skbuff.c 82.5 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3
/*
 *	Routines having to do with the 'struct sk_buff' memory handlers.
 *
4
 *	Authors:	Alan Cox <alan@lxorguk.ukuu.org.uk>
L
Linus Torvalds 已提交
5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41
 *			Florian La Roche <rzsfl@rz.uni-sb.de>
 *
 *	Fixes:
 *		Alan Cox	:	Fixed the worst of the load
 *					balancer bugs.
 *		Dave Platt	:	Interrupt stacking fix.
 *	Richard Kooijman	:	Timestamp fixes.
 *		Alan Cox	:	Changed buffer format.
 *		Alan Cox	:	destructor hook for AF_UNIX etc.
 *		Linus Torvalds	:	Better skb_clone.
 *		Alan Cox	:	Added skb_copy.
 *		Alan Cox	:	Added all the changed routines Linus
 *					only put in the headers
 *		Ray VanTassle	:	Fixed --skb->lock in free
 *		Alan Cox	:	skb_copy copy arp field
 *		Andi Kleen	:	slabified it.
 *		Robert Olsson	:	Removed skb_head_pool
 *
 *	NOTE:
 *		The __skb_ routines should be called with interrupts
 *	disabled, or you better be *real* sure that the operation is atomic
 *	with respect to whatever list is being frobbed (e.g. via lock_sock()
 *	or via disabling bottom half handlers, etc).
 *
 *	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.
 */

/*
 *	The functions in this file will not compile correctly with gcc 2.4.x
 */

#include <linux/module.h>
#include <linux/types.h>
#include <linux/kernel.h>
42
#include <linux/kmemcheck.h>
L
Linus Torvalds 已提交
43 44 45 46 47 48 49 50 51 52 53
#include <linux/mm.h>
#include <linux/interrupt.h>
#include <linux/in.h>
#include <linux/inet.h>
#include <linux/slab.h>
#include <linux/netdevice.h>
#ifdef CONFIG_NET_CLS_ACT
#include <net/pkt_sched.h>
#endif
#include <linux/string.h>
#include <linux/skbuff.h>
J
Jens Axboe 已提交
54
#include <linux/splice.h>
L
Linus Torvalds 已提交
55 56 57
#include <linux/cache.h>
#include <linux/rtnetlink.h>
#include <linux/init.h>
58
#include <linux/scatterlist.h>
59
#include <linux/errqueue.h>
60
#include <linux/prefetch.h>
L
Linus Torvalds 已提交
61 62 63 64 65 66 67 68

#include <net/protocol.h>
#include <net/dst.h>
#include <net/sock.h>
#include <net/checksum.h>
#include <net/xfrm.h>

#include <asm/uaccess.h>
69
#include <trace/events/skb.h>
E
Eric Dumazet 已提交
70
#include <linux/highmem.h>
A
Al Viro 已提交
71

72
struct kmem_cache *skbuff_head_cache __read_mostly;
73
static struct kmem_cache *skbuff_fclone_cache __read_mostly;
L
Linus Torvalds 已提交
74

J
Jens Axboe 已提交
75 76 77
static void sock_pipe_buf_release(struct pipe_inode_info *pipe,
				  struct pipe_buffer *buf)
{
78
	put_page(buf->page);
J
Jens Axboe 已提交
79 80 81 82 83
}

static void sock_pipe_buf_get(struct pipe_inode_info *pipe,
				struct pipe_buffer *buf)
{
84
	get_page(buf->page);
J
Jens Axboe 已提交
85 86 87 88 89 90 91 92 93 94
}

static int sock_pipe_buf_steal(struct pipe_inode_info *pipe,
			       struct pipe_buffer *buf)
{
	return 1;
}


/* Pipe buffer operations for a socket. */
95
static const struct pipe_buf_operations sock_pipe_buf_ops = {
J
Jens Axboe 已提交
96 97 98 99 100 101 102 103 104
	.can_merge = 0,
	.map = generic_pipe_buf_map,
	.unmap = generic_pipe_buf_unmap,
	.confirm = generic_pipe_buf_confirm,
	.release = sock_pipe_buf_release,
	.steal = sock_pipe_buf_steal,
	.get = sock_pipe_buf_get,
};

L
Linus Torvalds 已提交
105 106 107 108 109 110 111 112 113 114 115 116 117 118
/*
 *	Keep out-of-line to prevent kernel bloat.
 *	__builtin_return_address is not used because it is not always
 *	reliable.
 */

/**
 *	skb_over_panic	- 	private function
 *	@skb: buffer
 *	@sz: size
 *	@here: address
 *
 *	Out of line support code for skb_put(). Not user callable.
 */
119
static void skb_over_panic(struct sk_buff *skb, int sz, void *here)
L
Linus Torvalds 已提交
120
{
121
	printk(KERN_EMERG "skb_over_panic: text:%p len:%d put:%d head:%p "
122
			  "data:%p tail:%#lx end:%#lx dev:%s\n",
123
	       here, skb->len, sz, skb->head, skb->data,
124
	       (unsigned long)skb->tail, (unsigned long)skb->end,
125
	       skb->dev ? skb->dev->name : "<NULL>");
L
Linus Torvalds 已提交
126 127 128 129 130 131 132 133 134 135 136 137
	BUG();
}

/**
 *	skb_under_panic	- 	private function
 *	@skb: buffer
 *	@sz: size
 *	@here: address
 *
 *	Out of line support code for skb_push(). Not user callable.
 */

138
static void skb_under_panic(struct sk_buff *skb, int sz, void *here)
L
Linus Torvalds 已提交
139
{
140
	printk(KERN_EMERG "skb_under_panic: text:%p len:%d put:%d head:%p "
141
			  "data:%p tail:%#lx end:%#lx dev:%s\n",
142
	       here, skb->len, sz, skb->head, skb->data,
143
	       (unsigned long)skb->tail, (unsigned long)skb->end,
144
	       skb->dev ? skb->dev->name : "<NULL>");
L
Linus Torvalds 已提交
145 146 147 148 149 150 151 152 153 154
	BUG();
}

/* 	Allocate a new skbuff. We do this ourselves so we can fill in a few
 *	'private' fields and also do memory statistics to find all the
 *	[BEEP] leaks.
 *
 */

/**
155
 *	__alloc_skb	-	allocate a network buffer
L
Linus Torvalds 已提交
156 157
 *	@size: size to allocate
 *	@gfp_mask: allocation mask
158 159
 *	@fclone: allocate from fclone cache instead of head cache
 *		and allocate a cloned (child) skb
160
 *	@node: numa node to allocate memory on
L
Linus Torvalds 已提交
161 162 163 164 165 166 167 168
 *
 *	Allocate a new &sk_buff. The returned buffer has no headroom and a
 *	tail room of size bytes. The object has a reference count of one.
 *	The return is the buffer. On a failure the return is %NULL.
 *
 *	Buffers may only be allocated from interrupts using a @gfp_mask of
 *	%GFP_ATOMIC.
 */
A
Al Viro 已提交
169
struct sk_buff *__alloc_skb(unsigned int size, gfp_t gfp_mask,
170
			    int fclone, int node)
L
Linus Torvalds 已提交
171
{
172
	struct kmem_cache *cache;
B
Benjamin LaHaise 已提交
173
	struct skb_shared_info *shinfo;
L
Linus Torvalds 已提交
174 175 176
	struct sk_buff *skb;
	u8 *data;

177 178
	cache = fclone ? skbuff_fclone_cache : skbuff_head_cache;

L
Linus Torvalds 已提交
179
	/* Get the HEAD */
180
	skb = kmem_cache_alloc_node(cache, gfp_mask & ~__GFP_DMA, node);
L
Linus Torvalds 已提交
181 182
	if (!skb)
		goto out;
E
Eric Dumazet 已提交
183
	prefetchw(skb);
L
Linus Torvalds 已提交
184

E
Eric Dumazet 已提交
185 186 187 188 189
	/* We do our best to align skb_shared_info on a separate cache
	 * line. It usually works because kmalloc(X > SMP_CACHE_BYTES) gives
	 * aligned memory blocks, unless SLUB/SLAB debug is enabled.
	 * Both skb->head and skb_shared_info are cache line aligned.
	 */
190
	size = SKB_DATA_ALIGN(size);
E
Eric Dumazet 已提交
191 192
	size += SKB_DATA_ALIGN(sizeof(struct skb_shared_info));
	data = kmalloc_node_track_caller(size, gfp_mask, node);
L
Linus Torvalds 已提交
193 194
	if (!data)
		goto nodata;
E
Eric Dumazet 已提交
195 196 197 198 199
	/* kmalloc(size) might give us more room than requested.
	 * Put skb_shared_info exactly at the end of allocated zone,
	 * to allow max possible filling before reallocation.
	 */
	size = SKB_WITH_OVERHEAD(ksize(data));
E
Eric Dumazet 已提交
200
	prefetchw(data + size);
L
Linus Torvalds 已提交
201

202
	/*
203 204 205
	 * Only clear those fields we need to clear, not those that we will
	 * actually initialise below. Hence, don't put any more fields after
	 * the tail pointer in struct sk_buff!
206 207
	 */
	memset(skb, 0, offsetof(struct sk_buff, tail));
E
Eric Dumazet 已提交
208 209
	/* Account for allocated memory : skb + skb->head */
	skb->truesize = SKB_TRUESIZE(size);
L
Linus Torvalds 已提交
210 211 212
	atomic_set(&skb->users, 1);
	skb->head = data;
	skb->data = data;
213
	skb_reset_tail_pointer(skb);
214
	skb->end = skb->tail + size;
215 216 217 218
#ifdef NET_SKBUFF_DATA_USES_OFFSET
	skb->mac_header = ~0U;
#endif

B
Benjamin LaHaise 已提交
219 220
	/* make sure we initialize shinfo sequentially */
	shinfo = skb_shinfo(skb);
E
Eric Dumazet 已提交
221
	memset(shinfo, 0, offsetof(struct skb_shared_info, dataref));
B
Benjamin LaHaise 已提交
222
	atomic_set(&shinfo->dataref, 1);
223
	kmemcheck_annotate_variable(shinfo->destructor_arg);
B
Benjamin LaHaise 已提交
224

225 226 227
	if (fclone) {
		struct sk_buff *child = skb + 1;
		atomic_t *fclone_ref = (atomic_t *) (child + 1);
L
Linus Torvalds 已提交
228

229 230
		kmemcheck_annotate_bitfield(child, flags1);
		kmemcheck_annotate_bitfield(child, flags2);
231 232 233 234 235
		skb->fclone = SKB_FCLONE_ORIG;
		atomic_set(fclone_ref, 1);

		child->fclone = SKB_FCLONE_UNAVAILABLE;
	}
L
Linus Torvalds 已提交
236 237 238
out:
	return skb;
nodata:
239
	kmem_cache_free(cache, skb);
L
Linus Torvalds 已提交
240 241 242
	skb = NULL;
	goto out;
}
243
EXPORT_SYMBOL(__alloc_skb);
L
Linus Torvalds 已提交
244

E
Eric Dumazet 已提交
245 246 247
/**
 * build_skb - build a network buffer
 * @data: data buffer provided by caller
248
 * @frag_size: size of fragment, or 0 if head was kmalloced
E
Eric Dumazet 已提交
249 250 251 252 253 254 255 256 257 258 259 260 261
 *
 * Allocate a new &sk_buff. Caller provides space holding head and
 * skb_shared_info. @data must have been allocated by kmalloc()
 * The return is the new skb buffer.
 * On a failure the return is %NULL, and @data is not freed.
 * Notes :
 *  Before IO, driver allocates only data buffer where NIC put incoming frame
 *  Driver should add room at head (NET_SKB_PAD) and
 *  MUST add room at tail (SKB_DATA_ALIGN(skb_shared_info))
 *  After IO, driver calls build_skb(), to allocate sk_buff and populate it
 *  before giving packet to stack.
 *  RX rings only contains data buffers, not full skbs.
 */
262
struct sk_buff *build_skb(void *data, unsigned int frag_size)
E
Eric Dumazet 已提交
263 264 265
{
	struct skb_shared_info *shinfo;
	struct sk_buff *skb;
266
	unsigned int size = frag_size ? : ksize(data);
E
Eric Dumazet 已提交
267 268 269 270 271

	skb = kmem_cache_alloc(skbuff_head_cache, GFP_ATOMIC);
	if (!skb)
		return NULL;

272
	size -= SKB_DATA_ALIGN(sizeof(struct skb_shared_info));
E
Eric Dumazet 已提交
273 274 275

	memset(skb, 0, offsetof(struct sk_buff, tail));
	skb->truesize = SKB_TRUESIZE(size);
276
	skb->head_frag = frag_size != 0;
E
Eric Dumazet 已提交
277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295
	atomic_set(&skb->users, 1);
	skb->head = data;
	skb->data = data;
	skb_reset_tail_pointer(skb);
	skb->end = skb->tail + size;
#ifdef NET_SKBUFF_DATA_USES_OFFSET
	skb->mac_header = ~0U;
#endif

	/* make sure we initialize shinfo sequentially */
	shinfo = skb_shinfo(skb);
	memset(shinfo, 0, offsetof(struct skb_shared_info, dataref));
	atomic_set(&shinfo->dataref, 1);
	kmemcheck_annotate_variable(shinfo->destructor_arg);

	return skb;
}
EXPORT_SYMBOL(build_skb);

