skbuff.h 81.5 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
/*
 *	Definitions for the 'struct sk_buff' memory handlers.
 *
 *	Authors:
 *		Alan Cox, <gw4pts@gw4pts.ampr.org>
 *		Florian La Roche, <rzsfl@rz.uni-sb.de>
 *
 *	This program is free software; you can redistribute it and/or
 *	modify it under the terms of the GNU General Public License
 *	as published by the Free Software Foundation; either version
 *	2 of the License, or (at your option) any later version.
 */

#ifndef _LINUX_SKBUFF_H
#define _LINUX_SKBUFF_H

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

A
Arun Sharma 已提交
24
#include <linux/atomic.h>
L
Linus Torvalds 已提交
25 26 27
#include <asm/types.h>
#include <linux/spinlock.h>
#include <linux/net.h>
28
#include <linux/textsearch.h>
L
Linus Torvalds 已提交
29
#include <net/checksum.h>
30
#include <linux/rcupdate.h>
31
#include <linux/dmaengine.h>
32
#include <linux/hrtimer.h>
33
#include <linux/dma-mapping.h>
34
#include <linux/netdev_features.h>
35
#include <net/flow_keys.h>
L
Linus Torvalds 已提交
36

37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 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
/* 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
 *   for specific protocols e.g. TCP/UDP/SCTP, then, 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_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
 */

108
/* Don't change this without changing skb_csum_unnecessary! */
109 110 111 112
#define CHECKSUM_NONE		0
#define CHECKSUM_UNNECESSARY	1
#define CHECKSUM_COMPLETE	2
#define CHECKSUM_PARTIAL	3
L
Linus Torvalds 已提交
113 114 115

#define SKB_DATA_ALIGN(X)	(((X) + (SMP_CACHE_BYTES - 1)) & \
				 ~(SMP_CACHE_BYTES - 1))
116
#define SKB_WITH_OVERHEAD(X)	\
117
	((X) - SKB_DATA_ALIGN(sizeof(struct skb_shared_info)))
118 119
#define SKB_MAX_ORDER(X, ORDER) \
	SKB_WITH_OVERHEAD((PAGE_SIZE << (ORDER)) - (X))
L
Linus Torvalds 已提交
120 121 122
#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
123 124 125 126 127
/* 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
128
struct net_device;
129
struct scatterlist;
J
Jens Axboe 已提交
130
struct pipe_inode_info;
L
Linus Torvalds 已提交
131

132
#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE)
L
Linus Torvalds 已提交
133 134 135
struct nf_conntrack {
	atomic_t use;
};
136
#endif
L
Linus Torvalds 已提交
137 138 139

#ifdef CONFIG_BRIDGE_NETFILTER
struct nf_bridge_info {
140 141 142 143 144
	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
145 146 147 148 149 150 151 152 153 154 155 156 157 158
};
#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;

159 160 161 162 163 164
/* 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.
165
 */
166
#if (65536/PAGE_SIZE + 1) < 16
167
#define MAX_SKB_FRAGS 16UL
168
#else
169
#define MAX_SKB_FRAGS (65536/PAGE_SIZE + 1)
170
#endif
L
Linus Torvalds 已提交
171 172 173 174

typedef struct skb_frag_struct skb_frag_t;

struct skb_frag_struct {
175 176 177
	struct {
		struct page *p;
	} page;
178
#if (BITS_PER_LONG > 32) || (PAGE_SIZE >= 65536)
179 180
	__u32 page_offset;
	__u32 size;
181 182 183 184
#else
	__u16 page_offset;
	__u16 size;
#endif
L
Linus Torvalds 已提交
185 186
};

E
Eric Dumazet 已提交
187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206
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;
}

207 208 209
#define HAVE_HW_TIME_STAMP

/**
210
 * struct skb_shared_hwtstamps - hardware time stamps
211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236
 * @hwtstamp:	hardware time stamp transformed into duration
 *		since arbitrary point in time
 * @syststamp:	hwtstamp transformed to system time base
 *
 * Software time stamps generated by ktime_get_real() are stored in
 * skb->tstamp. The relation between the different kinds of time
 * stamps is as follows:
 *
 * syststamp and tstamp can be compared against each other in
 * arbitrary combinations.  The accuracy of a
 * syststamp/tstamp/"syststamp from other device" comparison is
 * limited by the accuracy of the transformation into system time
 * base. This depends on the device driver and its underlying
 * hardware.
 *
 * hwtstamps can only be compared against other hwtstamps from
 * the same device.
 *
 * This structure is attached to packets as part of the
 * &skb_shared_info. Use skb_hwtstamps() to get a pointer.
 */
struct skb_shared_hwtstamps {
	ktime_t	hwtstamp;
	ktime_t	syststamp;
};

237 238 239 240 241 242 243 244 245 246 247
/* Definitions for tx_flags in struct skb_shared_info */
enum {
	/* generate hardware time stamp */
	SKBTX_HW_TSTAMP = 1 << 0,

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

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

248
	/* device driver supports TX zero-copy buffers */
E
Eric Dumazet 已提交
249
	SKBTX_DEV_ZEROCOPY = 1 << 3,
250 251

	/* generate wifi status information (where possible) */
E
Eric Dumazet 已提交
252
	SKBTX_WIFI_STATUS = 1 << 4,
253 254 255 256 257 258 259

	/* 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,
260 261 262 263 264
};

/*
 * 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.
265 266
 * 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.
267 268
 * The ctx field is used to track device context.
 * The desc field is used to track userspace buffer index.
269 270
 */
struct ubuf_info {
271
	void (*callback)(struct ubuf_info *, bool zerocopy_success);
272
	void *ctx;
273
	unsigned long desc;
274 275
};

L
Linus Torvalds 已提交
276 277 278 279
/* This data is invariant across clones and lives at
 * the end of the header data, ie. at skb->end.
 */
struct skb_shared_info {
280 281
	unsigned char	nr_frags;
	__u8		tx_flags;
282 283 284 285
	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
286
	struct sk_buff	*frag_list;
287
	struct skb_shared_hwtstamps hwtstamps;
288
	__be32          ip6_frag_id;
E
Eric Dumazet 已提交
289 290 291 292 293 294

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

J
Johann Baudy 已提交
295 296 297
	/* Intermediate layers must ensure that destructor_arg
	 * remains valid until skb destructor */
	void *		destructor_arg;
298

299 300
	/* must be last field, see pskb_expand_head() */
	skb_frag_t	frags[MAX_SKB_FRAGS];
L
Linus Torvalds 已提交
301 302 303 304
};

/* 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
305 306
 * 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
307 308 309 310 311 312 313 314 315 316
 *
 * 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)

317 318 319 320 321 322 323

enum {
	SKB_FCLONE_UNAVAILABLE,
	SKB_FCLONE_ORIG,
	SKB_FCLONE_CLONE,
};

324 325
enum {
	SKB_GSO_TCPV4 = 1 << 0,
H
Herbert Xu 已提交
326
	SKB_GSO_UDP = 1 << 1,
327 328 329

	/* This indicates the skb is from an untrusted source. */
	SKB_GSO_DODGY = 1 << 2,
M
Michael Chan 已提交
330 331

	/* This indicates the tcp segment has CWR set. */
H
Herbert Xu 已提交
332 333 334
	SKB_GSO_TCP_ECN = 1 << 3,

	SKB_GSO_TCPV6 = 1 << 4,
335 336

	SKB_GSO_FCOE = 1 << 5,
337 338

	SKB_GSO_GRE = 1 << 6,
339

E
Eric Dumazet 已提交
340
	SKB_GSO_IPIP = 1 << 7,
S
Simon Horman 已提交
341

E
Eric Dumazet 已提交
342
	SKB_GSO_SIT = 1 << 8,
E
Eric Dumazet 已提交
343

E
Eric Dumazet 已提交
344 345 346
	SKB_GSO_UDP_TUNNEL = 1 << 9,

	SKB_GSO_MPLS = 1 << 10,
347 348
};

