 2984 2985 2986 2987 2988 2989 2990 2991 2992 2993 2994 2995 2996 2997 2998 2999 3000 3001 3002 3003 3004 3005 3006 3007 3008
		return false;
	}

	return true;
}

/* For small packets <= CHECKSUM_BREAK peform checksum complete directly
 * in checksum_init.
 */
#define CHECKSUM_BREAK 76

/* Validate (init) checksum based on checksum complete.
 *
 * Return values:
 *   0: checksum is validated or try to in skb_checksum_complete. In the latter
 *	case the ip_summed will not be CHECKSUM_UNNECESSARY and the pseudo
 *	checksum is stored in skb->csum for use in __skb_checksum_complete
 *   non-zero: value of invalid checksum
 *
 */
static inline __sum16 __skb_checksum_validate_complete(struct sk_buff *skb,
						       bool complete,
						       __wsum psum)
{
	if (skb->ip_summed == CHECKSUM_COMPLETE) {
		if (!csum_fold(csum_add(psum, skb->csum))) {
3009
			skb->csum_valid = 1;
3010 3011
			return 0;
		}
3012 3013 3014
	} else if (skb->csum_bad) {
		/* ip_summed == CHECKSUM_NONE in this case */
		return 1;
3015 3016 3017 3018
	}

	skb->csum = psum;

3019 3020 3021 3022 3023 3024 3025
	if (complete || skb->len <= CHECKSUM_BREAK) {
		__sum16 csum;

		csum = __skb_checksum_complete(skb);
		skb->csum_valid = !csum;
		return csum;
	}
3026 3027 3028 3029 3030 3031 3032 3033 3034 3035 3036 3037 3038 3039 3040 3041 3042 3043 3044 3045 3046 3047 3048

	return 0;
}

static inline __wsum null_compute_pseudo(struct sk_buff *skb, int proto)
{
	return 0;
}

/* Perform checksum validate (init). Note that this is a macro since we only
 * want to calculate the pseudo header which is an input function if necessary.
 * First we try to validate without any computation (checksum unnecessary) and
 * then calculate based on checksum complete calling the function to compute
 * pseudo header.
 *
 * Return values:
 *   0: checksum is validated or try to in skb_checksum_complete
 *   non-zero: value of invalid checksum
 */
#define __skb_checksum_validate(skb, proto, complete,			\
				zero_okay, check, compute_pseudo)	\
({									\
	__sum16 __ret = 0;						\
3049
	skb->csum_valid = 0;						\
3050 3051 3052 3053 3054 3055 3056 3057 3058 3059 3060 3061 3062 3063 3064 3065 3066 3067 3068 3069 3070 3071
	if (__skb_checksum_validate_needed(skb, zero_okay, check))	\
		__ret = __skb_checksum_validate_complete(skb,		\
				complete, compute_pseudo(skb, proto));	\
	__ret;								\
})

#define skb_checksum_init(skb, proto, compute_pseudo)			\
	__skb_checksum_validate(skb, proto, false, false, 0, compute_pseudo)

#define skb_checksum_init_zero_check(skb, proto, check, compute_pseudo)	\
	__skb_checksum_validate(skb, proto, false, true, check, compute_pseudo)

#define skb_checksum_validate(skb, proto, compute_pseudo)		\
	__skb_checksum_validate(skb, proto, true, false, 0, compute_pseudo)

#define skb_checksum_validate_zero_check(skb, proto, check,		\
					 compute_pseudo)		\
	__skb_checksum_validate_(skb, proto, true, true, check, compute_pseudo)

#define skb_checksum_simple_validate(skb)				\
	__skb_checksum_validate(skb, 0, true, false, 0, null_compute_pseudo)

3072 3073 3074 3075 3076 3077 3078 3079 3080 3081 3082 3083 3084 3085 3086 3087 3088 3089 3090 3091
static inline bool __skb_checksum_convert_check(struct sk_buff *skb)
{
	return (skb->ip_summed == CHECKSUM_NONE &&
		skb->csum_valid && !skb->csum_bad);
}

static inline void __skb_checksum_convert(struct sk_buff *skb,
					  __sum16 check, __wsum pseudo)
{
	skb->csum = ~pseudo;
	skb->ip_summed = CHECKSUM_COMPLETE;
}

#define skb_checksum_try_convert(skb, proto, check, compute_pseudo)	\
do {									\
	if (__skb_checksum_convert_check(skb))				\
		__skb_checksum_convert(skb, check,			\
				       compute_pseudo(skb, proto));	\
} while (0)