296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313
/**
 *	__netdev_alloc_skb - allocate an skbuff for rx on a specific device
 *	@dev: network device to receive on
 *	@length: length to allocate
 *	@gfp_mask: get_free_pages mask, passed to alloc_skb
 *
 *	Allocate a new &sk_buff and assign it a usage count of one. The
 *	buffer has unspecified headroom built in. Users should allocate
 *	the headroom they think they need without accounting for the
 *	built in space. The built in space is used for optimisations.
 *
 *	%NULL is returned if there is no free memory.
 */
struct sk_buff *__netdev_alloc_skb(struct net_device *dev,
		unsigned int length, gfp_t gfp_mask)
{
	struct sk_buff *skb;

E
Eric Dumazet 已提交
314
	skb = __alloc_skb(length + NET_SKB_PAD, gfp_mask, 0, NUMA_NO_NODE);
315
	if (likely(skb)) {
316
		skb_reserve(skb, NET_SKB_PAD);
317 318
		skb->dev = dev;
	}
319 320
	return skb;
}
321
EXPORT_SYMBOL(__netdev_alloc_skb);
L
Linus Torvalds 已提交
322

P
Peter Zijlstra 已提交
323
void skb_add_rx_frag(struct sk_buff *skb, int i, struct page *page, int off,
324
		     int size, unsigned int truesize)
P
Peter Zijlstra 已提交
325 326 327 328
{
	skb_fill_page_desc(skb, i, page, off, size);
	skb->len += size;
	skb->data_len += size;
329
	skb->truesize += truesize;
P
Peter Zijlstra 已提交
330 331 332
}
EXPORT_SYMBOL(skb_add_rx_frag);

333 334 335 336 337 338 339 340 341 342 343 344 345 346
/**
 *	dev_alloc_skb - allocate an skbuff for receiving
 *	@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.
 */
struct sk_buff *dev_alloc_skb(unsigned int length)
{
347 348
	/*
	 * There is more code here than it seems:
349
	 * __dev_alloc_skb is an inline
350
	 */
351 352 353 354
	return __dev_alloc_skb(length, GFP_ATOMIC);
}
EXPORT_SYMBOL(dev_alloc_skb);

355
static void skb_drop_list(struct sk_buff **listp)
L
Linus Torvalds 已提交
356
{
357
	struct sk_buff *list = *listp;
L
Linus Torvalds 已提交
358

359
	*listp = NULL;
L
Linus Torvalds 已提交
360 361 362 363 364 365 366 367

	do {
		struct sk_buff *this = list;
		list = list->next;
		kfree_skb(this);
	} while (list);
}

368 369 370 371 372
static inline void skb_drop_fraglist(struct sk_buff *skb)
{
	skb_drop_list(&skb_shinfo(skb)->frag_list);
}

L
Linus Torvalds 已提交
373 374 375 376
static void skb_clone_fraglist(struct sk_buff *skb)
{
	struct sk_buff *list;

377
	skb_walk_frags(skb, list)
L
Linus Torvalds 已提交
378 379 380
		skb_get(list);
}

381 382 383 384 385 386 387 388
static void skb_free_head(struct sk_buff *skb)
{
	if (skb->head_frag)
		put_page(virt_to_head_page(skb->head));
	else
		kfree(skb->head);
}

389
static void skb_release_data(struct sk_buff *skb)
L
Linus Torvalds 已提交
390 391 392 393 394 395 396
{
	if (!skb->cloned ||
	    !atomic_sub_return(skb->nohdr ? (1 << SKB_DATAREF_SHIFT) + 1 : 1,
			       &skb_shinfo(skb)->dataref)) {
		if (skb_shinfo(skb)->nr_frags) {
			int i;
			for (i = 0; i < skb_shinfo(skb)->nr_frags; i++)
397
				skb_frag_unref(skb, i);
L
Linus Torvalds 已提交
398 399
		}

400 401 402 403 404 405 406 407 408 409 410 411
		/*
		 * If skb buf is from userspace, we need to notify the caller
		 * the lower device DMA has done;
		 */
		if (skb_shinfo(skb)->tx_flags & SKBTX_DEV_ZEROCOPY) {
			struct ubuf_info *uarg;

			uarg = skb_shinfo(skb)->destructor_arg;
			if (uarg->callback)
				uarg->callback(uarg);
		}

412
		if (skb_has_frag_list(skb))
L
Linus Torvalds 已提交
413 414
			skb_drop_fraglist(skb);

415
		skb_free_head(skb);
L
Linus Torvalds 已提交
416 417 418 419 420 421
	}
}

/*
 *	Free an skbuff by memory without cleaning the state.
 */
422
static void kfree_skbmem(struct sk_buff *skb)
L
Linus Torvalds 已提交
423
{
424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449
	struct sk_buff *other;
	atomic_t *fclone_ref;

	switch (skb->fclone) {
	case SKB_FCLONE_UNAVAILABLE:
		kmem_cache_free(skbuff_head_cache, skb);
		break;

	case SKB_FCLONE_ORIG:
		fclone_ref = (atomic_t *) (skb + 2);
		if (atomic_dec_and_test(fclone_ref))
			kmem_cache_free(skbuff_fclone_cache, skb);
		break;

	case SKB_FCLONE_CLONE:
		fclone_ref = (atomic_t *) (skb + 1);
		other = skb - 1;

		/* The clone portion is available for
		 * fast-cloning again.
		 */
		skb->fclone = SKB_FCLONE_UNAVAILABLE;

		if (atomic_dec_and_test(fclone_ref))
			kmem_cache_free(skbuff_fclone_cache, other);
		break;
450
	}
L
Linus Torvalds 已提交
451 452
}

453
static void skb_release_head_state(struct sk_buff *skb)
L
Linus Torvalds 已提交
454
{
E
Eric Dumazet 已提交
455
	skb_dst_drop(skb);
L
Linus Torvalds 已提交
456 457 458
#ifdef CONFIG_XFRM
	secpath_put(skb->sp);
#endif
459 460
	if (skb->destructor) {
		WARN_ON(in_irq());
L
Linus Torvalds 已提交
461 462
		skb->destructor(skb);
	}
I
Igor Maravić 已提交
463
#if IS_ENABLED(CONFIG_NF_CONNTRACK)
464
	nf_conntrack_put(skb->nfct);
465 466
#endif
#ifdef NET_SKBUFF_NF_DEFRAG_NEEDED
467 468
	nf_conntrack_put_reasm(skb->nfct_reasm);
#endif
L
Linus Torvalds 已提交
469 470 471 472 473 474 475 476 477 478
#ifdef CONFIG_BRIDGE_NETFILTER
	nf_bridge_put(skb->nf_bridge);
#endif
/* XXX: IS this still necessary? - JHS */
#ifdef CONFIG_NET_SCHED
	skb->tc_index = 0;
#ifdef CONFIG_NET_CLS_ACT
	skb->tc_verd = 0;
#endif
#endif
479 480 481 482 483 484
}

/* Free everything but the sk_buff shell. */
static void skb_release_all(struct sk_buff *skb)
{
	skb_release_head_state(skb);
485 486 487 488 489 490 491 492 493 494 495
	skb_release_data(skb);
}

/**
 *	__kfree_skb - private function
 *	@skb: buffer
 *
 *	Free an sk_buff. Release anything attached to the buffer.
 *	Clean the state. This is an internal helper function. Users should
 *	always call kfree_skb
 */
L
Linus Torvalds 已提交
496

497 498 499
void __kfree_skb(struct sk_buff *skb)
{
	skb_release_all(skb);
L
Linus Torvalds 已提交
500 501
	kfree_skbmem(skb);
}
502
EXPORT_SYMBOL(__kfree_skb);
L
Linus Torvalds 已提交
503

504 505 506 507 508 509 510 511 512 513 514 515 516 517 518
/**
 *	kfree_skb - free an sk_buff
 *	@skb: buffer to free
 *
 *	Drop a reference to the buffer and free it if the usage count has
 *	hit zero.
 */
void kfree_skb(struct sk_buff *skb)
{
	if (unlikely(!skb))
		return;
	if (likely(atomic_read(&skb->users) == 1))
		smp_rmb();
	else if (likely(!atomic_dec_and_test(&skb->users)))
		return;
519
	trace_kfree_skb(skb, __builtin_return_address(0));
520 521
	__kfree_skb(skb);
}
522
EXPORT_SYMBOL(kfree_skb);
523

524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539
/**
 *	consume_skb - free an skbuff
 *	@skb: buffer to free
 *
 *	Drop a ref to the buffer and free it if the usage count has hit zero
 *	Functions identically to kfree_skb, but kfree_skb assumes that the frame
 *	is being dropped after a failure and notes that
 */
void consume_skb(struct sk_buff *skb)
{
	if (unlikely(!skb))
		return;
	if (likely(atomic_read(&skb->users) == 1))
		smp_rmb();
	else if (likely(!atomic_dec_and_test(&skb->users)))
		return;
540
	trace_consume_skb(skb);
541 542 543 544
	__kfree_skb(skb);
}
EXPORT_SYMBOL(consume_skb);

545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568
/**
 * 	skb_recycle - clean up an skb for reuse
 * 	@skb: buffer
 *
 * 	Recycles the skb to be reused as a receive buffer. This
 * 	function does any necessary reference count dropping, and
 * 	cleans up the skbuff as if it just came from __alloc_skb().
 */
void skb_recycle(struct sk_buff *skb)
{
	struct skb_shared_info *shinfo;

	skb_release_head_state(skb);

	shinfo = skb_shinfo(skb);
	memset(shinfo, 0, offsetof(struct skb_shared_info, dataref));
	atomic_set(&shinfo->dataref, 1);

	memset(skb, 0, offsetof(struct sk_buff, tail));
	skb->data = skb->head + NET_SKB_PAD;
	skb_reset_tail_pointer(skb);
}
EXPORT_SYMBOL(skb_recycle);

569 570 571 572 573 574 575 576 577 578 579 580
/**
 *	skb_recycle_check - check if skb can be reused for receive
 *	@skb: buffer
 *	@skb_size: minimum receive buffer size
 *
 *	Checks that the skb passed in is not shared or cloned, and
 *	that it is linear and its head portion at least as large as
 *	skb_size so that it can be recycled as a receive buffer.
 *	If these conditions are met, this function does any necessary
 *	reference count dropping and cleans up the skbuff as if it
 *	just came from __alloc_skb().
 */
581
bool skb_recycle_check(struct sk_buff *skb, int skb_size)
582
{
583
	if (!skb_is_recycleable(skb, skb_size))
584
		return false;
585

586
	skb_recycle(skb);
587

588
	return true;
589 590 591
}
EXPORT_SYMBOL(skb_recycle_check);

592 593 594 595 596 597 598
static void __copy_skb_header(struct sk_buff *new, const struct sk_buff *old)
{
	new->tstamp		= old->tstamp;
	new->dev		= old->dev;
	new->transport_header	= old->transport_header;
	new->network_header	= old->network_header;
	new->mac_header		= old->mac_header;
E
Eric Dumazet 已提交
599
	skb_dst_copy(new, old);
T
Tom Herbert 已提交
600
	new->rxhash		= old->rxhash;
601
	new->ooo_okay		= old->ooo_okay;
602
	new->l4_rxhash		= old->l4_rxhash;
603
	new->no_fcs		= old->no_fcs;
604
#ifdef CONFIG_XFRM
605 606 607
	new->sp			= secpath_get(old->sp);
#endif
	memcpy(new->cb, old->cb, sizeof(old->cb));
608
	new->csum		= old->csum;
609 610 611 612 613
	new->local_df		= old->local_df;
	new->pkt_type		= old->pkt_type;
	new->ip_summed		= old->ip_summed;
	skb_copy_queue_mapping(new, old);
	new->priority		= old->priority;
I
Igor Maravić 已提交
614
#if IS_ENABLED(CONFIG_IP_VS)
615 616 617 618
	new->ipvs_property	= old->ipvs_property;
#endif
	new->protocol		= old->protocol;
	new->mark		= old->mark;
619
	new->skb_iif		= old->skb_iif;
620
	__nf_copy(new, old);
I
Igor Maravić 已提交
621
#if IS_ENABLED(CONFIG_NETFILTER_XT_TARGET_TRACE)
622 623 624 625 626 627 628 629
	new->nf_trace		= old->nf_trace;
#endif
#ifdef CONFIG_NET_SCHED
	new->tc_index		= old->tc_index;
#ifdef CONFIG_NET_CLS_ACT
	new->tc_verd		= old->tc_verd;
#endif
#endif
630 631
	new->vlan_tci		= old->vlan_tci;

632 633 634
	skb_copy_secmark(new, old);
}

635 636 637 638
/*
 * You should not add any new code to this function.  Add it to
 * __copy_skb_header above instead.
 */
H
Herbert Xu 已提交
639
static struct sk_buff *__skb_clone(struct sk_buff *n, struct sk_buff *skb)
L
Linus Torvalds 已提交
640 641 642 643 644
{
#define C(x) n->x = skb->x

	n->next = n->prev = NULL;
	n->sk = NULL;
645 646
	__copy_skb_header(n, skb);