349 350 351 352 353 354 355 356 357 358
#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

L
Linus Torvalds 已提交
359 360 361 362
/** 
 *	struct sk_buff - socket buffer
 *	@next: Next buffer in list
 *	@prev: Previous buffer in list
H
Herbert Xu 已提交
363
 *	@tstamp: Time we arrived
364
 *	@sk: Socket we are owned by
L
Linus Torvalds 已提交
365
 *	@dev: Device we arrived on/are leaving by
366
 *	@cb: Control buffer. Free for use by every layer. Put private vars here
E
Eric Dumazet 已提交
367
 *	@_skb_refdst: destination entry (with norefcount bit)
368
 *	@sp: the security path, used for xfrm
L
Linus Torvalds 已提交
369 370 371
 *	@len: Length of actual data
 *	@data_len: Data length
 *	@mac_len: Length of link layer header
372
 *	@hdr_len: writable header length of cloned skb
373 374 375
 *	@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
376
 *	@priority: Packet queueing priority
377
 *	@local_df: allow local fragmentation
L
Linus Torvalds 已提交
378
 *	@cloned: Head may be cloned (check refcnt to be sure)
379
 *	@ip_summed: Driver fed us an IP checksum
L
Linus Torvalds 已提交
380
 *	@nohdr: Payload reference only, must not modify header
381
 *	@nfctinfo: Relationship of this skb to the connection
L
Linus Torvalds 已提交
382
 *	@pkt_type: Packet class
383 384
 *	@fclone: skbuff clone status
 *	@ipvs_property: skbuff is owned by ipvs
385 386
 *	@peeked: this packet has been seen already, so stats have been
 *		done for it, don't do them again
387
 *	@nf_trace: netfilter packet trace flag
388 389 390
 *	@protocol: Packet protocol from driver
 *	@destructor: Destruct function
 *	@nfct: Associated connection, if any
L
Linus Torvalds 已提交
391
 *	@nf_bridge: Saved data about a bridged frame - see br_netfilter.c
392
 *	@skb_iif: ifindex of device we arrived on
L
Linus Torvalds 已提交
393 394
 *	@tc_index: Traffic control index
 *	@tc_verd: traffic control verdict
395 396
 *	@rxhash: the packet hash computed on receive
 *	@queue_mapping: Queue mapping for multiqueue devices
397
 *	@ndisc_nodetype: router type (from link layer)
398
 *	@ooo_okay: allow the mapping of a socket to a queue to be changed
399 400
 *	@l4_rxhash: indicate rxhash is a canonical 4-tuple hash over transport
 *		ports.
401 402
 *	@wifi_acked_valid: wifi_acked was set
 *	@wifi_acked: whether frame was acked on wifi or not
403
 *	@no_fcs:  Request NIC to treat last 4 bytes as Ethernet FCS
R
Randy Dunlap 已提交
404 405
 *	@dma_cookie: a cookie to one of several possible DMA operations
 *		done by skb DMA functions
E
Eliezer Tamir 已提交
406
  *	@napi_id: id of the NAPI struct this skb came from
407
 *	@secmark: security marking
408 409
 *	@mark: Generic packet mark
 *	@dropcount: total number of sk_receive_queue overflows
410
 *	@vlan_proto: vlan encapsulation protocol
411
 *	@vlan_tci: vlan tag control information
S
Simon Horman 已提交
412
 *	@inner_protocol: Protocol (encapsulation)
413 414
 *	@inner_transport_header: Inner transport layer header (encapsulation)
 *	@inner_network_header: Network layer header (encapsulation)
415
 *	@inner_mac_header: Link layer header (encapsulation)
416 417 418 419 420 421 422 423 424
 *	@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
425 426 427 428 429 430 431
 */

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

432
	ktime_t			tstamp;
433 434

	struct sock		*sk;
L
Linus Torvalds 已提交
435 436 437 438 439 440 441 442
	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.
	 */
443
	char			cb[48] __aligned(8);
L
Linus Torvalds 已提交
444

E
Eric Dumazet 已提交
445
	unsigned long		_skb_refdst;
446 447 448
#ifdef CONFIG_XFRM
	struct	sec_path	*sp;
#endif
L
Linus Torvalds 已提交
449
	unsigned int		len,
450 451 452
				data_len;
	__u16			mac_len,
				hdr_len;
A
Al Viro 已提交
453 454
	union {
		__wsum		csum;
455 456 457 458
		struct {
			__u16	csum_start;
			__u16	csum_offset;
		};
A
Al Viro 已提交
459
	};
L
Linus Torvalds 已提交
460
	__u32			priority;
461
	kmemcheck_bitfield_begin(flags1);
462 463 464
	__u8			local_df:1,
				cloned:1,
				ip_summed:2,
465 466
				nohdr:1,
				nfctinfo:3;
467
	__u8			pkt_type:3,
468
				fclone:2,
469
				ipvs_property:1,
470
				peeked:1,
471
				nf_trace:1;
472
	kmemcheck_bitfield_end(flags1);
E
Eric Dumazet 已提交
473
	__be16			protocol;
L
Linus Torvalds 已提交
474 475

	void			(*destructor)(struct sk_buff *skb);
476
#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE)
477
	struct nf_conntrack	*nfct;
478
#endif
L
Linus Torvalds 已提交
479 480 481
#ifdef CONFIG_BRIDGE_NETFILTER
	struct nf_bridge_info	*nf_bridge;
#endif
482

483
	int			skb_iif;
484 485 486

	__u32			rxhash;

487
	__be16			vlan_proto;
488 489
	__u16			vlan_tci;

L
Linus Torvalds 已提交
490
#ifdef CONFIG_NET_SCHED
491
	__u16			tc_index;	/* traffic control index */
L
Linus Torvalds 已提交
492
#ifdef CONFIG_NET_CLS_ACT
493
	__u16			tc_verd;	/* traffic control verdict */
L
Linus Torvalds 已提交
494 495
#endif
#endif
496

497
	__u16			queue_mapping;
498
	kmemcheck_bitfield_begin(flags2);
499
#ifdef CONFIG_IPV6_NDISC_NODETYPE
500
	__u8			ndisc_nodetype:2;
501
#endif
502
	__u8			pfmemalloc:1;
503
	__u8			ooo_okay:1;
504
	__u8			l4_rxhash:1;
505 506
	__u8			wifi_acked_valid:1;
	__u8			wifi_acked:1;
507
	__u8			no_fcs:1;
508
	__u8			head_frag:1;
509 510 511 512 513 514
	/* Encapsulation protocol and NIC drivers should use
	 * this flag to indicate to each other if the skb contains
	 * encapsulated packet or not and maybe use the inner packet
	 * headers if needed
	 */
	__u8			encapsulation:1;
515
	/* 6/8 bit hole (depending on ndisc_nodetype presence) */
516 517
	kmemcheck_bitfield_end(flags2);

518
#if defined CONFIG_NET_DMA || defined CONFIG_NET_RX_BUSY_POLL
E
Eliezer Tamir 已提交
519 520 521 522
	union {
		unsigned int	napi_id;
		dma_cookie_t	dma_cookie;
	};
523
#endif
524 525 526
#ifdef CONFIG_NETWORK_SECMARK
	__u32			secmark;
#endif
527 528 529
	union {
		__u32		mark;
		__u32		dropcount;
E
Eric Dumazet 已提交
530
		__u32		reserved_tailroom;
531
	};
L
Linus Torvalds 已提交
532