3092
#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE)
3093
void nf_conntrack_destroy(struct nf_conntrack *nfct);
L
Linus Torvalds 已提交
3094 3095 3096
static inline void nf_conntrack_put(struct nf_conntrack *nfct)
{
	if (nfct && atomic_dec_and_test(&nfct->use))
3097
		nf_conntrack_destroy(nfct);
L
Linus Torvalds 已提交
3098 3099 3100 3101 3102 3103
}
static inline void nf_conntrack_get(struct nf_conntrack *nfct)
{
	if (nfct)
		atomic_inc(&nfct->use);
}
3104
#endif
3105
#if IS_ENABLED(CONFIG_BRIDGE_NETFILTER)
L
Linus Torvalds 已提交
3106 3107 3108 3109 3110 3111 3112 3113 3114 3115 3116
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 */
3117 3118
static inline void nf_reset(struct sk_buff *skb)
{
3119
#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE)
3120 3121
	nf_conntrack_put(skb->nfct);
	skb->nfct = NULL;
3122
#endif
3123
#if IS_ENABLED(CONFIG_BRIDGE_NETFILTER)
3124 3125 3126 3127 3128
	nf_bridge_put(skb->nf_bridge);
	skb->nf_bridge = NULL;
#endif
}

3129 3130
static inline void nf_reset_trace(struct sk_buff *skb)
{
3131
#if IS_ENABLED(CONFIG_NETFILTER_XT_TARGET_TRACE) || defined(CONFIG_NF_TABLES)
G
Gao feng 已提交
3132 3133
	skb->nf_trace = 0;
#endif
3134 3135
}

3136
/* Note: This doesn't put any conntrack and bridge info in dst. */
3137 3138
static inline void __nf_copy(struct sk_buff *dst, const struct sk_buff *src,
			     bool copy)
3139
{
3140
#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE)
3141 3142
	dst->nfct = src->nfct;
	nf_conntrack_get(src->nfct);
3143 3144
	if (copy)
		dst->nfctinfo = src->nfctinfo;
3145
#endif
3146
#if IS_ENABLED(CONFIG_BRIDGE_NETFILTER)
3147 3148 3149
	dst->nf_bridge  = src->nf_bridge;
	nf_bridge_get(src->nf_bridge);
#endif
3150
#if IS_ENABLED(CONFIG_NETFILTER_XT_TARGET_TRACE) || defined(CONFIG_NF_TABLES)
3151 3152
	if (copy)
		dst->nf_trace = src->nf_trace;
3153
#endif
3154 3155
}

3156 3157 3158
static inline void nf_copy(struct sk_buff *dst, const struct sk_buff *src)
{
#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE)
3159
	nf_conntrack_put(dst->nfct);
3160
#endif
3161
#if IS_ENABLED(CONFIG_BRIDGE_NETFILTER)
3162 3163
	nf_bridge_put(dst->nf_bridge);
#endif
3164
	__nf_copy(dst, src, true);
3165 3166
}

3167 3168 3169 3170 3171 3172 3173 3174 3175 3176 3177 3178 3179 3180 3181 3182 3183 3184
#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

3185 3186 3187 3188 3189 3190 3191 3192 3193 3194 3195 3196 3197
static inline bool skb_irq_freeable(const struct sk_buff *skb)
{
	return !skb->destructor &&
#if IS_ENABLED(CONFIG_XFRM)
		!skb->sp &&
#endif
#if IS_ENABLED(CONFIG_NF_CONNTRACK)
		!skb->nfct &&
#endif
		!skb->_skb_refdst &&
		!skb_has_frag_list(skb);
}

3198 3199 3200 3201 3202
static inline void skb_set_queue_mapping(struct sk_buff *skb, u16 queue_mapping)
{
	skb->queue_mapping = queue_mapping;
}

3203
static inline u16 skb_get_queue_mapping(const struct sk_buff *skb)
3204 3205 3206 3207
{
	return skb->queue_mapping;
}

3208 3209 3210 3211 3212
static inline void skb_copy_queue_mapping(struct sk_buff *to, const struct sk_buff *from)
{
	to->queue_mapping = from->queue_mapping;
}

3213 3214 3215 3216 3217
static inline void skb_record_rx_queue(struct sk_buff *skb, u16 rx_queue)
{
	skb->queue_mapping = rx_queue + 1;
}

3218
static inline u16 skb_get_rx_queue(const struct sk_buff *skb)
3219 3220 3221 3222
{
	return skb->queue_mapping - 1;
}

3223
static inline bool skb_rx_queue_recorded(const struct sk_buff *skb)
3224
{
E
Eric Dumazet 已提交
3225
	return skb->queue_mapping != 0;
3226 3227
}

3228
u16 __skb_tx_hash(const struct net_device *dev, struct sk_buff *skb,
3229
		  unsigned int num_tx_queues);
3230

3231 3232
static inline struct sec_path *skb_sec_path(struct sk_buff *skb)
{
3233
#ifdef CONFIG_XFRM
3234 3235 3236 3237
	return skb->sp;
#else
	return NULL;
#endif
3238
}
3239