L
Linus Torvalds 已提交
647 648
	C(len);
	C(data_len);
649
	C(mac_len);
650
	n->hdr_len = skb->nohdr ? skb_headroom(skb) : skb->hdr_len;
651
	n->cloned = 1;
L
Linus Torvalds 已提交
652 653 654 655
	n->nohdr = 0;
	n->destructor = NULL;
	C(tail);
	C(end);
656
	C(head);
657
	C(head_frag);
658 659 660
	C(data);
	C(truesize);
	atomic_set(&n->users, 1);
L
Linus Torvalds 已提交
661 662 663 664 665

	atomic_inc(&(skb_shinfo(skb)->dataref));
	skb->cloned = 1;

	return n;
H
Herbert Xu 已提交
666 667 668 669 670 671 672 673 674 675 676 677 678 679 680
#undef C
}

/**
 *	skb_morph	-	morph one skb into another
 *	@dst: the skb to receive the contents
 *	@src: the skb to supply the contents
 *
 *	This is identical to skb_clone except that the target skb is
 *	supplied by the user.
 *
 *	The target skb is returned upon exit.
 */
struct sk_buff *skb_morph(struct sk_buff *dst, struct sk_buff *src)
{
681
	skb_release_all(dst);
H
Herbert Xu 已提交
682 683 684 685
	return __skb_clone(dst, src);
}
EXPORT_SYMBOL_GPL(skb_morph);

686 687 688 689 690 691 692 693 694 695 696 697 698 699 700
/*	skb_copy_ubufs	-	copy userspace skb frags buffers to kernel
 *	@skb: the skb to modify
 *	@gfp_mask: allocation priority
 *
 *	This must be called on SKBTX_DEV_ZEROCOPY skb.
 *	It will copy all frags into kernel and drop the reference
 *	to userspace pages.
 *
 *	If this function is called from an interrupt gfp_mask() must be
 *	%GFP_ATOMIC.
 *
 *	Returns 0 on success or a negative error code on failure
 *	to allocate kernel memory to copy to.
 */
int skb_copy_ubufs(struct sk_buff *skb, gfp_t gfp_mask)
701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719
{
	int i;
	int num_frags = skb_shinfo(skb)->nr_frags;
	struct page *page, *head = NULL;
	struct ubuf_info *uarg = skb_shinfo(skb)->destructor_arg;

	for (i = 0; i < num_frags; i++) {
		u8 *vaddr;
		skb_frag_t *f = &skb_shinfo(skb)->frags[i];

		page = alloc_page(GFP_ATOMIC);
		if (!page) {
			while (head) {
				struct page *next = (struct page *)head->private;
				put_page(head);
				head = next;
			}
			return -ENOMEM;
		}
E
Eric Dumazet 已提交
720
		vaddr = kmap_atomic(skb_frag_page(f));
721
		memcpy(page_address(page),
E
Eric Dumazet 已提交
722
		       vaddr + f->page_offset, skb_frag_size(f));
E
Eric Dumazet 已提交
723
		kunmap_atomic(vaddr);
724 725 726 727 728 729
		page->private = (unsigned long)head;
		head = page;
	}

	/* skb frags release userspace buffers */
	for (i = 0; i < skb_shinfo(skb)->nr_frags; i++)
730
		skb_frag_unref(skb, i);
731 732 733 734 735

	uarg->callback(uarg);

	/* skb frags point to kernel buffers */
	for (i = skb_shinfo(skb)->nr_frags; i > 0; i--) {
736 737
		__skb_fill_page_desc(skb, i-1, head, 0,
				     skb_shinfo(skb)->frags[i - 1].size);
738 739
		head = (struct page *)head->private;
	}
740 741

	skb_shinfo(skb)->tx_flags &= ~SKBTX_DEV_ZEROCOPY;
742 743 744 745
	return 0;
}


H
Herbert Xu 已提交
746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763
/**
 *	skb_clone	-	duplicate an sk_buff
 *	@skb: buffer to clone
 *	@gfp_mask: allocation priority
 *
 *	Duplicate an &sk_buff. The new one is not owned by a socket. Both
 *	copies share the same packet data but not structure. The new
 *	buffer has a reference count of 1. If the allocation fails the
 *	function returns %NULL otherwise the new buffer is returned.
 *
 *	If this function is called from an interrupt gfp_mask() must be
 *	%GFP_ATOMIC.
 */

struct sk_buff *skb_clone(struct sk_buff *skb, gfp_t gfp_mask)
{
	struct sk_buff *n;

764 765 766 767 768
	if (skb_shinfo(skb)->tx_flags & SKBTX_DEV_ZEROCOPY) {
		if (skb_copy_ubufs(skb, gfp_mask))
			return NULL;
	}

H
Herbert Xu 已提交
769 770 771 772 773 774 775 776 777 778
	n = skb + 1;
	if (skb->fclone == SKB_FCLONE_ORIG &&
	    n->fclone == SKB_FCLONE_UNAVAILABLE) {
		atomic_t *fclone_ref = (atomic_t *) (n + 1);
		n->fclone = SKB_FCLONE_CLONE;
		atomic_inc(fclone_ref);
	} else {
		n = kmem_cache_alloc(skbuff_head_cache, gfp_mask);
		if (!n)
			return NULL;
779 780 781

		kmemcheck_annotate_bitfield(n, flags1);
		kmemcheck_annotate_bitfield(n, flags2);
H
Herbert Xu 已提交
782 783 784 785
		n->fclone = SKB_FCLONE_UNAVAILABLE;
	}

	return __skb_clone(n, skb);
L
Linus Torvalds 已提交
786
}
787
EXPORT_SYMBOL(skb_clone);
L
Linus Torvalds 已提交
788 789 790

static void copy_skb_header(struct sk_buff *new, const struct sk_buff *old)
{
791
#ifndef NET_SKBUFF_DATA_USES_OFFSET
L
Linus Torvalds 已提交
792 793 794 795
	/*
	 *	Shift between the two data areas in bytes
	 */
	unsigned long offset = new->data - old->data;
796
#endif
797 798 799

	__copy_skb_header(new, old);

800 801 802 803
#ifndef NET_SKBUFF_DATA_USES_OFFSET
	/* {transport,network,mac}_header are relative to skb->head */
	new->transport_header += offset;
	new->network_header   += offset;
804 805
	if (skb_mac_header_was_set(new))
		new->mac_header	      += offset;
806
#endif
807 808 809
	skb_shinfo(new)->gso_size = skb_shinfo(old)->gso_size;
	skb_shinfo(new)->gso_segs = skb_shinfo(old)->gso_segs;
	skb_shinfo(new)->gso_type = skb_shinfo(old)->gso_type;
L
Linus Torvalds 已提交
810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828
}

/**
 *	skb_copy	-	create private copy of an sk_buff
 *	@skb: buffer to copy
 *	@gfp_mask: allocation priority
 *
 *	Make a copy of both an &sk_buff and its data. This is used when the
 *	caller wishes to modify the data and needs a private copy of the
 *	data to alter. Returns %NULL on failure or the pointer to the buffer
 *	on success. The returned buffer has a reference count of 1.
 *
 *	As by-product this function converts non-linear &sk_buff to linear
 *	one, so that &sk_buff becomes completely private and caller is allowed
 *	to modify all the data of returned buffer. This means that this
 *	function is not recommended for use in circumstances when only
 *	header is going to be modified. Use pskb_copy() instead.
 */

A
Al Viro 已提交
829
struct sk_buff *skb_copy(const struct sk_buff *skb, gfp_t gfp_mask)
L
Linus Torvalds 已提交
830
{
E
Eric Dumazet 已提交
831
	int headerlen = skb_headroom(skb);
832
	unsigned int size = skb_end_offset(skb) + skb->data_len;
E
Eric Dumazet 已提交
833 834
	struct sk_buff *n = alloc_skb(size, gfp_mask);

L
Linus Torvalds 已提交
835 836 837 838 839 840 841 842 843 844 845 846 847 848
	if (!n)
		return NULL;

	/* Set the data pointer */
	skb_reserve(n, headerlen);
	/* Set the tail pointer and length */
	skb_put(n, skb->len);

	if (skb_copy_bits(skb, -headerlen, n->head, headerlen + skb->len))
		BUG();

	copy_skb_header(n, skb);
	return n;
}
849
EXPORT_SYMBOL(skb_copy);
L
Linus Torvalds 已提交
850 851

/**
E
Eric Dumazet 已提交
852
 *	__pskb_copy	-	create copy of an sk_buff with private head.
L
Linus Torvalds 已提交
853
 *	@skb: buffer to copy
E
Eric Dumazet 已提交
854
 *	@headroom: headroom of new skb
L
Linus Torvalds 已提交
855 856 857 858 859 860 861 862 863 864
 *	@gfp_mask: allocation priority
 *
 *	Make a copy of both an &sk_buff and part of its data, located
 *	in header. Fragmented data remain shared. This is used when
 *	the caller wishes to modify only header of &sk_buff and needs
 *	private copy of the header to alter. Returns %NULL on failure
 *	or the pointer to the buffer on success.
 *	The returned buffer has a reference count of 1.
 */

E
Eric Dumazet 已提交
865
struct sk_buff *__pskb_copy(struct sk_buff *skb, int headroom, gfp_t gfp_mask)
L
Linus Torvalds 已提交
866
{
E
Eric Dumazet 已提交
867
	unsigned int size = skb_headlen(skb) + headroom;
E
Eric Dumazet 已提交
868 869
	struct sk_buff *n = alloc_skb(size, gfp_mask);

L
Linus Torvalds 已提交
870 871 872 873
	if (!n)
		goto out;

	/* Set the data pointer */
E
Eric Dumazet 已提交
874
	skb_reserve(n, headroom);
L
Linus Torvalds 已提交
875 876 877
	/* Set the tail pointer and length */
	skb_put(n, skb_headlen(skb));
	/* Copy the bytes */
878
	skb_copy_from_linear_data(skb, n->data, n->len);
L
Linus Torvalds 已提交
879

H
Herbert Xu 已提交
880
	n->truesize += skb->data_len;
L
Linus Torvalds 已提交
881 882 883 884 885 886
	n->data_len  = skb->data_len;
	n->len	     = skb->len;

	if (skb_shinfo(skb)->nr_frags) {
		int i;

887 888
		if (skb_shinfo(skb)->tx_flags & SKBTX_DEV_ZEROCOPY) {
			if (skb_copy_ubufs(skb, gfp_mask)) {
889 890
				kfree_skb(n);
				n = NULL;
891 892 893
				goto out;
			}
		}
L
Linus Torvalds 已提交
894 895
		for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
			skb_shinfo(n)->frags[i] = skb_shinfo(skb)->frags[i];
896
			skb_frag_ref(skb, i);
L
Linus Torvalds 已提交
897 898 899 900
		}
		skb_shinfo(n)->nr_frags = i;
	}

901
	if (skb_has_frag_list(skb)) {
L
Linus Torvalds 已提交
902 903 904 905 906 907 908 909
		skb_shinfo(n)->frag_list = skb_shinfo(skb)->frag_list;
		skb_clone_fraglist(n);
	}

	copy_skb_header(n, skb);
out:
	return n;
}
E
Eric Dumazet 已提交
910
EXPORT_SYMBOL(__pskb_copy);
L
Linus Torvalds 已提交
911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927

/**
 *	pskb_expand_head - reallocate header of &sk_buff
 *	@skb: buffer to reallocate
 *	@nhead: room to add at head
 *	@ntail: room to add at tail
 *	@gfp_mask: allocation priority
 *
 *	Expands (or creates identical copy, if &nhead and &ntail are zero)
 *	header of skb. &sk_buff itself is not changed. &sk_buff MUST have
 *	reference count of 1. Returns zero in the case of success or error,
 *	if expansion failed. In the last case, &sk_buff is not changed.
 *
 *	All the pointers pointing into skb header may change and must be
 *	reloaded after call to this function.
 */

V
Victor Fusco 已提交
928
int pskb_expand_head(struct sk_buff *skb, int nhead, int ntail,
A
Al Viro 已提交
929
		     gfp_t gfp_mask)
L
Linus Torvalds 已提交
930 931 932
{
	int i;
	u8 *data;
933
	int size = nhead + skb_end_offset(skb) + ntail;
L
Linus Torvalds 已提交
934 935
	long off;

936 937
	BUG_ON(nhead < 0);

L
Linus Torvalds 已提交
938 939 940 941 942
	if (skb_shared(skb))
		BUG();

	size = SKB_DATA_ALIGN(size);

943 944
	data = kmalloc(size + SKB_DATA_ALIGN(sizeof(struct skb_shared_info)),
		       gfp_mask);
L
Linus Torvalds 已提交
945 946
	if (!data)
		goto nodata;
947
	size = SKB_WITH_OVERHEAD(ksize(data));
L
Linus Torvalds 已提交
948 949

	/* Copy only real data... and, alas, header. This should be
E
Eric Dumazet 已提交
950 951 952 953 954 955
	 * optimized for the cases when header is void.
	 */
	memcpy(data + nhead, skb->head, skb_tail_pointer(skb) - skb->head);

	memcpy((struct skb_shared_info *)(data + size),
	       skb_shinfo(skb),
956
	       offsetof(struct skb_shared_info, frags[skb_shinfo(skb)->nr_frags]));
L
Linus Torvalds 已提交
957

958 959 960 961 962 963
	/*
	 * if shinfo is shared we must drop the old head gracefully, but if it
	 * is not we can just drop the old head and let the existing refcount
	 * be since all we did is relocate the values
	 */
	if (skb_cloned(skb)) {
964 965 966 967 968
		/* copy this zero copy skb frags */
		if (skb_shinfo(skb)->tx_flags & SKBTX_DEV_ZEROCOPY) {
			if (skb_copy_ubufs(skb, gfp_mask))
				goto nofrags;
		}
969
		for (i = 0; i < skb_shinfo(skb)->nr_frags; i++)
970
			skb_frag_ref(skb, i);
L
Linus Torvalds 已提交
971