S
Simon Horman 已提交
533
	__be16			inner_protocol;
534 535 536 537 538 539
	__u16			inner_transport_header;
	__u16			inner_network_header;
	__u16			inner_mac_header;
	__u16			transport_header;
	__u16			network_header;
	__u16			mac_header;
L
Linus Torvalds 已提交
540
	/* These elements must be at the end, see alloc_skb() for details.  */
541
	sk_buff_data_t		tail;
542
	sk_buff_data_t		end;
L
Linus Torvalds 已提交
543
	unsigned char		*head,
544
				*data;
545 546
	unsigned int		truesize;
	atomic_t		users;
L
Linus Torvalds 已提交
547 548 549 550 551 552 553 554 555
};

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


556 557 558 559 560 561 562 563 564
#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
565 566 567 568 569 570 571 572 573 574 575 576 577
/*
 * 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
578 579
static inline struct dst_entry *skb_dst(const struct sk_buff *skb)
{
E
Eric Dumazet 已提交
580 581 582 583 584 585 586
	/* 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
587 588
}

E
Eric Dumazet 已提交
589 590 591 592 593 594 595 596
/**
 * 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
597 598
static inline void skb_dst_set(struct sk_buff *skb, struct dst_entry *dst)
{
E
Eric Dumazet 已提交
599 600 601
	skb->_skb_refdst = (unsigned long)dst;
}

602 603
void __skb_dst_set_noref(struct sk_buff *skb, struct dst_entry *dst,
			 bool force);
604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635

/**
 * 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
636 637

/**
L
Lucas De Marchi 已提交
638
 * skb_dst_is_noref - Test if skb dst isn't refcounted
E
Eric Dumazet 已提交
639 640 641 642 643
 * @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
644 645
}

E
Eric Dumazet 已提交
646 647
static inline struct rtable *skb_rtable(const struct sk_buff *skb)
{
E
Eric Dumazet 已提交
648
	return (struct rtable *)skb_dst(skb);
E
Eric Dumazet 已提交
649 650
}

651 652 653 654 655
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);
656
extern struct kmem_cache *skbuff_head_cache;
E
Eric Dumazet 已提交
657

658 659 660
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
661

662 663 664
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);
665
static inline struct sk_buff *alloc_skb(unsigned int size,
A
Al Viro 已提交
666
					gfp_t priority)
667
{
E
Eric Dumazet 已提交
668
	return __alloc_skb(size, priority, 0, NUMA_NO_NODE);
669 670 671
}

static inline struct sk_buff *alloc_skb_fclone(unsigned int size,
A
Al Viro 已提交
672
					       gfp_t priority)
673
{
674
	return __alloc_skb(size, priority, SKB_ALLOC_FCLONE, NUMA_NO_NODE);
675 676
}

677
struct sk_buff *__alloc_skb_head(gfp_t priority, int node);
678 679 680 681 682
static inline struct sk_buff *alloc_skb_head(gfp_t priority)
{
	return __alloc_skb_head(priority, -1);
}

683 684 685 686 687 688 689 690 691 692 693 694 695 696 697
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);
struct sk_buff *__pskb_copy(struct sk_buff *skb, int headroom, gfp_t gfp_mask);

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);
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);
698
#define dev_kfree_skb(a)	consume_skb(a)
L
Linus Torvalds 已提交
699

700 701 702 703
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);
704

E
Eric Dumazet 已提交
705
struct skb_seq_state {
706 707 708 709 710 711 712 713 714
	__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;
};

715 716 717 718 719
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);
720

721 722 723
unsigned int skb_find_text(struct sk_buff *skb, unsigned int from,
			   unsigned int to, struct ts_config *config,
			   struct ts_state *state);
724

T
Tom Herbert 已提交
725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764
/*
 * 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)
{
	skb->l4_rxhash = (type == PKT_HASH_TYPE_L4);
	skb->rxhash = hash;
}

765 766
void __skb_get_hash(struct sk_buff *skb);
static inline __u32 skb_get_hash(struct sk_buff *skb)
767
{
768
	if (!skb->l4_rxhash)
769
		__skb_get_hash(skb);
770 771 772 773

	return skb->rxhash;
}

774 775 776 777 778 779 780 781 782 783 784 785
static inline void skb_clear_hash(struct sk_buff *skb)
{
	skb->rxhash = 0;
	skb->l4_rxhash = 0;
}

static inline void skb_clear_hash_if_not_l4(struct sk_buff *skb)
{
	if (!skb->l4_rxhash)
		skb_clear_hash(skb);
}

786 787 788 789 790 791
static inline void skb_copy_hash(struct sk_buff *to, const struct sk_buff *from)
{
	to->rxhash = from->rxhash;
	to->l4_rxhash = from->l4_rxhash;
};

792 793 794 795 796
#ifdef NET_SKBUFF_DATA_USES_OFFSET
static inline unsigned char *skb_end_pointer(const struct sk_buff *skb)
{
	return skb->head + skb->end;
}
797 798 799 800 801

static inline unsigned int skb_end_offset(const struct sk_buff *skb)
{
	return skb->end;
}
802 803 804 805 806
#else
static inline unsigned char *skb_end_pointer(const struct sk_buff *skb)
{
	return skb->end;
}
807 808 809 810 811

static inline unsigned int skb_end_offset(const struct sk_buff *skb)
{
	return skb->end - skb->head;
}
812 813
#endif

L
Linus Torvalds 已提交
814
/* Internal */
815
#define skb_shinfo(SKB)	((struct skb_shared_info *)(skb_end_pointer(SKB)))
L
Linus Torvalds 已提交
816

817 818 819 820 821
static inline struct skb_shared_hwtstamps *skb_hwtstamps(struct sk_buff *skb)
{
	return &skb_shinfo(skb)->hwtstamps;
}

L
Linus Torvalds 已提交
822 823 824 825 826 827 828 829 830 831 832
/**
 *	skb_queue_empty - check if a queue is empty
 *	@list: queue head
 *
 *	Returns true if the queue is empty, false otherwise.
 */
static inline int skb_queue_empty(const struct sk_buff_head *list)
{
	return list->next == (struct sk_buff *)list;
}

D
David S. Miller 已提交
833 834 835 836 837 838 839 840 841 842
/**
 *	skb_queue_is_last - check if skb is the last entry in the queue
 *	@list: queue head
 *	@skb: buffer
 *
 *	Returns true if @skb is the last buffer on the list.
 */
static inline bool skb_queue_is_last(const struct sk_buff_head *list,
				     const struct sk_buff *skb)
{
E
Eric Dumazet 已提交
843
	return skb->next == (struct sk_buff *)list;
D
David S. Miller 已提交
844 845
}

846 847 848 849 850 851 852 853 854 855
/**
 *	skb_queue_is_first - check if skb is the first entry in the queue
 *	@list: queue head
 *	@skb: buffer
 *
 *	Returns true if @skb is the first buffer on the list.
 */
static inline bool skb_queue_is_first(const struct sk_buff_head *list,
				      const struct sk_buff *skb)
{
E
Eric Dumazet 已提交
856
	return skb->prev == (struct sk_buff *)list;
857 858
}

D
David S. Miller 已提交
859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876
/**
 *	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;
}

877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894
/**
 *	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
895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926
/**
 *	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;
}

927 928 929 930 931 932 933 934 935 936
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
937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995
/**
 *	skb_header_cloned - is the header a clone
 *	@skb: buffer to check
 *
 *	Returns true if modifying the header part of the buffer requires
 *	the data to be copied.
 */
static inline int skb_header_cloned(const struct sk_buff *skb)
{
	int dataref;

	if (!skb->cloned)
		return 0;