3240 3241 3242
/* 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
3243 3244 3245
 * tunnel skb it points to outer mac header.
 * Keeps track of level of encapsulation of network headers.
 */
3246
struct skb_gso_cb {
3247 3248
	int	mac_offset;
	int	encap_level;
3249
	__u16	csum_start;
3250 3251 3252 3253 3254 3255 3256 3257 3258
};
#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;
}

3259 3260 3261 3262 3263 3264 3265 3266 3267 3268 3269 3270 3271 3272 3273
static inline int gso_pskb_expand_head(struct sk_buff *skb, int extra)
{
	int new_headroom, headroom;
	int ret;

	headroom = skb_headroom(skb);
	ret = pskb_expand_head(skb, extra, 0, GFP_ATOMIC);
	if (ret)
		return ret;

	new_headroom = skb_headroom(skb);
	SKB_GSO_CB(skb)->mac_offset += (new_headroom - headroom);
	return 0;
}

3274 3275 3276 3277 3278 3279 3280 3281 3282 3283 3284 3285 3286 3287 3288 3289 3290 3291 3292 3293 3294 3295
/* Compute the checksum for a gso segment. First compute the checksum value
 * from the start of transport header to SKB_GSO_CB(skb)->csum_start, and
 * then add in skb->csum (checksum from csum_start to end of packet).
 * skb->csum and csum_start are then updated to reflect the checksum of the
 * resultant packet starting from the transport header-- the resultant checksum
 * is in the res argument (i.e. normally zero or ~ of checksum of a pseudo
 * header.
 */
static inline __sum16 gso_make_checksum(struct sk_buff *skb, __wsum res)
{
	int plen = SKB_GSO_CB(skb)->csum_start - skb_headroom(skb) -
	    skb_transport_offset(skb);
	__u16 csum;

	csum = csum_fold(csum_partial(skb_transport_header(skb),
				      plen, skb->csum));
	skb->csum = res;
	SKB_GSO_CB(skb)->csum_start -= plen;

	return csum;
}

3296
static inline bool skb_is_gso(const struct sk_buff *skb)
H
Herbert Xu 已提交
3297 3298 3299 3300
{
	return skb_shinfo(skb)->gso_size;
}

3301
/* Note: Should be called only if skb_is_gso(skb) is true */
3302
static inline bool skb_is_gso_v6(const struct sk_buff *skb)
B
Brice Goglin 已提交
3303 3304 3305 3306
{
	return skb_shinfo(skb)->gso_type & SKB_GSO_TCPV6;
}

3307
void __skb_warn_lro_forwarding(const struct sk_buff *skb);
3308 3309 3310 3311 3312

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. */
3313 3314
	const struct skb_shared_info *shinfo = skb_shinfo(skb);

3315 3316
	if (skb_is_nonlinear(skb) && shinfo->gso_size != 0 &&
	    unlikely(shinfo->gso_type == 0)) {
3317 3318 3319 3320 3321 3322
		__skb_warn_lro_forwarding(skb);
		return true;
	}
	return false;
}

3323 3324 3325 3326 3327 3328 3329
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;
}

3330 3331 3332 3333 3334 3335 3336 3337
/**
 * 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.
 */
3338
static inline void skb_checksum_none_assert(const struct sk_buff *skb)
3339 3340 3341 3342 3343 3344
{
#ifdef DEBUG
	BUG_ON(skb->ip_summed != CHECKSUM_NONE);
#endif
}

3345
bool skb_partial_csum_set(struct sk_buff *skb, u16 start, u16 off);
3346

P
Paul Durrant 已提交
3347 3348
int skb_checksum_setup(struct sk_buff *skb, bool recalculate);

3349 3350 3351
u32 skb_get_poff(const struct sk_buff *skb);
u32 __skb_get_poff(const struct sk_buff *skb, void *data,
		   const struct flow_keys *keys, int hlen);
3352

3353 3354 3355 3356 3357 3358 3359 3360 3361 3362 3363 3364 3365
/**
 * 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);
}
3366 3367 3368 3369 3370 3371 3372 3373 3374 3375 3376 3377 3378 3379 3380 3381 3382

/**
 * skb_gso_network_seglen - Return length of individual segments of a gso packet
 *
 * @skb: GSO skb
 *
 * skb_gso_network_seglen is used to determine the real size of the
 * individual segments, including Layer3 (IP, IPv6) and L4 headers (TCP/UDP).
 *
 * The MAC/L2 header is not accounted for.
 */
static inline unsigned int skb_gso_network_seglen(const struct sk_buff *skb)
{
	unsigned int hdr_len = skb_transport_header(skb) -
			       skb_network_header(skb);
	return hdr_len + skb_gso_transport_seglen(skb);
}
L
Linus Torvalds 已提交
3383 3384
#endif	/* __KERNEL__ */
#endif	/* _LINUX_SKBUFF_H */