972 973
		if (skb_has_frag_list(skb))
			skb_clone_fraglist(skb);
L
Linus Torvalds 已提交
974

975
		skb_release_data(skb);
976 977
	} else {
		skb_free_head(skb);
978
	}
L
Linus Torvalds 已提交
979 980 981
	off = (data + nhead) - skb->head;

	skb->head     = data;
982
	skb->head_frag = 0;
L
Linus Torvalds 已提交
983
	skb->data    += off;
984 985
#ifdef NET_SKBUFF_DATA_USES_OFFSET
	skb->end      = size;
986
	off           = nhead;
987 988
#else
	skb->end      = skb->head + size;
989
#endif
990 991
	/* {transport,network,mac}_header and tail are relative to skb->head */
	skb->tail	      += off;
992 993
	skb->transport_header += off;
	skb->network_header   += off;
994 995
	if (skb_mac_header_was_set(skb))
		skb->mac_header += off;
996 997 998
	/* Only adjust this if it actually is csum_start rather than csum */
	if (skb->ip_summed == CHECKSUM_PARTIAL)
		skb->csum_start += nhead;
L
Linus Torvalds 已提交
999
	skb->cloned   = 0;
1000
	skb->hdr_len  = 0;
L
Linus Torvalds 已提交
1001 1002 1003 1004
	skb->nohdr    = 0;
	atomic_set(&skb_shinfo(skb)->dataref, 1);
	return 0;

1005 1006
nofrags:
	kfree(data);
L
Linus Torvalds 已提交
1007 1008 1009
nodata:
	return -ENOMEM;
}
1010
EXPORT_SYMBOL(pskb_expand_head);
L
Linus Torvalds 已提交
1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030

/* Make private copy of skb with writable head and some headroom */

struct sk_buff *skb_realloc_headroom(struct sk_buff *skb, unsigned int headroom)
{
	struct sk_buff *skb2;
	int delta = headroom - skb_headroom(skb);

	if (delta <= 0)
		skb2 = pskb_copy(skb, GFP_ATOMIC);
	else {
		skb2 = skb_clone(skb, GFP_ATOMIC);
		if (skb2 && pskb_expand_head(skb2, SKB_DATA_ALIGN(delta), 0,
					     GFP_ATOMIC)) {
			kfree_skb(skb2);
			skb2 = NULL;
		}
	}
	return skb2;
}
1031
EXPORT_SYMBOL(skb_realloc_headroom);
L
Linus Torvalds 已提交
1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051

/**
 *	skb_copy_expand	-	copy and expand sk_buff
 *	@skb: buffer to copy
 *	@newheadroom: new free bytes at head
 *	@newtailroom: new free bytes at tail
 *	@gfp_mask: allocation priority
 *
 *	Make a copy of both an &sk_buff and its data and while doing so
 *	allocate additional space.
 *
 *	This is used when the caller wishes to modify the data and needs a
 *	private copy of the data to alter as well as more space for new fields.
 *	Returns %NULL on failure or the pointer to the buffer
 *	on success. The returned buffer has a reference count of 1.
 *
 *	You must pass %GFP_ATOMIC as the allocation priority if this function
 *	is called from an interrupt.
 */
struct sk_buff *skb_copy_expand(const struct sk_buff *skb,
V
Victor Fusco 已提交
1052
				int newheadroom, int newtailroom,
A
Al Viro 已提交
1053
				gfp_t gfp_mask)
L
Linus Torvalds 已提交
1054 1055 1056 1057 1058 1059
{
	/*
	 *	Allocate the copy buffer
	 */
	struct sk_buff *n = alloc_skb(newheadroom + skb->len + newtailroom,
				      gfp_mask);
1060
	int oldheadroom = skb_headroom(skb);
L
Linus Torvalds 已提交
1061
	int head_copy_len, head_copy_off;
1062
	int off;
L
Linus Torvalds 已提交
1063 1064 1065 1066 1067 1068 1069 1070 1071

	if (!n)
		return NULL;

	skb_reserve(n, newheadroom);

	/* Set the tail pointer and length */
	skb_put(n, skb->len);

1072
	head_copy_len = oldheadroom;
L
Linus Torvalds 已提交
1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085
	head_copy_off = 0;
	if (newheadroom <= head_copy_len)
		head_copy_len = newheadroom;
	else
		head_copy_off = newheadroom - head_copy_len;

	/* Copy the linear header and data. */
	if (skb_copy_bits(skb, -head_copy_len, n->head + head_copy_off,
			  skb->len + head_copy_len))
		BUG();

	copy_skb_header(n, skb);

1086
	off                  = newheadroom - oldheadroom;
1087 1088
	if (n->ip_summed == CHECKSUM_PARTIAL)
		n->csum_start += off;
1089
#ifdef NET_SKBUFF_DATA_USES_OFFSET
1090 1091
	n->transport_header += off;
	n->network_header   += off;
1092 1093
	if (skb_mac_header_was_set(skb))
		n->mac_header += off;
1094
#endif
1095

L
Linus Torvalds 已提交
1096 1097
	return n;
}
1098
EXPORT_SYMBOL(skb_copy_expand);
L
Linus Torvalds 已提交
1099 1100 1101 1102 1103 1104 1105 1106 1107 1108

/**
 *	skb_pad			-	zero pad the tail of an skb
 *	@skb: buffer to pad
 *	@pad: space to pad
 *
 *	Ensure that a buffer is followed by a padding area that is zero
 *	filled. Used by network drivers which may DMA or transfer data
 *	beyond the buffer end onto the wire.
 *
1109
 *	May return error in out of memory cases. The skb is freed on error.
L
Linus Torvalds 已提交
1110
 */
1111

1112
int skb_pad(struct sk_buff *skb, int pad)
L
Linus Torvalds 已提交
1113
{
1114 1115
	int err;
	int ntail;
1116

L
Linus Torvalds 已提交
1117
	/* If the skbuff is non linear tailroom is always zero.. */
1118
	if (!skb_cloned(skb) && skb_tailroom(skb) >= pad) {
L
Linus Torvalds 已提交
1119
		memset(skb->data+skb->len, 0, pad);
1120
		return 0;
L
Linus Torvalds 已提交
1121
	}
1122

1123
	ntail = skb->data_len + pad - (skb->end - skb->tail);
1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140
	if (likely(skb_cloned(skb) || ntail > 0)) {
		err = pskb_expand_head(skb, 0, ntail, GFP_ATOMIC);
		if (unlikely(err))
			goto free_skb;
	}

	/* FIXME: The use of this function with non-linear skb's really needs
	 * to be audited.
	 */
	err = skb_linearize(skb);
	if (unlikely(err))
		goto free_skb;

	memset(skb->data + skb->len, 0, pad);
	return 0;

free_skb:
L
Linus Torvalds 已提交
1141
	kfree_skb(skb);
1142
	return err;
1143
}
1144
EXPORT_SYMBOL(skb_pad);
1145

1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166
/**
 *	skb_put - add data to a buffer
 *	@skb: buffer to use
 *	@len: amount of data to add
 *
 *	This function extends the used data area of the buffer. If this would
 *	exceed the total buffer size the kernel will panic. A pointer to the
 *	first byte of the extra data is returned.
 */
unsigned char *skb_put(struct sk_buff *skb, unsigned int len)
{
	unsigned char *tmp = skb_tail_pointer(skb);
	SKB_LINEAR_ASSERT(skb);
	skb->tail += len;
	skb->len  += len;
	if (unlikely(skb->tail > skb->end))
		skb_over_panic(skb, len, __builtin_return_address(0));
	return tmp;
}
EXPORT_SYMBOL(skb_put);

1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185
/**
 *	skb_push - add data to the start of a buffer
 *	@skb: buffer to use
 *	@len: amount of data to add
 *
 *	This function extends the used data area of the buffer at the buffer
 *	start. If this would exceed the total buffer headroom the kernel will
 *	panic. A pointer to the first byte of the extra data is returned.
 */
unsigned char *skb_push(struct sk_buff *skb, unsigned int len)
{
	skb->data -= len;
	skb->len  += len;
	if (unlikely(skb->data<skb->head))
		skb_under_panic(skb, len, __builtin_return_address(0));
	return skb->data;
}
EXPORT_SYMBOL(skb_push);

1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197
/**
 *	skb_pull - remove data from the start of a buffer
 *	@skb: buffer to use
 *	@len: amount of data to remove
 *
 *	This function removes data from the start of a buffer, returning
 *	the memory to the headroom. A pointer to the next data in the buffer
 *	is returned. Once the data has been pulled future pushes will overwrite
 *	the old data.
 */
unsigned char *skb_pull(struct sk_buff *skb, unsigned int len)
{
1198
	return skb_pull_inline(skb, len);
1199 1200 1201
}
EXPORT_SYMBOL(skb_pull);

1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217
/**
 *	skb_trim - remove end from a buffer
 *	@skb: buffer to alter
 *	@len: new length
 *
 *	Cut the length of a buffer down by removing data from the tail. If
 *	the buffer is already under the length specified it is not modified.
 *	The skb must be linear.
 */
void skb_trim(struct sk_buff *skb, unsigned int len)
{
	if (skb->len > len)
		__skb_trim(skb, len);
}
EXPORT_SYMBOL(skb_trim);

1218
/* Trims skb to length len. It can change skb pointers.
L
Linus Torvalds 已提交
1219 1220
 */

1221
int ___pskb_trim(struct sk_buff *skb, unsigned int len)
L
Linus Torvalds 已提交
1222
{
1223 1224
	struct sk_buff **fragp;
	struct sk_buff *frag;
L
Linus Torvalds 已提交
1225 1226 1227
	int offset = skb_headlen(skb);
	int nfrags = skb_shinfo(skb)->nr_frags;
	int i;
1228 1229 1230 1231 1232
	int err;

	if (skb_cloned(skb) &&
	    unlikely((err = pskb_expand_head(skb, 0, 0, GFP_ATOMIC))))
		return err;
L
Linus Torvalds 已提交
1233

1234 1235 1236 1237 1238
	i = 0;
	if (offset >= len)
		goto drop_pages;

	for (; i < nfrags; i++) {
E
Eric Dumazet 已提交
1239
		int end = offset + skb_frag_size(&skb_shinfo(skb)->frags[i]);
1240 1241 1242 1243 1244 1245

		if (end < len) {
			offset = end;
			continue;
		}

E
Eric Dumazet 已提交
1246
		skb_frag_size_set(&skb_shinfo(skb)->frags[i++], len - offset);
1247

1248
drop_pages:
1249 1250 1251
		skb_shinfo(skb)->nr_frags = i;

		for (; i < nfrags; i++)
1252
			skb_frag_unref(skb, i);
1253

1254
		if (skb_has_frag_list(skb))
1255
			skb_drop_fraglist(skb);
1256
		goto done;
1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270
	}

	for (fragp = &skb_shinfo(skb)->frag_list; (frag = *fragp);
	     fragp = &frag->next) {
		int end = offset + frag->len;

		if (skb_shared(frag)) {
			struct sk_buff *nfrag;

			nfrag = skb_clone(frag, GFP_ATOMIC);
			if (unlikely(!nfrag))
				return -ENOMEM;

			nfrag->next = frag->next;
1271
			consume_skb(frag);
1272 1273
			frag = nfrag;
			*fragp = frag;
L
Linus Torvalds 已提交
1274
		}
1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287

		if (end < len) {
			offset = end;
			continue;
		}

		if (end > len &&
		    unlikely((err = pskb_trim(frag, len - offset))))
			return err;

		if (frag->next)
			skb_drop_list(&frag->next);
		break;
L
Linus Torvalds 已提交
1288 1289
	}

1290
done:
1291
	if (len > skb_headlen(skb)) {
L
Linus Torvalds 已提交
1292 1293 1294
		skb->data_len -= skb->len - len;
		skb->len       = len;
	} else {
1295 1296
		skb->len       = len;
		skb->data_len  = 0;
1297
		skb_set_tail_pointer(skb, len);
L
Linus Torvalds 已提交
1298 1299 1300 1301
	}

	return 0;
}
1302
EXPORT_SYMBOL(___pskb_trim);
L
Linus Torvalds 已提交
1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334

/**
 *	__pskb_pull_tail - advance tail of skb header
 *	@skb: buffer to reallocate
 *	@delta: number of bytes to advance tail
 *
 *	The function makes a sense only on a fragmented &sk_buff,
 *	it expands header moving its tail forward and copying necessary
 *	data from fragmented part.
 *
 *	&sk_buff MUST have reference count of 1.
 *
 *	Returns %NULL (and &sk_buff does not change) if pull failed
 *	or value of new tail of skb in the case of success.
 *
 *	All the pointers pointing into skb header may change and must be
 *	reloaded after call to this function.
 */