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

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

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

/**
 *	skb_share_check - check if buffer is shared and if so clone it
 *	@skb: buffer to check
 *	@pri: priority for memory allocation
 *
 *	If the buffer is shared the buffer is cloned and the old copy
 *	drops a reference. A new clone with a single reference is returned.
 *	If the buffer is not shared the original buffer is returned. When
 *	being called from interrupt status or with spinlocks held pri must
 *	be GFP_ATOMIC.
 *
 *	NULL is returned on a memory allocation failure.
 */
996
static inline struct sk_buff *skb_share_check(struct sk_buff *skb, gfp_t pri)
L
Linus Torvalds 已提交
997 998 999 1000
{
	might_sleep_if(pri & __GFP_WAIT);
	if (skb_shared(skb)) {
		struct sk_buff *nskb = skb_clone(skb, pri);
1001 1002 1003 1004 1005

		if (likely(nskb))
			consume_skb(skb);
		else
			kfree_skb(skb);
L
Linus Torvalds 已提交
1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030
		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.
 */
1031
static inline struct sk_buff *skb_unshare(struct sk_buff *skb,
A
Al Viro 已提交
1032
					  gfp_t pri)
L
Linus Torvalds 已提交
1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043
{
	might_sleep_if(pri & __GFP_WAIT);
	if (skb_cloned(skb)) {
		struct sk_buff *nskb = skb_copy(skb, pri);
		kfree_skb(skb);	/* Free our shared copy */
		skb = nskb;
	}
	return skb;
}

/**
1044
 *	skb_peek - peek at the head of an &sk_buff_head
L
Linus Torvalds 已提交
1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055
 *	@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.
 */
1056
static inline struct sk_buff *skb_peek(const struct sk_buff_head *list_)
L
Linus Torvalds 已提交
1057
{
1058 1059 1060 1061 1062
	struct sk_buff *skb = list_->next;

	if (skb == (struct sk_buff *)list_)
		skb = NULL;
	return skb;
L
Linus Torvalds 已提交
1063 1064
}

P
Pavel Emelyanov 已提交
1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077
/**
 *	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;
1078

P
Pavel Emelyanov 已提交
1079 1080 1081 1082 1083
	if (next == (struct sk_buff *)list_)
		next = NULL;
	return next;
}

L
Linus Torvalds 已提交
1084
/**
1085
 *	skb_peek_tail - peek at the tail of an &sk_buff_head
L
Linus Torvalds 已提交
1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096
 *	@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.
 */
1097
static inline struct sk_buff *skb_peek_tail(const struct sk_buff_head *list_)
L
Linus Torvalds 已提交
1098
{
1099 1100 1101 1102 1103 1104
	struct sk_buff *skb = list_->prev;

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

L
Linus Torvalds 已提交
1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117
}

/**
 *	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;
}

1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133
/**
 *	__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;
}

1134 1135 1136 1137 1138 1139 1140 1141
/*
 * 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
1142 1143 1144
static inline void skb_queue_head_init(struct sk_buff_head *list)
{
	spin_lock_init(&list->lock);
1145
	__skb_queue_head_init(list);
L
Linus Torvalds 已提交
1146 1147
}

1148 1149 1150 1151 1152 1153 1154
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
1155
/*
1156
 *	Insert an sk_buff on a list.
L
Linus Torvalds 已提交
1157 1158 1159 1160
 *
 *	The "__skb_xxxx()" functions are the non-atomic ones that
 *	can only be called with interrupts disabled.
 */
1161 1162
void skb_insert(struct sk_buff *old, struct sk_buff *newsk,
		struct sk_buff_head *list);
1163 1164 1165 1166 1167 1168 1169 1170 1171
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
1172

1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196
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);
1197
		head->qlen += list->qlen;
1198 1199 1200 1201
	}
}

/**
E
Eric Dumazet 已提交
1202
 *	skb_queue_splice_init - join two skb lists and reinitialise the emptied list
1203 1204 1205 1206 1207 1208 1209 1210 1211 1212
 *	@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);
1213
		head->qlen += list->qlen;
1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227
		__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);
1228
		head->qlen += list->qlen;
1229 1230 1231 1232
	}
}

/**
E
Eric Dumazet 已提交
1233
 *	skb_queue_splice_tail_init - join two skb lists and reinitialise the emptied list
1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244
 *	@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);
1245
		head->qlen += list->qlen;
1246 1247 1248 1249
		__skb_queue_head_init(list);
	}
}

L
Linus Torvalds 已提交
1250
/**
1251
 *	__skb_queue_after - queue a buffer at the list head
L
Linus Torvalds 已提交
1252
 *	@list: list to use
1253
 *	@prev: place after this buffer
L
Linus Torvalds 已提交
1254 1255
 *	@newsk: buffer to queue
 *
1256
 *	Queue a buffer int the middle of a list. This function takes no locks
L
Linus Torvalds 已提交
1257 1258 1259 1260
 *	and you must therefore hold required locks before calling it.
 *
 *	A buffer cannot be placed on two lists at the same time.
 */
1261 1262 1263
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
1264
{
1265
	__skb_insert(newsk, prev, prev->next, list);
L
Linus Torvalds 已提交
1266 1267
}

1268 1269
void skb_append(struct sk_buff *old, struct sk_buff *newsk,
		struct sk_buff_head *list);
1270

1271 1272 1273 1274 1275 1276 1277
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);
}

1278 1279 1280 1281 1282 1283 1284 1285 1286 1287
/**
 *	__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.
 */
1288
void skb_queue_head(struct sk_buff_head *list, struct sk_buff *newsk);
1289 1290 1291 1292 1293 1294
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
1295 1296 1297 1298 1299 1300 1301 1302 1303 1304
/**
 *	__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.
 */
1305
void skb_queue_tail(struct sk_buff_head *list, struct sk_buff *newsk);
L
Linus Torvalds 已提交
1306 1307 1308
static inline void __skb_queue_tail(struct sk_buff_head *list,
				   struct sk_buff *newsk)
{
1309
	__skb_queue_before(list, (struct sk_buff *)list, newsk);
L
Linus Torvalds 已提交
1310 1311 1312 1313 1314 1315
}

/*
 * remove sk_buff from list. _Must_ be called atomically, and with
 * the list known..
 */
1316
void skb_unlink(struct sk_buff *skb, struct sk_buff_head *list);
L
Linus Torvalds 已提交
1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328
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;
}

1329 1330 1331 1332 1333 1334 1335 1336
/**
 *	__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.
 */
1337
struct sk_buff *skb_dequeue(struct sk_buff_head *list);
1338 1339 1340 1341 1342 1343 1344
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
1345 1346 1347 1348 1349 1350 1351 1352 1353

/**
 *	__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.
 */
1354
struct sk_buff *skb_dequeue_tail(struct sk_buff_head *list);
L
Linus Torvalds 已提交
1355 1356 1357 1358 1359 1360 1361 1362 1363
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;
}


1364
static inline bool skb_is_nonlinear(const struct sk_buff *skb)
L
Linus Torvalds 已提交
1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378
{
	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
1379
		len += skb_frag_size(&skb_shinfo(skb)->frags[i]);
L
Linus Torvalds 已提交
1380 1381 1382
	return len + skb_headlen(skb);
}

1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397
/**
 * __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
1398 1399 1400
{
	skb_frag_t *frag = &skb_shinfo(skb)->frags[i];

1401 1402 1403 1404 1405 1406 1407 1408 1409
	/*
	 * 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.
	 */
1410
	frag->page.p		  = page;
L
Linus Torvalds 已提交
1411
	frag->page_offset	  = off;
E
Eric Dumazet 已提交
1412
	skb_frag_size_set(frag, size);
1413 1414 1415 1416

	page = compound_head(page);
	if (page->pfmemalloc && !page->mapping)
		skb->pfmemalloc	= true;
1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427
}

/**
 * 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
1428
 * @skb to point to @size bytes at offset @off within @page. In
1429 1430 1431 1432 1433 1434 1435 1436
 * 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
1437 1438 1439
	skb_shinfo(skb)->nr_frags = i + 1;
}

1440 1441
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
1442

J
Jason Wang 已提交
1443 1444 1445
void skb_coalesce_rx_frag(struct sk_buff *skb, int i, int size,
			  unsigned int truesize);

L
Linus Torvalds 已提交
1446
#define SKB_PAGE_ASSERT(skb) 	BUG_ON(skb_shinfo(skb)->nr_frags)
1447
#define SKB_FRAG_ASSERT(skb) 	BUG_ON(skb_has_frag_list(skb))
L
Linus Torvalds 已提交
1448 1449
#define SKB_LINEAR_ASSERT(skb)  BUG_ON(skb_is_nonlinear(skb))

1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465
#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;
}
1466

1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481
#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;
}
1482

1483 1484
#endif /* NET_SKBUFF_DATA_USES_OFFSET */

L
Linus Torvalds 已提交
1485 1486 1487
/*
 *	Add data to an sk_buff
 */
M
Mathias Krause 已提交
1488
unsigned char *pskb_put(struct sk_buff *skb, struct sk_buff *tail, int len);
1489
unsigned char *skb_put(struct sk_buff *skb, unsigned int len);
L
Linus Torvalds 已提交
1490 1491
static inline unsigned char *__skb_put(struct sk_buff *skb, unsigned int len)
{
1492
	unsigned char *tmp = skb_tail_pointer(skb);
L
Linus Torvalds 已提交
1493 1494 1495 1496 1497 1498
	SKB_LINEAR_ASSERT(skb);
	skb->tail += len;
	skb->len  += len;
	return tmp;
}

1499
unsigned char *skb_push(struct sk_buff *skb, unsigned int len);
L
Linus Torvalds 已提交
1500 1501 1502 1503 1504 1505 1506
static inline unsigned char *__skb_push(struct sk_buff *skb, unsigned int len)
{
	skb->data -= len;
	skb->len  += len;
	return skb->data;
}

1507
unsigned char *skb_pull(struct sk_buff *skb, unsigned int len);
L
Linus Torvalds 已提交
1508 1509 1510 1511 1512 1513 1514
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;
}

1515 1516 1517 1518 1519
static inline unsigned char *skb_pull_inline(struct sk_buff *skb, unsigned int len)
{
	return unlikely(len > skb->len) ? NULL : __skb_pull(skb, len);
}

1520
unsigned char *__pskb_pull_tail(struct sk_buff *skb, int delta);
L
Linus Torvalds 已提交
1521 1522 1523 1524

static inline unsigned char *__pskb_pull(struct sk_buff *skb, unsigned int len)
{
	if (len > skb_headlen(skb) &&
G
Gerrit Renker 已提交
1525
	    !__pskb_pull_tail(skb, len - skb_headlen(skb)))
L
Linus Torvalds 已提交
1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541
		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
1542
	return __pskb_pull_tail(skb, len - skb_headlen(skb)) != NULL;
L
Linus Torvalds 已提交
1543 1544 1545 1546 1547 1548 1549 1550
}

/**
 *	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.
 */
1551
static inline unsigned int skb_headroom(const struct sk_buff *skb)
L
Linus Torvalds 已提交
1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563
{
	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)
{
1564
	return skb_is_nonlinear(skb) ? 0 : skb->end - skb->tail;
L
Linus Torvalds 已提交
1565 1566
}

1567 1568 1569 1570 1571 1572 1573 1574 1575
/**
 *	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
1576 1577 1578 1579
	if (skb_is_nonlinear(skb))
		return 0;

	return skb->end - skb->tail - skb->reserved_tailroom;
1580 1581
}

L
Linus Torvalds 已提交
1582 1583 1584 1585 1586 1587 1588 1589
/**
 *	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.
 */
1590
static inline void skb_reserve(struct sk_buff *skb, int len)
L
Linus Torvalds 已提交
1591 1592 1593 1594 1595
{
	skb->data += len;
	skb->tail += len;
}

1596 1597
static inline void skb_reset_inner_headers(struct sk_buff *skb)
{
1598
	skb->inner_mac_header = skb->mac_header;
1599 1600 1601 1602
	skb->inner_network_header = skb->network_header;
	skb->inner_transport_header = skb->transport_header;
}

1603 1604 1605 1606 1607
static inline void skb_reset_mac_len(struct sk_buff *skb)
{
	skb->mac_len = skb->network_header - skb->mac_header;
}

1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642
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;
}

1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658
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;
}
1659 1660
static inline bool skb_transport_header_was_set(const struct sk_buff *skb)
{
C
Cong Wang 已提交
1661
	return skb->transport_header != (typeof(skb->transport_header))~0U;
1662 1663
}

1664 1665
static inline unsigned char *skb_transport_header(const struct sk_buff *skb)
{
1666
	return skb->head + skb->transport_header;
1667 1668
}

1669 1670
static inline void skb_reset_transport_header(struct sk_buff *skb)
{
1671
	skb->transport_header = skb->data - skb->head;
1672 1673
}

1674 1675 1676
static inline void skb_set_transport_header(struct sk_buff *skb,
					    const int offset)
{
1677 1678
	skb_reset_transport_header(skb);
	skb->transport_header += offset;
1679 1680
}

1681 1682
static inline unsigned char *skb_network_header(const struct sk_buff *skb)
{
1683
	return skb->head + skb->network_header;
1684 1685
}

1686 1687
static inline void skb_reset_network_header(struct sk_buff *skb)
{
1688
	skb->network_header = skb->data - skb->head;
1689 1690
}

1691 1692
static inline void skb_set_network_header(struct sk_buff *skb, const int offset)
{
1693 1694
	skb_reset_network_header(skb);
	skb->network_header += offset;
1695 1696
}

1697
static inline unsigned char *skb_mac_header(const struct sk_buff *skb)
1698
{
1699
	return skb->head + skb->mac_header;
1700 1701
}

1702
static inline int skb_mac_header_was_set(const struct sk_buff *skb)
1703
{
C
Cong Wang 已提交
1704
	return skb->mac_header != (typeof(skb->mac_header))~0U;
1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717
}

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;
}

1718 1719 1720 1721 1722
static inline void skb_pop_mac_header(struct sk_buff *skb)
{
	skb->mac_header = skb->network_header;
}

1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735
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);
}

1736 1737 1738 1739 1740 1741 1742 1743 1744 1745
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);
	}
}

1746 1747 1748 1749 1750
static inline int skb_checksum_start_offset(const struct sk_buff *skb)
{
	return skb->csum_start - skb_headroom(skb);
}

1751 1752 1753 1754 1755 1756 1757 1758 1759 1760
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;
}

1761 1762 1763 1764 1765
static inline u32 skb_inner_network_header_len(const struct sk_buff *skb)
{
	return skb->inner_transport_header - skb->inner_network_header;
}

1766 1767 1768 1769
static inline int skb_network_offset(const struct sk_buff *skb)
{
	return skb_network_header(skb) - skb->data;
}
1770

1771 1772 1773 1774 1775
static inline int skb_inner_network_offset(const struct sk_buff *skb)
{
	return skb_inner_network_header(skb) - skb->data;
}

1776 1777 1778 1779 1780
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
1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791
/*
 * 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:
 *
1792
 * skb_reserve(skb, NET_IP_ALIGN);
L
Linus Torvalds 已提交
1793 1794 1795 1796
 *
 * 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.
1797
 *
L
Linus Torvalds 已提交
1798 1799 1800 1801 1802 1803 1804
 * 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

1805 1806 1807 1808
/*
 * 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
1809
 * 32 bytes or less we avoid the reallocation.
1810 1811 1812 1813 1814 1815 1816
 *
 * 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.
 *
1817
 * Various parts of the networking layer expect at least 32 bytes of
1818
 * headroom, you should not reduce this.
1819 1820 1821 1822
 *
 * 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 :
1823
 * NET_IP_ALIGN(2) + ethernet_header(14) + IP_header(20/40) + ports(8)
1824 1825
 */
#ifndef NET_SKB_PAD
1826
#define NET_SKB_PAD	max(32, L1_CACHE_BYTES)
1827 1828
#endif