/* Moves tail of skb head forward, copying data from fragmented part,
 * when it is necessary.
 * 1. It may fail due to malloc failure.
 * 2. It may change skb pointers.
 *
 * It is pretty complicated. Luckily, it is called only in exceptional cases.
 */
unsigned char *__pskb_pull_tail(struct sk_buff *skb, int delta)
{
	/* If skb has not enough free space at tail, get new one
	 * plus 128 bytes for future expansions. If we have enough
	 * room at tail, reallocate without expansion only if skb is cloned.
	 */
1335
	int i, k, eat = (skb->tail + delta) - skb->end;
L
Linus Torvalds 已提交
1336 1337 1338 1339 1340 1341 1342

	if (eat > 0 || skb_cloned(skb)) {
		if (pskb_expand_head(skb, 0, eat > 0 ? eat + 128 : 0,
				     GFP_ATOMIC))
			return NULL;
	}

1343
	if (skb_copy_bits(skb, skb_headlen(skb), skb_tail_pointer(skb), delta))
L
Linus Torvalds 已提交
1344 1345 1346 1347 1348
		BUG();

	/* Optimization: no fragments, no reasons to preestimate
	 * size of pulled pages. Superb.
	 */
1349
	if (!skb_has_frag_list(skb))
L
Linus Torvalds 已提交
1350 1351 1352 1353 1354
		goto pull_pages;

	/* Estimate size of pulled pages. */
	eat = delta;
	for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
E
Eric Dumazet 已提交
1355 1356 1357
		int size = skb_frag_size(&skb_shinfo(skb)->frags[i]);

		if (size >= eat)
L
Linus Torvalds 已提交
1358
			goto pull_pages;
E
Eric Dumazet 已提交
1359
		eat -= size;
L
Linus Torvalds 已提交
1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374
	}

	/* If we need update frag list, we are in troubles.
	 * Certainly, it possible to add an offset to skb data,
	 * but taking into account that pulling is expected to
	 * be very rare operation, it is worth to fight against
	 * further bloating skb head and crucify ourselves here instead.
	 * Pure masohism, indeed. 8)8)
	 */
	if (eat) {
		struct sk_buff *list = skb_shinfo(skb)->frag_list;
		struct sk_buff *clone = NULL;
		struct sk_buff *insp = NULL;

		do {
1375
			BUG_ON(!list);
L
Linus Torvalds 已提交
1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397

			if (list->len <= eat) {
				/* Eaten as whole. */
				eat -= list->len;
				list = list->next;
				insp = list;
			} else {
				/* Eaten partially. */

				if (skb_shared(list)) {
					/* Sucks! We need to fork list. :-( */
					clone = skb_clone(list, GFP_ATOMIC);
					if (!clone)
						return NULL;
					insp = list->next;
					list = clone;
				} else {
					/* This may be pulled without
					 * problems. */
					insp = list;
				}
				if (!pskb_pull(list, eat)) {
1398
					kfree_skb(clone);
L
Linus Torvalds 已提交
1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421
					return NULL;
				}
				break;
			}
		} while (eat);

		/* Free pulled out fragments. */
		while ((list = skb_shinfo(skb)->frag_list) != insp) {
			skb_shinfo(skb)->frag_list = list->next;
			kfree_skb(list);
		}
		/* And insert new clone at head. */
		if (clone) {
			clone->next = list;
			skb_shinfo(skb)->frag_list = clone;
		}
	}
	/* Success! Now we may commit changes to skb data. */

pull_pages:
	eat = delta;
	k = 0;
	for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
E
Eric Dumazet 已提交
1422 1423 1424
		int size = skb_frag_size(&skb_shinfo(skb)->frags[i]);

		if (size <= eat) {
1425
			skb_frag_unref(skb, i);
E
Eric Dumazet 已提交
1426
			eat -= size;
L
Linus Torvalds 已提交
1427 1428 1429 1430
		} else {
			skb_shinfo(skb)->frags[k] = skb_shinfo(skb)->frags[i];
			if (eat) {
				skb_shinfo(skb)->frags[k].page_offset += eat;
E
Eric Dumazet 已提交
1431
				skb_frag_size_sub(&skb_shinfo(skb)->frags[k], eat);
L
Linus Torvalds 已提交
1432 1433 1434 1435 1436 1437 1438 1439 1440 1441
				eat = 0;
			}
			k++;
		}
	}
	skb_shinfo(skb)->nr_frags = k;

	skb->tail     += delta;
	skb->data_len -= delta;

1442
	return skb_tail_pointer(skb);
L
Linus Torvalds 已提交
1443
}
1444
EXPORT_SYMBOL(__pskb_pull_tail);
L
Linus Torvalds 已提交
1445

1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460
/**
 *	skb_copy_bits - copy bits from skb to kernel buffer
 *	@skb: source skb
 *	@offset: offset in source
 *	@to: destination buffer
 *	@len: number of bytes to copy
 *
 *	Copy the specified number of bytes from the source skb to the
 *	destination buffer.
 *
 *	CAUTION ! :
 *		If its prototype is ever changed,
 *		check arch/{*}/net/{*}.S files,
 *		since it is called from BPF assembly code.
 */
L
Linus Torvalds 已提交
1461 1462
int skb_copy_bits(const struct sk_buff *skb, int offset, void *to, int len)
{
1463
	int start = skb_headlen(skb);
1464 1465
	struct sk_buff *frag_iter;
	int i, copy;
L
Linus Torvalds 已提交
1466 1467 1468 1469 1470

	if (offset > (int)skb->len - len)
		goto fault;

	/* Copy header. */
1471
	if ((copy = start - offset) > 0) {
L
Linus Torvalds 已提交
1472 1473
		if (copy > len)
			copy = len;
1474
		skb_copy_from_linear_data_offset(skb, offset, to, copy);
L
Linus Torvalds 已提交
1475 1476 1477 1478 1479 1480 1481
		if ((len -= copy) == 0)
			return 0;
		offset += copy;
		to     += copy;
	}

	for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
1482
		int end;
E
Eric Dumazet 已提交
1483
		skb_frag_t *f = &skb_shinfo(skb)->frags[i];
L
Linus Torvalds 已提交
1484

1485
		WARN_ON(start > offset + len);
1486

E
Eric Dumazet 已提交
1487
		end = start + skb_frag_size(f);
L
Linus Torvalds 已提交
1488 1489 1490 1491 1492 1493
		if ((copy = end - offset) > 0) {
			u8 *vaddr;

			if (copy > len)
				copy = len;

E
Eric Dumazet 已提交
1494
			vaddr = kmap_atomic(skb_frag_page(f));
L
Linus Torvalds 已提交
1495
			memcpy(to,
E
Eric Dumazet 已提交
1496 1497 1498
			       vaddr + f->page_offset + offset - start,
			       copy);
			kunmap_atomic(vaddr);
L
Linus Torvalds 已提交
1499 1500 1501 1502 1503 1504

			if ((len -= copy) == 0)
				return 0;
			offset += copy;
			to     += copy;
		}
1505
		start = end;
L
Linus Torvalds 已提交
1506 1507
	}

1508 1509
	skb_walk_frags(skb, frag_iter) {
		int end;
L
Linus Torvalds 已提交
1510

1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522
		WARN_ON(start > offset + len);

		end = start + frag_iter->len;
		if ((copy = end - offset) > 0) {
			if (copy > len)
				copy = len;
			if (skb_copy_bits(frag_iter, offset - start, to, copy))
				goto fault;
			if ((len -= copy) == 0)
				return 0;
			offset += copy;
			to     += copy;
L
Linus Torvalds 已提交
1523
		}
1524
		start = end;
L
Linus Torvalds 已提交
1525
	}
1526

L
Linus Torvalds 已提交
1527 1528 1529 1530 1531 1532
	if (!len)
		return 0;

fault:
	return -EFAULT;
}
1533
EXPORT_SYMBOL(skb_copy_bits);
L
Linus Torvalds 已提交
1534

J
Jens Axboe 已提交
1535 1536 1537 1538 1539 1540
/*
 * Callback from splice_to_pipe(), if we need to release some pages
 * at the end of the spd in case we error'ed out in filling the pipe.
 */
static void sock_spd_release(struct splice_pipe_desc *spd, unsigned int i)
{
1541 1542
	put_page(spd->pages[i]);
}
J
Jens Axboe 已提交
1543

1544 1545 1546
static struct page *linear_to_page(struct page *page, unsigned int *len,
				   unsigned int *offset,
				   struct sk_buff *skb, struct sock *sk)
1547
{
1548 1549 1550 1551 1552 1553 1554 1555
	struct page *p = sk->sk_sndmsg_page;
	unsigned int off;

	if (!p) {
new_page:
		p = sk->sk_sndmsg_page = alloc_pages(sk->sk_allocation, 0);
		if (!p)
			return NULL;
1556

1557 1558 1559 1560 1561
		off = sk->sk_sndmsg_off = 0;
		/* hold one ref to this page until it's full */
	} else {
		unsigned int mlen;

1562 1563 1564
		/* If we are the only user of the page, we can reset offset */
		if (page_count(p) == 1)
			sk->sk_sndmsg_off = 0;
1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577
		off = sk->sk_sndmsg_off;
		mlen = PAGE_SIZE - off;
		if (mlen < 64 && mlen < *len) {
			put_page(p);
			goto new_page;
		}

		*len = min_t(unsigned int, *len, mlen);
	}

	memcpy(page_address(p) + off, page_address(page) + *offset, *len);
	sk->sk_sndmsg_off += *len;
	*offset = off;
1578 1579

	return p;
J
Jens Axboe 已提交
1580 1581
}

E
Eric Dumazet 已提交
1582 1583 1584 1585 1586 1587 1588 1589 1590 1591
static bool spd_can_coalesce(const struct splice_pipe_desc *spd,
			     struct page *page,
			     unsigned int offset)
{
	return	spd->nr_pages &&
		spd->pages[spd->nr_pages - 1] == page &&
		(spd->partial[spd->nr_pages - 1].offset +
		 spd->partial[spd->nr_pages - 1].len == offset);
}

J
Jens Axboe 已提交
1592 1593 1594
/*
 * Fill page/offset/length into spd, if it can hold more pages.
 */
1595 1596 1597
static bool spd_fill_page(struct splice_pipe_desc *spd,
			  struct pipe_inode_info *pipe, struct page *page,
			  unsigned int *len, unsigned int offset,
1598
			  struct sk_buff *skb, bool linear,
1599
			  struct sock *sk)
J
Jens Axboe 已提交
1600
{
E
Eric Dumazet 已提交
1601
	if (unlikely(spd->nr_pages == MAX_SKB_FRAGS))
1602
		return true;
J
Jens Axboe 已提交
1603

1604
	if (linear) {
1605
		page = linear_to_page(page, len, &offset, skb, sk);
1606
		if (!page)
1607
			return true;
E
Eric Dumazet 已提交
1608 1609 1610
	}
	if (spd_can_coalesce(spd, page, offset)) {
		spd->partial[spd->nr_pages - 1].len += *len;
1611
		return false;
E
Eric Dumazet 已提交
1612 1613
	}
	get_page(page);
J
Jens Axboe 已提交
1614
	spd->pages[spd->nr_pages] = page;
1615
	spd->partial[spd->nr_pages].len = *len;
J
Jens Axboe 已提交
1616 1617
	spd->partial[spd->nr_pages].offset = offset;
	spd->nr_pages++;
1618

1619
	return false;
J
Jens Axboe 已提交
1620 1621
}

1622 1623 1624
static inline void __segment_seek(struct page **page, unsigned int *poff,
				  unsigned int *plen, unsigned int off)
{
1625 1626
	unsigned long n;

1627
	*poff += off;
1628 1629 1630 1631
	n = *poff / PAGE_SIZE;
	if (n)
		*page = nth_page(*page, n);

1632 1633 1634 1635
	*poff = *poff % PAGE_SIZE;
	*plen -= off;
}

1636 1637 1638
static bool __splice_segment(struct page *page, unsigned int poff,
			     unsigned int plen, unsigned int *off,
			     unsigned int *len, struct sk_buff *skb,
1639
			     struct splice_pipe_desc *spd, bool linear,
1640 1641
			     struct sock *sk,
			     struct pipe_inode_info *pipe)
J
Jens Axboe 已提交
1642
{
1643
	if (!*len)
1644
		return true;
1645 1646 1647 1648

	/* skip this segment if already processed */
	if (*off >= plen) {
		*off -= plen;
1649
		return false;
1650
	}
J
Jens Axboe 已提交
1651

1652 1653 1654 1655
	/* ignore any bits we already processed */
	if (*off) {
		__segment_seek(&page, &poff, &plen, *off);
		*off = 0;
J
Jens Axboe 已提交
1656 1657
	}

1658 1659
	do {
		unsigned int flen = min(*len, plen);
J
Jens Axboe 已提交
1660

1661 1662
		/* the linear region may spread across several pages  */
		flen = min_t(unsigned int, flen, PAGE_SIZE - poff);
J
Jens Axboe 已提交
1663

1664
		if (spd_fill_page(spd, pipe, page, &flen, poff, skb, linear, sk))
1665
			return true;
J
Jens Axboe 已提交
1666

1667 1668 1669 1670 1671
		__segment_seek(&page, &poff, &plen, flen);
		*len -= flen;

	} while (*len && plen);

1672
	return false;
1673 1674 1675
}

/*
1676
 * Map linear and fragment data from the skb to spd. It reports true if the
1677 1678
 * pipe is full or if we already spliced the requested length.
 */
1679 1680 1681
static bool __skb_splice_bits(struct sk_buff *skb, struct pipe_inode_info *pipe,
			      unsigned int *offset, unsigned int *len,
			      struct splice_pipe_desc *spd, struct sock *sk)
1682 1683 1684
{
	int seg;