1829
int ___pskb_trim(struct sk_buff *skb, unsigned int len);
L
Linus Torvalds 已提交
1830 1831 1832

static inline void __skb_trim(struct sk_buff *skb, unsigned int len)
{
1833
	if (unlikely(skb_is_nonlinear(skb))) {
1834 1835 1836
		WARN_ON(1);
		return;
	}
1837 1838
	skb->len = len;
	skb_set_tail_pointer(skb, len);
L
Linus Torvalds 已提交
1839 1840
}

1841
void skb_trim(struct sk_buff *skb, unsigned int len);
L
Linus Torvalds 已提交
1842 1843 1844

static inline int __pskb_trim(struct sk_buff *skb, unsigned int len)
{
1845 1846 1847 1848
	if (skb->data_len)
		return ___pskb_trim(skb, len);
	__skb_trim(skb, len);
	return 0;
L
Linus Torvalds 已提交
1849 1850 1851 1852 1853 1854 1855
}

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

1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870
/**
 *	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
1871 1872 1873 1874 1875 1876 1877 1878 1879 1880
/**
 *	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
1881
	if (skb->destructor) {
L
Linus Torvalds 已提交
1882
		skb->destructor(skb);
E
Eric Dumazet 已提交
1883 1884
		skb->destructor = NULL;
		skb->sk		= NULL;
1885 1886
	} else {
		BUG_ON(skb->sk);
E
Eric Dumazet 已提交
1887
	}
L
Linus Torvalds 已提交
1888 1889
}

1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905
/**
 *	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
1906 1907 1908 1909 1910 1911 1912 1913
/**
 *	__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.
 */
1914
void skb_queue_purge(struct sk_buff_head *list);
L
Linus Torvalds 已提交
1915 1916 1917 1918 1919 1920 1921
static inline void __skb_queue_purge(struct sk_buff_head *list)
{
	struct sk_buff *skb;
	while ((skb = __skb_dequeue(list)) != NULL)
		kfree_skb(skb);
}

1922 1923 1924 1925
#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

1926
void *netdev_alloc_frag(unsigned int fragsz);
L
Linus Torvalds 已提交
1927

1928 1929
struct sk_buff *__netdev_alloc_skb(struct net_device *dev, unsigned int length,
				   gfp_t gfp_mask);
1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944

/**
 *	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,
1945
					       unsigned int length)
1946 1947 1948 1949
{
	return __netdev_alloc_skb(dev, length, GFP_ATOMIC);
}

1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963
/* 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);
}


1964 1965
static inline struct sk_buff *__netdev_alloc_skb_ip_align(struct net_device *dev,
		unsigned int length, gfp_t gfp)
1966
{
1967
	struct sk_buff *skb = __netdev_alloc_skb(dev, length + NET_IP_ALIGN, gfp);
1968 1969 1970 1971 1972 1973

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

1974 1975 1976 1977 1978 1979
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);
}

1980 1981
/**
 *	__skb_alloc_pages - allocate pages for ps-rx on a skb and preserve pfmemalloc data
1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034
 *	@gfp_mask: alloc_pages_node mask. Set __GFP_NOMEMALLOC if not for network packet RX
 *	@skb: skb to set pfmemalloc on if __GFP_MEMALLOC is used
 *	@order: size of the allocation
 *
 * 	Allocate a new page.
 *
 * 	%NULL is returned if there is no free memory.
*/
static inline struct page *__skb_alloc_pages(gfp_t gfp_mask,
					      struct sk_buff *skb,
					      unsigned int order)
{
	struct page *page;

	gfp_mask |= __GFP_COLD;

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

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

	return page;
}

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

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

2035 2036 2037 2038 2039 2040 2041 2042
/**
 * skb_frag_page - retrieve the page refered to by a paged fragment
 * @frag: the paged fragment
 *
 * Returns the &struct page associated with @frag.
 */
static inline struct page *skb_frag_page(const skb_frag_t *frag)
{
2043
	return frag->page.p;
2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 2106 2107 2108 2109 2110 2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128
}

/**
 * __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)
{
2129
	frag->page.p = page;
2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145
}

/**
 * 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
2146 2147
bool skb_page_frag_refill(unsigned int sz, struct page_frag *pfrag, gfp_t prio);

2148 2149
/**
 * skb_frag_dma_map - maps a paged fragment via the DMA API
2150
 * @dev: the device to map the fragment to
2151 2152 2153 2154
 * @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
2155
 * @dir: the direction of the mapping (%PCI_DMA_*)
2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167
 *
 * 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
2168 2169 2170 2171 2172 2173
static inline struct sk_buff *pskb_copy(struct sk_buff *skb,
					gfp_t gfp_mask)
{
	return __pskb_copy(skb, skb_headroom(skb), gfp_mask);
}

2174 2175 2176 2177 2178 2179 2180 2181
/**
 *	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.
 */
2182
static inline int skb_clone_writable(const struct sk_buff *skb, unsigned int len)
2183 2184 2185 2186 2187
{
	return !skb_header_cloned(skb) &&
	       skb_headroom(skb) + len <= skb->hdr_len;
}

H
Herbert Xu 已提交
2188 2189 2190 2191 2192 2193 2194 2195 2196 2197 2198 2199 2200 2201
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
2202 2203 2204 2205 2206 2207 2208 2209 2210 2211 2212 2213 2214 2215
/**
 *	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
2216 2217
	return __skb_cow(skb, headroom, skb_cloned(skb));
}
L
Linus Torvalds 已提交
2218

H
Herbert Xu 已提交
2219 2220 2221 2222 2223 2224 2225 2226 2227 2228 2229 2230 2231
/**
 *	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
2232 2233 2234 2235 2236 2237 2238 2239 2240
}

/**
 *	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
2241 2242
 *	is untouched. Otherwise it is extended. Returns zero on
 *	success. The skb is freed on error.
L
Linus Torvalds 已提交
2243 2244
 */
 
2245
static inline int skb_padto(struct sk_buff *skb, unsigned int len)
L
Linus Torvalds 已提交
2246 2247 2248
{
	unsigned int size = skb->len;
	if (likely(size >= len))
2249
		return 0;
G
Gerrit Renker 已提交
2250
	return skb_pad(skb, len - size);
L
Linus Torvalds 已提交
2251 2252 2253 2254 2255 2256 2257 2258 2259
}

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;
2260
		__wsum csum = csum_and_copy_from_user(from, skb_put(skb, copy),
L
Linus Torvalds 已提交
2261 2262 2263 2264 2265 2266 2267 2268 2269 2270 2271 2272
							    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;
}

2273 2274
static inline bool skb_can_coalesce(struct sk_buff *skb, int i,
				    const struct page *page, int off)
L
Linus Torvalds 已提交
2275 2276
{
	if (i) {
E
Eric Dumazet 已提交
2277
		const struct skb_frag_struct *frag = &skb_shinfo(skb)->frags[i - 1];
L
Linus Torvalds 已提交
2278

2279
		return page == skb_frag_page(frag) &&
E
Eric Dumazet 已提交
2280
		       off == frag->page_offset + skb_frag_size(frag);
L
Linus Torvalds 已提交
2281
	}
2282
	return false;
L
Linus Torvalds 已提交
2283 2284
}

H
Herbert Xu 已提交
2285 2286 2287 2288 2289
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
2290 2291 2292 2293 2294 2295 2296
/**
 *	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
2297 2298 2299 2300 2301
static inline int skb_linearize(struct sk_buff *skb)
{
	return skb_is_nonlinear(skb) ? __skb_linearize(skb) : 0;
}

2302 2303 2304 2305 2306 2307 2308 2309 2310
/**
 * 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)
{
2311 2312
	return skb_is_nonlinear(skb) &&
	       skb_shinfo(skb)->tx_flags & SKBTX_SHARED_FRAG;
2313 2314
}