1685
	/* map the linear part :
1686 1687 1688
	 * If skb->head_frag is set, this 'linear' part is backed by a
	 * fragment, and if the head is not shared with any clones then
	 * we can avoid a copy since we own the head portion of this page.
1689 1690 1691 1692
	 */
	if (__splice_segment(virt_to_page(skb->data),
			     (unsigned long) skb->data & (PAGE_SIZE - 1),
			     skb_headlen(skb),
1693
			     offset, len, skb, spd,
1694
			     skb_head_is_locked(skb),
1695
			     sk, pipe))
1696
		return true;
J
Jens Axboe 已提交
1697 1698 1699 1700 1701 1702 1703

	/*
	 * then map the fragments
	 */
	for (seg = 0; seg < skb_shinfo(skb)->nr_frags; seg++) {
		const skb_frag_t *f = &skb_shinfo(skb)->frags[seg];

1704
		if (__splice_segment(skb_frag_page(f),
E
Eric Dumazet 已提交
1705
				     f->page_offset, skb_frag_size(f),
1706
				     offset, len, skb, spd, false, sk, pipe))
1707
			return true;
J
Jens Axboe 已提交
1708 1709
	}

1710
	return false;
J
Jens Axboe 已提交
1711 1712 1713 1714 1715 1716 1717 1718
}

/*
 * Map data from the skb to a pipe. Should handle both the linear part,
 * the fragments, and the frag list. It does NOT handle frag lists within
 * the frag list, if such a thing exists. We'd probably need to recurse to
 * handle that cleanly.
 */
1719
int skb_splice_bits(struct sk_buff *skb, unsigned int offset,
J
Jens Axboe 已提交
1720 1721 1722
		    struct pipe_inode_info *pipe, unsigned int tlen,
		    unsigned int flags)
{
E
Eric Dumazet 已提交
1723 1724
	struct partial_page partial[MAX_SKB_FRAGS];
	struct page *pages[MAX_SKB_FRAGS];
J
Jens Axboe 已提交
1725 1726 1727 1728 1729 1730 1731
	struct splice_pipe_desc spd = {
		.pages = pages,
		.partial = partial,
		.flags = flags,
		.ops = &sock_pipe_buf_ops,
		.spd_release = sock_spd_release,
	};
1732
	struct sk_buff *frag_iter;
1733
	struct sock *sk = skb->sk;
1734 1735
	int ret = 0;

J
Jens Axboe 已提交
1736 1737 1738 1739
	/*
	 * __skb_splice_bits() only fails if the output has no room left,
	 * so no point in going over the frag_list for the error case.
	 */
1740
	if (__skb_splice_bits(skb, pipe, &offset, &tlen, &spd, sk))
J
Jens Axboe 已提交
1741 1742 1743 1744 1745 1746 1747
		goto done;
	else if (!tlen)
		goto done;

	/*
	 * now see if we have a frag_list to map
	 */
1748 1749 1750
	skb_walk_frags(skb, frag_iter) {
		if (!tlen)
			break;
1751
		if (__skb_splice_bits(frag_iter, pipe, &offset, &tlen, &spd, sk))
1752
			break;
J
Jens Axboe 已提交
1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765
	}

done:
	if (spd.nr_pages) {
		/*
		 * Drop the socket lock, otherwise we have reverse
		 * locking dependencies between sk_lock and i_mutex
		 * here as compared to sendfile(). We enter here
		 * with the socket lock held, and splice_to_pipe() will
		 * grab the pipe inode lock. For sendfile() emulation,
		 * we call into ->sendpage() with the i_mutex lock held
		 * and networking will grab the socket lock.
		 */
1766
		release_sock(sk);
J
Jens Axboe 已提交
1767
		ret = splice_to_pipe(pipe, &spd);
1768
		lock_sock(sk);
J
Jens Axboe 已提交
1769 1770
	}

1771
	return ret;
J
Jens Axboe 已提交
1772 1773
}

1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785
/**
 *	skb_store_bits - store bits from kernel buffer to skb
 *	@skb: destination buffer
 *	@offset: offset in destination
 *	@from: source buffer
 *	@len: number of bytes to copy
 *
 *	Copy the specified number of bytes from the source buffer to the
 *	destination skb.  This function handles all the messy bits of
 *	traversing fragment lists and such.
 */

1786
int skb_store_bits(struct sk_buff *skb, int offset, const void *from, int len)
1787
{
1788
	int start = skb_headlen(skb);
1789 1790
	struct sk_buff *frag_iter;
	int i, copy;
1791 1792 1793 1794

	if (offset > (int)skb->len - len)
		goto fault;

1795
	if ((copy = start - offset) > 0) {
1796 1797
		if (copy > len)
			copy = len;
1798
		skb_copy_to_linear_data_offset(skb, offset, from, copy);
1799 1800 1801 1802 1803 1804 1805 1806
		if ((len -= copy) == 0)
			return 0;
		offset += copy;
		from += copy;
	}

	for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
		skb_frag_t *frag = &skb_shinfo(skb)->frags[i];
1807 1808
		int end;

1809
		WARN_ON(start > offset + len);
1810

E
Eric Dumazet 已提交
1811
		end = start + skb_frag_size(frag);
1812 1813 1814 1815 1816 1817
		if ((copy = end - offset) > 0) {
			u8 *vaddr;

			if (copy > len)
				copy = len;

E
Eric Dumazet 已提交
1818
			vaddr = kmap_atomic(skb_frag_page(frag));
1819 1820
			memcpy(vaddr + frag->page_offset + offset - start,
			       from, copy);
E
Eric Dumazet 已提交
1821
			kunmap_atomic(vaddr);
1822 1823 1824 1825 1826 1827

			if ((len -= copy) == 0)
				return 0;
			offset += copy;
			from += copy;
		}
1828
		start = end;
1829 1830
	}

1831 1832
	skb_walk_frags(skb, frag_iter) {
		int end;
1833

1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846
		WARN_ON(start > offset + len);

		end = start + frag_iter->len;
		if ((copy = end - offset) > 0) {
			if (copy > len)
				copy = len;
			if (skb_store_bits(frag_iter, offset - start,
					   from, copy))
				goto fault;
			if ((len -= copy) == 0)
				return 0;
			offset += copy;
			from += copy;
1847
		}
1848
		start = end;
1849 1850 1851 1852 1853 1854 1855 1856 1857
	}
	if (!len)
		return 0;

fault:
	return -EFAULT;
}
EXPORT_SYMBOL(skb_store_bits);

L
Linus Torvalds 已提交
1858 1859
/* Checksum skb data. */

1860 1861
__wsum skb_checksum(const struct sk_buff *skb, int offset,
			  int len, __wsum csum)
L
Linus Torvalds 已提交
1862
{
1863 1864
	int start = skb_headlen(skb);
	int i, copy = start - offset;
1865
	struct sk_buff *frag_iter;
L
Linus Torvalds 已提交
1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879
	int pos = 0;

	/* Checksum header. */
	if (copy > 0) {
		if (copy > len)
			copy = len;
		csum = csum_partial(skb->data + offset, copy, csum);
		if ((len -= copy) == 0)
			return csum;
		offset += copy;
		pos	= copy;
	}

	for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
1880
		int end;
E
Eric Dumazet 已提交
1881
		skb_frag_t *frag = &skb_shinfo(skb)->frags[i];
1882

1883
		WARN_ON(start > offset + len);
L
Linus Torvalds 已提交
1884

E
Eric Dumazet 已提交
1885
		end = start + skb_frag_size(frag);
L
Linus Torvalds 已提交
1886
		if ((copy = end - offset) > 0) {
1887
			__wsum csum2;
L
Linus Torvalds 已提交
1888 1889 1890 1891
			u8 *vaddr;

			if (copy > len)
				copy = len;
E
Eric Dumazet 已提交
1892
			vaddr = kmap_atomic(skb_frag_page(frag));
1893 1894
			csum2 = csum_partial(vaddr + frag->page_offset +
					     offset - start, copy, 0);
E
Eric Dumazet 已提交
1895
			kunmap_atomic(vaddr);
L
Linus Torvalds 已提交
1896 1897 1898 1899 1900 1901
			csum = csum_block_add(csum, csum2, pos);
			if (!(len -= copy))
				return csum;
			offset += copy;
			pos    += copy;
		}
1902
		start = end;
L
Linus Torvalds 已提交
1903 1904
	}

1905 1906
	skb_walk_frags(skb, frag_iter) {
		int end;
L
Linus Torvalds 已提交
1907

1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921
		WARN_ON(start > offset + len);

		end = start + frag_iter->len;
		if ((copy = end - offset) > 0) {
			__wsum csum2;
			if (copy > len)
				copy = len;
			csum2 = skb_checksum(frag_iter, offset - start,
					     copy, 0);
			csum = csum_block_add(csum, csum2, pos);
			if ((len -= copy) == 0)
				return csum;
			offset += copy;
			pos    += copy;
L
Linus Torvalds 已提交
1922
		}
1923
		start = end;
L
Linus Torvalds 已提交
1924
	}
1925
	BUG_ON(len);
L
Linus Torvalds 已提交
1926 1927 1928

	return csum;
}
1929
EXPORT_SYMBOL(skb_checksum);
L
Linus Torvalds 已提交
1930 1931 1932

/* Both of above in one bottle. */

1933 1934
__wsum skb_copy_and_csum_bits(const struct sk_buff *skb, int offset,
				    u8 *to, int len, __wsum csum)
L
Linus Torvalds 已提交
1935
{
1936 1937
	int start = skb_headlen(skb);
	int i, copy = start - offset;
1938
	struct sk_buff *frag_iter;
L
Linus Torvalds 已提交
1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954
	int pos = 0;

	/* Copy header. */
	if (copy > 0) {
		if (copy > len)
			copy = len;
		csum = csum_partial_copy_nocheck(skb->data + offset, to,
						 copy, csum);
		if ((len -= copy) == 0)
			return csum;
		offset += copy;
		to     += copy;
		pos	= copy;
	}

	for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
1955 1956
		int end;

1957
		WARN_ON(start > offset + len);
L
Linus Torvalds 已提交
1958

E
Eric Dumazet 已提交
1959
		end = start + skb_frag_size(&skb_shinfo(skb)->frags[i]);
L
Linus Torvalds 已提交
1960
		if ((copy = end - offset) > 0) {
1961
			__wsum csum2;
L
Linus Torvalds 已提交
1962 1963 1964 1965 1966
			u8 *vaddr;
			skb_frag_t *frag = &skb_shinfo(skb)->frags[i];

			if (copy > len)
				copy = len;
E
Eric Dumazet 已提交
1967
			vaddr = kmap_atomic(skb_frag_page(frag));
L
Linus Torvalds 已提交
1968
			csum2 = csum_partial_copy_nocheck(vaddr +
1969 1970 1971
							  frag->page_offset +
							  offset - start, to,
							  copy, 0);
E
Eric Dumazet 已提交
1972
			kunmap_atomic(vaddr);
L
Linus Torvalds 已提交
1973 1974 1975 1976 1977 1978 1979
			csum = csum_block_add(csum, csum2, pos);
			if (!(len -= copy))
				return csum;
			offset += copy;
			to     += copy;
			pos    += copy;
		}
1980
		start = end;
L
Linus Torvalds 已提交
1981 1982
	}

1983 1984 1985
	skb_walk_frags(skb, frag_iter) {
		__wsum csum2;
		int end;
L
Linus Torvalds 已提交
1986

1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001
		WARN_ON(start > offset + len);

		end = start + frag_iter->len;
		if ((copy = end - offset) > 0) {
			if (copy > len)
				copy = len;
			csum2 = skb_copy_and_csum_bits(frag_iter,
						       offset - start,
						       to, copy, 0);
			csum = csum_block_add(csum, csum2, pos);
			if ((len -= copy) == 0)
				return csum;
			offset += copy;
			to     += copy;
			pos    += copy;
L
Linus Torvalds 已提交
2002
		}
2003
		start = end;
L
Linus Torvalds 已提交
2004
	}
2005
	BUG_ON(len);
L
Linus Torvalds 已提交
2006 2007
	return csum;
}
2008
EXPORT_SYMBOL(skb_copy_and_csum_bits);
L
Linus Torvalds 已提交
2009 2010 2011

void skb_copy_and_csum_dev(const struct sk_buff *skb, u8 *to)
{
2012
	__wsum csum;
L
Linus Torvalds 已提交
2013 2014
	long csstart;

2015
	if (skb->ip_summed == CHECKSUM_PARTIAL)
2016
		csstart = skb_checksum_start_offset(skb);
L
Linus Torvalds 已提交
2017 2018 2019
	else
		csstart = skb_headlen(skb);

2020
	BUG_ON(csstart > skb_headlen(skb));
L
Linus Torvalds 已提交
2021

2022
	skb_copy_from_linear_data(skb, to, csstart);
L
Linus Torvalds 已提交
2023 2024 2025 2026 2027 2028

	csum = 0;
	if (csstart != skb->len)
		csum = skb_copy_and_csum_bits(skb, csstart, to + csstart,
					      skb->len - csstart, 0);

2029
	if (skb->ip_summed == CHECKSUM_PARTIAL) {
A
Al Viro 已提交
2030
		long csstuff = csstart + skb->csum_offset;
L
Linus Torvalds 已提交
2031