H
Herbert Xu 已提交
2315 2316 2317 2318 2319 2320 2321 2322
/**
 *	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
2323
{
H
Herbert Xu 已提交
2324 2325
	return skb_is_nonlinear(skb) || skb_cloned(skb) ?
	       __skb_linearize(skb) : 0;
L
Linus Torvalds 已提交
2326 2327 2328 2329 2330 2331 2332 2333 2334
}

/**
 *	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
2335 2336
 *	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
2337 2338 2339
 */

static inline void skb_postpull_rcsum(struct sk_buff *skb,
2340
				      const void *start, unsigned int len)
L
Linus Torvalds 已提交
2341
{
2342
	if (skb->ip_summed == CHECKSUM_COMPLETE)
L
Linus Torvalds 已提交
2343 2344 2345
		skb->csum = csum_sub(skb->csum, csum_partial(start, len, 0));
}

2346 2347
unsigned char *skb_pull_rcsum(struct sk_buff *skb, unsigned int len);

2348 2349 2350 2351 2352 2353 2354 2355 2356 2357 2358 2359 2360 2361 2362 2363 2364 2365
/**
 *	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
2366 2367
#define skb_queue_walk(queue, skb) \
		for (skb = (queue)->next;					\
2368
		     skb != (struct sk_buff *)(queue);				\
L
Linus Torvalds 已提交
2369 2370
		     skb = skb->next)

2371 2372 2373 2374 2375
#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)

2376
#define skb_queue_walk_from(queue, skb)						\
2377
		for (; skb != (struct sk_buff *)(queue);			\
2378 2379 2380 2381 2382 2383 2384
		     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)

2385 2386
#define skb_queue_reverse_walk(queue, skb) \
		for (skb = (queue)->prev;					\
2387
		     skb != (struct sk_buff *)(queue);				\
2388 2389
		     skb = skb->prev)

2390 2391 2392 2393 2394 2395 2396 2397 2398
#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
2399

2400
static inline bool skb_has_frag_list(const struct sk_buff *skb)
2401 2402 2403 2404 2405 2406 2407 2408 2409 2410 2411 2412 2413 2414 2415 2416 2417 2418
{
	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)

2419 2420 2421 2422 2423 2424 2425 2426 2427 2428 2429 2430 2431 2432 2433 2434 2435 2436 2437 2438 2439 2440 2441 2442 2443 2444 2445 2446 2447
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);
int skb_copy_and_csum_datagram_iovec(struct sk_buff *skb, int hlen,
				     struct iovec *iov);
int skb_copy_datagram_from_iovec(struct sk_buff *skb, int offset,
				 const struct iovec *from, int from_offset,
				 int len);
int zerocopy_sg_from_iovec(struct sk_buff *skb, const struct iovec *frm,
			   int offset, size_t count);
int skb_copy_datagram_const_iovec(const struct sk_buff *from, int offset,
				  const struct iovec *to, int to_offset,
				  int size);
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);
2448 2449 2450
unsigned int skb_zerocopy_headlen(const struct sk_buff *from);
void skb_zerocopy(struct sk_buff *to, const struct sk_buff *from,
		  int len, int hlen);
2451 2452 2453 2454
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);
struct sk_buff *skb_segment(struct sk_buff *skb, netdev_features_t features);
2455

2456 2457 2458 2459 2460 2461 2462 2463 2464 2465
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);

L
Linus Torvalds 已提交
2466 2467 2468 2469 2470
static inline void *skb_header_pointer(const struct sk_buff *skb, int offset,
				       int len, void *buffer)
{
	int hlen = skb_headlen(skb);

2471
	if (hlen - offset >= len)
L
Linus Torvalds 已提交
2472 2473 2474 2475 2476 2477 2478 2479
		return skb->data + offset;

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

	return buffer;
}

2480 2481 2482 2483 2484 2485 2486 2487 2488 2489 2490 2491 2492 2493 2494 2495 2496 2497
/**
 *	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)));
}

2498 2499 2500 2501 2502 2503 2504 2505 2506 2507 2508 2509 2510 2511
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);
}

2512 2513 2514 2515 2516 2517 2518 2519 2520 2521 2522 2523 2524 2525 2526
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);
}

2527
void skb_init(void);
L
Linus Torvalds 已提交
2528

2529 2530 2531 2532 2533
static inline ktime_t skb_get_ktime(const struct sk_buff *skb)
{
	return skb->tstamp;
}

2534 2535 2536 2537 2538 2539 2540 2541 2542
/**
 *	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.
 */
2543 2544
static inline void skb_get_timestamp(const struct sk_buff *skb,
				     struct timeval *stamp)
2545
{
2546
	*stamp = ktime_to_timeval(skb->tstamp);
2547 2548
}

2549 2550 2551 2552 2553 2554
static inline void skb_get_timestampns(const struct sk_buff *skb,
				       struct timespec *stamp)
{
	*stamp = ktime_to_timespec(skb->tstamp);
}

2555
static inline void __net_timestamp(struct sk_buff *skb)
2556
{
2557
	skb->tstamp = ktime_get_real();
2558 2559
}

2560 2561 2562 2563 2564
static inline ktime_t net_timedelta(ktime_t t)
{
	return ktime_sub(ktime_get_real(), t);
}

2565 2566 2567 2568
static inline ktime_t net_invalid_timestamp(void)
{
	return ktime_set(0, 0);
}
2569

2570
void skb_timestamping_init(void);
2571 2572 2573

#ifdef CONFIG_NETWORK_PHY_TIMESTAMPING

2574 2575
void skb_clone_tx_timestamp(struct sk_buff *skb);
bool skb_defer_rx_timestamp(struct sk_buff *skb);
2576 2577 2578 2579 2580 2581 2582 2583 2584 2585 2586 2587 2588 2589 2590 2591 2592

#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
 *
2593 2594 2595 2596 2597
 * 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.
 *
2598
 * @skb: clone of the the original outgoing packet
2599
 * @hwtstamps: hardware time stamps, may be NULL if not available
2600 2601 2602 2603 2604
 *
 */
void skb_complete_tx_timestamp(struct sk_buff *skb,
			       struct skb_shared_hwtstamps *hwtstamps);

2605 2606 2607 2608 2609 2610 2611 2612 2613 2614 2615
/**
 * 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.
 */
2616 2617
void skb_tstamp_tx(struct sk_buff *orig_skb,
		   struct skb_shared_hwtstamps *hwtstamps);
2618

2619 2620
static inline void sw_tx_timestamp(struct sk_buff *skb)
{
2621 2622
	if (skb_shinfo(skb)->tx_flags & SKBTX_SW_TSTAMP &&
	    !(skb_shinfo(skb)->tx_flags & SKBTX_IN_PROGRESS))
2623 2624 2625 2626 2627 2628 2629
		skb_tstamp_tx(skb, NULL);
}

/**
 * skb_tx_timestamp() - Driver hook for transmit timestamping
 *
 * Ethernet MAC Drivers should call this function in their hard_xmit()
2630
 * function immediately before giving the sk_buff to the MAC hardware.
2631
 *
2632 2633 2634 2635
 * 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.
 *
2636 2637 2638 2639
 * @skb: A socket buffer.
 */
static inline void skb_tx_timestamp(struct sk_buff *skb)
{
2640
	skb_clone_tx_timestamp(skb);
2641 2642 2643
	sw_tx_timestamp(skb);
}

2644 2645 2646 2647 2648 2649 2650 2651 2652
/**
 * 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);

2653 2654
__sum16 __skb_checksum_complete_head(struct sk_buff *skb, int len);
__sum16 __skb_checksum_complete(struct sk_buff *skb);
2655