2032
		*((__sum16 *)(to + csstuff)) = csum_fold(csum);
L
Linus Torvalds 已提交
2033 2034
	}
}
2035
EXPORT_SYMBOL(skb_copy_and_csum_dev);
L
Linus Torvalds 已提交
2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055

/**
 *	skb_dequeue - remove from the head of the queue
 *	@list: list to dequeue from
 *
 *	Remove the head of the list. The list lock is taken so the function
 *	may be used safely with other locking list functions. The head item is
 *	returned or %NULL if the list is empty.
 */

struct sk_buff *skb_dequeue(struct sk_buff_head *list)
{
	unsigned long flags;
	struct sk_buff *result;

	spin_lock_irqsave(&list->lock, flags);
	result = __skb_dequeue(list);
	spin_unlock_irqrestore(&list->lock, flags);
	return result;
}
2056
EXPORT_SYMBOL(skb_dequeue);
L
Linus Torvalds 已提交
2057 2058 2059 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075

/**
 *	skb_dequeue_tail - remove from the tail of the queue
 *	@list: list to dequeue from
 *
 *	Remove the tail of the list. The list lock is taken so the function
 *	may be used safely with other locking list functions. The tail item is
 *	returned or %NULL if the list is empty.
 */
struct sk_buff *skb_dequeue_tail(struct sk_buff_head *list)
{
	unsigned long flags;
	struct sk_buff *result;

	spin_lock_irqsave(&list->lock, flags);
	result = __skb_dequeue_tail(list);
	spin_unlock_irqrestore(&list->lock, flags);
	return result;
}
2076
EXPORT_SYMBOL(skb_dequeue_tail);
L
Linus Torvalds 已提交
2077 2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091

/**
 *	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 takes the list
 *	lock and is atomic with respect to other list locking functions.
 */
void skb_queue_purge(struct sk_buff_head *list)
{
	struct sk_buff *skb;
	while ((skb = skb_dequeue(list)) != NULL)
		kfree_skb(skb);
}
2092
EXPORT_SYMBOL(skb_queue_purge);
L
Linus Torvalds 已提交
2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 2106 2107 2108 2109 2110 2111 2112

/**
 *	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 the list. This function takes the
 *	list lock and can be used safely with other locking &sk_buff functions
 *	safely.
 *
 *	A buffer cannot be placed on two lists at the same time.
 */
void skb_queue_head(struct sk_buff_head *list, struct sk_buff *newsk)
{
	unsigned long flags;

	spin_lock_irqsave(&list->lock, flags);
	__skb_queue_head(list, newsk);
	spin_unlock_irqrestore(&list->lock, flags);
}
2113
EXPORT_SYMBOL(skb_queue_head);
L
Linus Torvalds 已提交
2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131 2132 2133

/**
 *	skb_queue_tail - queue a buffer at the list tail
 *	@list: list to use
 *	@newsk: buffer to queue
 *
 *	Queue a buffer at the tail of the list. This function takes the
 *	list lock and can be used safely with other locking &sk_buff functions
 *	safely.
 *
 *	A buffer cannot be placed on two lists at the same time.
 */
void skb_queue_tail(struct sk_buff_head *list, struct sk_buff *newsk)
{
	unsigned long flags;

	spin_lock_irqsave(&list->lock, flags);
	__skb_queue_tail(list, newsk);
	spin_unlock_irqrestore(&list->lock, flags);
}
2134
EXPORT_SYMBOL(skb_queue_tail);
D
David S. Miller 已提交
2135

L
Linus Torvalds 已提交
2136 2137 2138
/**
 *	skb_unlink	-	remove a buffer from a list
 *	@skb: buffer to remove
D
David S. Miller 已提交
2139
 *	@list: list to use
L
Linus Torvalds 已提交
2140
 *
D
David S. Miller 已提交
2141 2142
 *	Remove a packet from a list. The list locks are taken and this
 *	function is atomic with respect to other list locked calls
L
Linus Torvalds 已提交
2143
 *
D
David S. Miller 已提交
2144
 *	You must know what list the SKB is on.
L
Linus Torvalds 已提交
2145
 */
D
David S. Miller 已提交
2146
void skb_unlink(struct sk_buff *skb, struct sk_buff_head *list)
L
Linus Torvalds 已提交
2147
{
D
David S. Miller 已提交
2148
	unsigned long flags;
L
Linus Torvalds 已提交
2149

D
David S. Miller 已提交
2150 2151 2152
	spin_lock_irqsave(&list->lock, flags);
	__skb_unlink(skb, list);
	spin_unlock_irqrestore(&list->lock, flags);
L
Linus Torvalds 已提交
2153
}
2154
EXPORT_SYMBOL(skb_unlink);
L
Linus Torvalds 已提交
2155 2156 2157 2158 2159

/**
 *	skb_append	-	append a buffer
 *	@old: buffer to insert after
 *	@newsk: buffer to insert
D
David S. Miller 已提交
2160
 *	@list: list to use
L
Linus Torvalds 已提交
2161 2162 2163 2164 2165
 *
 *	Place a packet after a given packet in a list. The list locks are taken
 *	and this function is atomic with respect to other list locked calls.
 *	A buffer cannot be placed on two lists at the same time.
 */
D
David S. Miller 已提交
2166
void skb_append(struct sk_buff *old, struct sk_buff *newsk, struct sk_buff_head *list)
L
Linus Torvalds 已提交
2167 2168 2169
{
	unsigned long flags;

D
David S. Miller 已提交
2170
	spin_lock_irqsave(&list->lock, flags);
2171
	__skb_queue_after(list, old, newsk);
D
David S. Miller 已提交
2172
	spin_unlock_irqrestore(&list->lock, flags);
L
Linus Torvalds 已提交
2173
}
2174
EXPORT_SYMBOL(skb_append);
L
Linus Torvalds 已提交
2175 2176 2177 2178 2179

/**
 *	skb_insert	-	insert a buffer
 *	@old: buffer to insert before
 *	@newsk: buffer to insert
D
David S. Miller 已提交
2180 2181 2182 2183 2184
 *	@list: list to use
 *
 *	Place a packet before a given packet in a list. The list locks are
 * 	taken and this function is atomic with respect to other list locked
 *	calls.
L
Linus Torvalds 已提交
2185 2186 2187
 *
 *	A buffer cannot be placed on two lists at the same time.
 */
D
David S. Miller 已提交
2188
void skb_insert(struct sk_buff *old, struct sk_buff *newsk, struct sk_buff_head *list)
L
Linus Torvalds 已提交
2189 2190 2191
{
	unsigned long flags;

D
David S. Miller 已提交
2192 2193 2194
	spin_lock_irqsave(&list->lock, flags);
	__skb_insert(newsk, old->prev, old, list);
	spin_unlock_irqrestore(&list->lock, flags);
L
Linus Torvalds 已提交
2195
}
2196
EXPORT_SYMBOL(skb_insert);
L
Linus Torvalds 已提交
2197 2198 2199 2200 2201 2202 2203

static inline void skb_split_inside_header(struct sk_buff *skb,
					   struct sk_buff* skb1,
					   const u32 len, const int pos)
{
	int i;

2204 2205
	skb_copy_from_linear_data_offset(skb, len, skb_put(skb1, pos - len),
					 pos - len);
L
Linus Torvalds 已提交
2206 2207 2208 2209 2210 2211 2212 2213 2214 2215
	/* And move data appendix as is. */
	for (i = 0; i < skb_shinfo(skb)->nr_frags; i++)
		skb_shinfo(skb1)->frags[i] = skb_shinfo(skb)->frags[i];

	skb_shinfo(skb1)->nr_frags = skb_shinfo(skb)->nr_frags;
	skb_shinfo(skb)->nr_frags  = 0;
	skb1->data_len		   = skb->data_len;
	skb1->len		   += skb1->data_len;
	skb->data_len		   = 0;
	skb->len		   = len;
2216
	skb_set_tail_pointer(skb, len);
L
Linus Torvalds 已提交
2217 2218 2219 2220 2221 2222 2223 2224 2225 2226 2227 2228 2229 2230 2231
}

static inline void skb_split_no_header(struct sk_buff *skb,
				       struct sk_buff* skb1,
				       const u32 len, int pos)
{
	int i, k = 0;
	const int nfrags = skb_shinfo(skb)->nr_frags;

	skb_shinfo(skb)->nr_frags = 0;
	skb1->len		  = skb1->data_len = skb->len - len;
	skb->len		  = len;
	skb->data_len		  = len - pos;

	for (i = 0; i < nfrags; i++) {
E
Eric Dumazet 已提交
2232
		int size = skb_frag_size(&skb_shinfo(skb)->frags[i]);
L
Linus Torvalds 已提交
2233 2234 2235 2236 2237 2238 2239 2240 2241 2242 2243 2244 2245

		if (pos + size > len) {
			skb_shinfo(skb1)->frags[k] = skb_shinfo(skb)->frags[i];

			if (pos < len) {
				/* Split frag.
				 * We have two variants in this case:
				 * 1. Move all the frag to the second
				 *    part, if it is possible. F.e.
				 *    this approach is mandatory for TUX,
				 *    where splitting is expensive.
				 * 2. Split is accurately. We make this.
				 */
2246
				skb_frag_ref(skb, i);
L
Linus Torvalds 已提交
2247
				skb_shinfo(skb1)->frags[0].page_offset += len - pos;
E
Eric Dumazet 已提交
2248 2249
				skb_frag_size_sub(&skb_shinfo(skb1)->frags[0], len - pos);
				skb_frag_size_set(&skb_shinfo(skb)->frags[i], len - pos);
L
Linus Torvalds 已提交
2250 2251 2252 2253 2254 2255 2256 2257 2258 2259 2260 2261 2262 2263 2264 2265 2266 2267 2268 2269 2270 2271 2272 2273 2274
				skb_shinfo(skb)->nr_frags++;
			}
			k++;
		} else
			skb_shinfo(skb)->nr_frags++;
		pos += size;
	}
	skb_shinfo(skb1)->nr_frags = k;
}

/**
 * skb_split - Split fragmented skb to two parts at length len.
 * @skb: the buffer to split
 * @skb1: the buffer to receive the second part
 * @len: new length for skb
 */
void skb_split(struct sk_buff *skb, struct sk_buff *skb1, const u32 len)
{
	int pos = skb_headlen(skb);

	if (len < pos)	/* Split line is inside header. */
		skb_split_inside_header(skb, skb1, len, pos);
	else		/* Second chunk has no header, nothing to copy. */
		skb_split_no_header(skb, skb1, len, pos);
}
2275
EXPORT_SYMBOL(skb_split);
L
Linus Torvalds 已提交
2276

2277 2278 2279 2280
/* Shifting from/to a cloned skb is a no-go.
 *
 * Caller cannot keep skb_shinfo related pointers past calling here!
 */
2281 2282
static int skb_prepare_for_shift(struct sk_buff *skb)
{
2283
	return skb_cloned(skb) && pskb_expand_head(skb, 0, 0, GFP_ATOMIC);
2284 2285 2286 2287 2288 2289 2290 2291 2292
}

/**
 * skb_shift - Shifts paged data partially from skb to another
 * @tgt: buffer into which tail data gets added
 * @skb: buffer from which the paged data comes from
 * @shiftlen: shift up to this many bytes
 *
 * Attempts to shift up to shiftlen worth of bytes, which may be less than
F
Feng King 已提交
2293
 * the length of the skb, from skb to tgt. Returns number bytes shifted.
2294 2295 2296 2297 2298 2299 2300 2301 2302 2303 2304 2305 2306 2307 2308 2309 2310 2311 2312 2313 2314 2315 2316 2317 2318 2319 2320
 * It's up to caller to free skb if everything was shifted.
 *
 * If @tgt runs out of frags, the whole operation is aborted.
 *
 * Skb cannot include anything else but paged data while tgt is allowed
 * to have non-paged data as well.
 *
 * TODO: full sized shift could be optimized but that would need
 * specialized skb free'er to handle frags without up-to-date nr_frags.
 */
int skb_shift(struct sk_buff *tgt, struct sk_buff *skb, int shiftlen)
{
	int from, to, merge, todo;
	struct skb_frag_struct *fragfrom, *fragto;

	BUG_ON(shiftlen > skb->len);
	BUG_ON(skb_headlen(skb));	/* Would corrupt stream */

	todo = shiftlen;
	from = 0;
	to = skb_shinfo(tgt)->nr_frags;
	fragfrom = &skb_shinfo(skb)->frags[from];

	/* Actual merge is delayed until the point when we know we can
	 * commit all, so that we don't have to undo partial changes
	 */
	if (!to ||
2321 2322
	    !skb_can_coalesce(tgt, to, skb_frag_page(fragfrom),
			      fragfrom->page_offset)) {
2323 2324 2325 2326
		merge = -1;
	} else {
		merge = to - 1;

E
Eric Dumazet 已提交
2327
		todo -= skb_frag_size(fragfrom);
2328 2329 2330 2331 2332
		if (todo < 0) {
			if (skb_prepare_for_shift(skb) ||
			    skb_prepare_for_shift(tgt))
				return 0;

2333 2334
			/* All previous frag pointers might be stale! */
			fragfrom = &skb_shinfo(skb)->frags[from];
2335 2336
			fragto = &skb_shinfo(tgt)->frags[merge];