2656 2657 2658 2659 2660
static inline int skb_csum_unnecessary(const struct sk_buff *skb)
{
	return skb->ip_summed & CHECKSUM_UNNECESSARY;
}

2661 2662 2663 2664 2665 2666 2667 2668 2669 2670 2671 2672 2673 2674 2675 2676
/**
 *	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.
 */
2677
static inline __sum16 skb_checksum_complete(struct sk_buff *skb)
2678
{
2679 2680
	return skb_csum_unnecessary(skb) ?
	       0 : __skb_checksum_complete(skb);
2681 2682
}

2683
#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE)
2684
void nf_conntrack_destroy(struct nf_conntrack *nfct);
L
Linus Torvalds 已提交
2685 2686 2687
static inline void nf_conntrack_put(struct nf_conntrack *nfct)
{
	if (nfct && atomic_dec_and_test(&nfct->use))
2688
		nf_conntrack_destroy(nfct);
L
Linus Torvalds 已提交
2689 2690 2691 2692 2693 2694
}
static inline void nf_conntrack_get(struct nf_conntrack *nfct)
{
	if (nfct)
		atomic_inc(&nfct->use);
}
2695
#endif
L
Linus Torvalds 已提交
2696 2697 2698 2699 2700 2701 2702 2703 2704 2705 2706 2707
#ifdef CONFIG_BRIDGE_NETFILTER
static inline void nf_bridge_put(struct nf_bridge_info *nf_bridge)
{
	if (nf_bridge && atomic_dec_and_test(&nf_bridge->use))
		kfree(nf_bridge);
}
static inline void nf_bridge_get(struct nf_bridge_info *nf_bridge)
{
	if (nf_bridge)
		atomic_inc(&nf_bridge->use);
}
#endif /* CONFIG_BRIDGE_NETFILTER */
2708 2709
static inline void nf_reset(struct sk_buff *skb)
{
2710
#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE)
2711 2712
	nf_conntrack_put(skb->nfct);
	skb->nfct = NULL;
2713
#endif
2714 2715 2716 2717 2718 2719
#ifdef CONFIG_BRIDGE_NETFILTER
	nf_bridge_put(skb->nf_bridge);
	skb->nf_bridge = NULL;
#endif
}

2720 2721
static inline void nf_reset_trace(struct sk_buff *skb)
{
G
Gao feng 已提交
2722 2723 2724
#if IS_ENABLED(CONFIG_NETFILTER_XT_TARGET_TRACE)
	skb->nf_trace = 0;
#endif
2725 2726
}

2727 2728 2729
/* Note: This doesn't put any conntrack and bridge info in dst. */
static inline void __nf_copy(struct sk_buff *dst, const struct sk_buff *src)
{
2730
#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE)
2731 2732 2733
	dst->nfct = src->nfct;
	nf_conntrack_get(src->nfct);
	dst->nfctinfo = src->nfctinfo;
2734
#endif
2735 2736 2737 2738 2739 2740
#ifdef CONFIG_BRIDGE_NETFILTER
	dst->nf_bridge  = src->nf_bridge;
	nf_bridge_get(src->nf_bridge);
#endif
}

2741 2742 2743
static inline void nf_copy(struct sk_buff *dst, const struct sk_buff *src)
{
#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE)
2744
	nf_conntrack_put(dst->nfct);
2745
#endif
2746 2747 2748 2749 2750 2751
#ifdef CONFIG_BRIDGE_NETFILTER
	nf_bridge_put(dst->nf_bridge);
#endif
	__nf_copy(dst, src);
}

2752 2753 2754 2755 2756 2757 2758 2759 2760 2761 2762 2763 2764 2765 2766 2767 2768 2769
#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

2770 2771 2772 2773 2774
static inline void skb_set_queue_mapping(struct sk_buff *skb, u16 queue_mapping)
{
	skb->queue_mapping = queue_mapping;
}

2775
static inline u16 skb_get_queue_mapping(const struct sk_buff *skb)
2776 2777 2778 2779
{
	return skb->queue_mapping;
}

2780 2781 2782 2783 2784
static inline void skb_copy_queue_mapping(struct sk_buff *to, const struct sk_buff *from)
{
	to->queue_mapping = from->queue_mapping;
}

2785 2786 2787 2788 2789
static inline void skb_record_rx_queue(struct sk_buff *skb, u16 rx_queue)
{
	skb->queue_mapping = rx_queue + 1;
}

2790
static inline u16 skb_get_rx_queue(const struct sk_buff *skb)
2791 2792 2793 2794
{
	return skb->queue_mapping - 1;
}

2795
static inline bool skb_rx_queue_recorded(const struct sk_buff *skb)
2796
{
E
Eric Dumazet 已提交
2797
	return skb->queue_mapping != 0;
2798 2799
}

2800 2801
u16 __skb_tx_hash(const struct net_device *dev, const struct sk_buff *skb,
		  unsigned int num_tx_queues);
2802

2803 2804
static inline struct sec_path *skb_sec_path(struct sk_buff *skb)
{
2805
#ifdef CONFIG_XFRM
2806 2807 2808 2809
	return skb->sp;
#else
	return NULL;
#endif
2810
}
2811

2812 2813 2814
/* 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
2815 2816 2817
 * tunnel skb it points to outer mac header.
 * Keeps track of level of encapsulation of network headers.
 */
2818
struct skb_gso_cb {
2819 2820
	int	mac_offset;
	int	encap_level;
2821 2822 2823 2824 2825 2826 2827 2828 2829
};
#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;
}

2830 2831 2832 2833 2834 2835 2836 2837 2838 2839 2840 2841 2842 2843 2844
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;
}

2845
static inline bool skb_is_gso(const struct sk_buff *skb)
H
Herbert Xu 已提交
2846 2847 2848 2849
{
	return skb_shinfo(skb)->gso_size;
}

2850
/* Note: Should be called only if skb_is_gso(skb) is true */
2851
static inline bool skb_is_gso_v6(const struct sk_buff *skb)
B
Brice Goglin 已提交
2852 2853 2854 2855
{
	return skb_shinfo(skb)->gso_type & SKB_GSO_TCPV6;
}

2856
void __skb_warn_lro_forwarding(const struct sk_buff *skb);
2857 2858 2859 2860 2861

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. */
2862 2863
	const struct skb_shared_info *shinfo = skb_shinfo(skb);

2864 2865
	if (skb_is_nonlinear(skb) && shinfo->gso_size != 0 &&
	    unlikely(shinfo->gso_type == 0)) {
2866 2867 2868 2869 2870 2871
		__skb_warn_lro_forwarding(skb);
		return true;
	}
	return false;
}

2872 2873 2874 2875 2876 2877 2878
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;
}

2879 2880 2881 2882 2883 2884 2885 2886
/**
 * 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.
 */
2887
static inline void skb_checksum_none_assert(const struct sk_buff *skb)
2888 2889 2890 2891 2892 2893
{
#ifdef DEBUG
	BUG_ON(skb->ip_summed != CHECKSUM_NONE);
#endif
}

2894
bool skb_partial_csum_set(struct sk_buff *skb, u16 start, u16 off);
2895

2896 2897
u32 __skb_get_poff(const struct sk_buff *skb);

2898 2899 2900 2901 2902 2903 2904 2905 2906 2907 2908 2909 2910
/**
 * skb_head_is_locked - Determine if the skb->head is locked down
 * @skb: skb to check
 *
 * The head on skbs build around a head frag can be removed if they are
 * not cloned.  This function returns true if the skb head is locked down
 * due to either being allocated via kmalloc, or by being a clone with
 * multiple references to the head.
 */
static inline bool skb_head_is_locked(const struct sk_buff *skb)
{
	return !skb->head_frag || skb_cloned(skb);
}
L
Linus Torvalds 已提交
2911 2912
#endif	/* __KERNEL__ */
#endif	/* _LINUX_SKBUFF_H */