E
Eric Dumazet 已提交
2337 2338
			skb_frag_size_add(fragto, shiftlen);
			skb_frag_size_sub(fragfrom, shiftlen);
2339 2340 2341 2342 2343 2344 2345 2346 2347 2348 2349 2350 2351 2352 2353 2354 2355 2356 2357 2358 2359 2360 2361
			fragfrom->page_offset += shiftlen;

			goto onlymerged;
		}

		from++;
	}

	/* Skip full, not-fitting skb to avoid expensive operations */
	if ((shiftlen == skb->len) &&
	    (skb_shinfo(skb)->nr_frags - from) > (MAX_SKB_FRAGS - to))
		return 0;

	if (skb_prepare_for_shift(skb) || skb_prepare_for_shift(tgt))
		return 0;

	while ((todo > 0) && (from < skb_shinfo(skb)->nr_frags)) {
		if (to == MAX_SKB_FRAGS)
			return 0;

		fragfrom = &skb_shinfo(skb)->frags[from];
		fragto = &skb_shinfo(tgt)->frags[to];

E
Eric Dumazet 已提交
2362
		if (todo >= skb_frag_size(fragfrom)) {
2363
			*fragto = *fragfrom;
E
Eric Dumazet 已提交
2364
			todo -= skb_frag_size(fragfrom);
2365 2366 2367 2368
			from++;
			to++;

		} else {
2369
			__skb_frag_ref(fragfrom);
2370 2371
			fragto->page = fragfrom->page;
			fragto->page_offset = fragfrom->page_offset;
E
Eric Dumazet 已提交
2372
			skb_frag_size_set(fragto, todo);
2373 2374

			fragfrom->page_offset += todo;
E
Eric Dumazet 已提交
2375
			skb_frag_size_sub(fragfrom, todo);
2376 2377 2378 2379 2380 2381 2382 2383 2384 2385 2386 2387 2388 2389
			todo = 0;

			to++;
			break;
		}
	}

	/* Ready to "commit" this state change to tgt */
	skb_shinfo(tgt)->nr_frags = to;

	if (merge >= 0) {
		fragfrom = &skb_shinfo(skb)->frags[0];
		fragto = &skb_shinfo(tgt)->frags[merge];

E
Eric Dumazet 已提交
2390
		skb_frag_size_add(fragto, skb_frag_size(fragfrom));
2391
		__skb_frag_unref(fragfrom);
2392 2393 2394 2395 2396 2397 2398 2399 2400 2401 2402 2403 2404 2405 2406 2407 2408 2409 2410 2411 2412 2413 2414 2415 2416 2417 2418 2419
	}

	/* Reposition in the original skb */
	to = 0;
	while (from < skb_shinfo(skb)->nr_frags)
		skb_shinfo(skb)->frags[to++] = skb_shinfo(skb)->frags[from++];
	skb_shinfo(skb)->nr_frags = to;

	BUG_ON(todo > 0 && !skb_shinfo(skb)->nr_frags);

onlymerged:
	/* Most likely the tgt won't ever need its checksum anymore, skb on
	 * the other hand might need it if it needs to be resent
	 */
	tgt->ip_summed = CHECKSUM_PARTIAL;
	skb->ip_summed = CHECKSUM_PARTIAL;

	/* Yak, is it really working this way? Some helper please? */
	skb->len -= shiftlen;
	skb->data_len -= shiftlen;
	skb->truesize -= shiftlen;
	tgt->len += shiftlen;
	tgt->data_len += shiftlen;
	tgt->truesize += shiftlen;

	return shiftlen;
}

2420 2421 2422 2423 2424 2425 2426 2427 2428 2429 2430 2431 2432 2433 2434 2435 2436 2437 2438
/**
 * skb_prepare_seq_read - Prepare a sequential read of skb data
 * @skb: the buffer to read
 * @from: lower offset of data to be read
 * @to: upper offset of data to be read
 * @st: state variable
 *
 * Initializes the specified state variable. Must be called before
 * invoking skb_seq_read() for the first time.
 */
void skb_prepare_seq_read(struct sk_buff *skb, unsigned int from,
			  unsigned int to, struct skb_seq_state *st)
{
	st->lower_offset = from;
	st->upper_offset = to;
	st->root_skb = st->cur_skb = skb;
	st->frag_idx = st->stepped_offset = 0;
	st->frag_data = NULL;
}
2439
EXPORT_SYMBOL(skb_prepare_seq_read);
2440 2441 2442 2443 2444 2445 2446 2447 2448 2449 2450 2451 2452 2453 2454 2455 2456 2457

/**
 * skb_seq_read - Sequentially read skb data
 * @consumed: number of bytes consumed by the caller so far
 * @data: destination pointer for data to be returned
 * @st: state variable
 *
 * Reads a block of skb data at &consumed relative to the
 * lower offset specified to skb_prepare_seq_read(). Assigns
 * the head of the data block to &data and returns the length
 * of the block or 0 if the end of the skb data or the upper
 * offset has been reached.
 *
 * The caller is not required to consume all of the data
 * returned, i.e. &consumed is typically set to the number
 * of bytes already consumed and the next call to
 * skb_seq_read() will return the remaining part of the block.
 *
L
Lucas De Marchi 已提交
2458
 * Note 1: The size of each block of data returned can be arbitrary,
2459 2460 2461
 *       this limitation is the cost for zerocopy seqeuental
 *       reads of potentially non linear data.
 *
2462
 * Note 2: Fragment lists within fragments are not implemented
2463 2464 2465 2466 2467 2468 2469 2470 2471 2472 2473 2474 2475
 *       at the moment, state->root_skb could be replaced with
 *       a stack for this purpose.
 */
unsigned int skb_seq_read(unsigned int consumed, const u8 **data,
			  struct skb_seq_state *st)
{
	unsigned int block_limit, abs_offset = consumed + st->lower_offset;
	skb_frag_t *frag;

	if (unlikely(abs_offset >= st->upper_offset))
		return 0;

next_skb:
2476
	block_limit = skb_headlen(st->cur_skb) + st->stepped_offset;
2477

2478
	if (abs_offset < block_limit && !st->frag_data) {
2479
		*data = st->cur_skb->data + (abs_offset - st->stepped_offset);
2480 2481 2482 2483 2484 2485 2486 2487
		return block_limit - abs_offset;
	}

	if (st->frag_idx == 0 && !st->frag_data)
		st->stepped_offset += skb_headlen(st->cur_skb);

	while (st->frag_idx < skb_shinfo(st->cur_skb)->nr_frags) {
		frag = &skb_shinfo(st->cur_skb)->frags[st->frag_idx];
E
Eric Dumazet 已提交
2488
		block_limit = skb_frag_size(frag) + st->stepped_offset;
2489 2490 2491

		if (abs_offset < block_limit) {
			if (!st->frag_data)
E
Eric Dumazet 已提交
2492
				st->frag_data = kmap_atomic(skb_frag_page(frag));
2493 2494 2495 2496 2497 2498 2499 2500

			*data = (u8 *) st->frag_data + frag->page_offset +
				(abs_offset - st->stepped_offset);

			return block_limit - abs_offset;
		}

		if (st->frag_data) {
E
Eric Dumazet 已提交
2501
			kunmap_atomic(st->frag_data);
2502 2503 2504 2505
			st->frag_data = NULL;
		}

		st->frag_idx++;
E
Eric Dumazet 已提交
2506
		st->stepped_offset += skb_frag_size(frag);
2507 2508
	}

2509
	if (st->frag_data) {
E
Eric Dumazet 已提交
2510
		kunmap_atomic(st->frag_data);
2511 2512 2513
		st->frag_data = NULL;
	}

2514
	if (st->root_skb == st->cur_skb && skb_has_frag_list(st->root_skb)) {
S
Shyam Iyer 已提交
2515
		st->cur_skb = skb_shinfo(st->root_skb)->frag_list;
2516 2517
		st->frag_idx = 0;
		goto next_skb;
S
Shyam Iyer 已提交
2518 2519
	} else if (st->cur_skb->next) {
		st->cur_skb = st->cur_skb->next;
2520
		st->frag_idx = 0;
2521 2522 2523 2524 2525
		goto next_skb;
	}

	return 0;
}
2526
EXPORT_SYMBOL(skb_seq_read);
2527 2528 2529 2530 2531 2532 2533 2534 2535 2536 2537

/**
 * skb_abort_seq_read - Abort a sequential read of skb data
 * @st: state variable
 *
 * Must be called if skb_seq_read() was not called until it
 * returned 0.
 */
void skb_abort_seq_read(struct skb_seq_state *st)
{
	if (st->frag_data)
E
Eric Dumazet 已提交
2538
		kunmap_atomic(st->frag_data);
2539
}
2540
EXPORT_SYMBOL(skb_abort_seq_read);
2541

2542 2543 2544 2545 2546 2547 2548 2549 2550 2551 2552 2553 2554 2555 2556 2557 2558 2559 2560 2561 2562 2563 2564 2565 2566 2567 2568 2569 2570 2571 2572
#define TS_SKB_CB(state)	((struct skb_seq_state *) &((state)->cb))

static unsigned int skb_ts_get_next_block(unsigned int offset, const u8 **text,
					  struct ts_config *conf,
					  struct ts_state *state)
{
	return skb_seq_read(offset, text, TS_SKB_CB(state));
}

static void skb_ts_finish(struct ts_config *conf, struct ts_state *state)
{
	skb_abort_seq_read(TS_SKB_CB(state));
}

/**
 * skb_find_text - Find a text pattern in skb data
 * @skb: the buffer to look in
 * @from: search offset
 * @to: search limit
 * @config: textsearch configuration
 * @state: uninitialized textsearch state variable
 *
 * Finds a pattern in the skb data according to the specified
 * textsearch configuration. Use textsearch_next() to retrieve
 * subsequent occurrences of the pattern. Returns the offset
 * to the first occurrence or UINT_MAX if no match was found.
 */
unsigned int skb_find_text(struct sk_buff *skb, unsigned int from,
			   unsigned int to, struct ts_config *config,
			   struct ts_state *state)
{
2573 2574
	unsigned int ret;

2575 2576 2577 2578 2579
	config->get_next_block = skb_ts_get_next_block;
	config->finish = skb_ts_finish;

	skb_prepare_seq_read(skb, from, to, TS_SKB_CB(state));

2580 2581
	ret = textsearch_find(config, state);
	return (ret <= to - from ? ret : UINT_MAX);
2582
}
2583
EXPORT_SYMBOL(skb_find_text);
2584

2585 2586 2587 2588 2589 2590 2591 2592 2593 2594 2595 2596
/**
 * skb_append_datato_frags: - append the user data to a skb
 * @sk: sock  structure
 * @skb: skb structure to be appened with user data.
 * @getfrag: call back function to be used for getting the user data
 * @from: pointer to user message iov
 * @length: length of the iov message
 *
 * Description: This procedure append the user data in the fragment part
 * of the skb if any page alloc fails user this procedure returns  -ENOMEM
 */
int skb_append_datato_frags(struct sock *sk, struct sk_buff *skb,
2597
			int (*getfrag)(void *from, char *to, int offset,
2598 2599 2600 2601 2602 2603 2604 2605 2606 2607 2608 2609 2610 2611 2612 2613 2614 2615 2616 2617 2618 2619 2620 2621 2622 2623 2624 2625 2626 2627 2628 2629 2630 2631 2632 2633 2634 2635
					int len, int odd, struct sk_buff *skb),
			void *from, int length)
{
	int frg_cnt = 0;
	skb_frag_t *frag = NULL;
	struct page *page = NULL;
	int copy, left;
	int offset = 0;
	int ret;

	do {
		/* Return error if we don't have space for new frag */
		frg_cnt = skb_shinfo(skb)->nr_frags;
		if (frg_cnt >= MAX_SKB_FRAGS)
			return -EFAULT;

		/* allocate a new page for next frag */
		page = alloc_pages(sk->sk_allocation, 0);

		/* If alloc_page fails just return failure and caller will
		 * free previous allocated pages by doing kfree_skb()
		 */
		if (page == NULL)
			return -ENOMEM;

		/* initialize the next frag */
		skb_fill_page_desc(skb, frg_cnt, page, 0, 0);
		skb->truesize += PAGE_SIZE;
		atomic_add(PAGE_SIZE, &sk->sk_wmem_alloc);

		/* get the new initialized frag */
		frg_cnt = skb_shinfo(skb)->nr_frags;
		frag = &skb_shinfo(skb)->frags[frg_cnt - 1];

		/* copy the user data to page */
		left = PAGE_SIZE - frag->page_offset;
		copy = (length > left)? left : length;

E
Eric Dumazet 已提交
2636
		ret = getfrag(from, skb_frag_address(frag) + skb_frag_size(frag),
2637 2638 2639 2640 2641
			    offset, copy, 0, skb);
		if (ret < 0)
			return -EFAULT;

		/* copy was successful so update the size parameters */
E
Eric Dumazet 已提交
2642
		skb_frag_size_add(frag, copy);
2643 2644 2645 2646 2647 2648 2649 2650 2651
		skb->len += copy;
		skb->data_len += copy;
		offset += copy;
		length -= copy;

	} while (length > 0);

	return 0;
}
2652
EXPORT_SYMBOL(skb_append_datato_frags);
2653

2654 2655 2656 2657 2658 2659
/**
 *	skb_pull_rcsum - pull skb and update receive checksum
 *	@skb: buffer to update
 *	@len: length of data pulled
 *
 *	This function performs an skb_pull on the packet and updates
2660
 *	the CHECKSUM_COMPLETE checksum.  It should be used on
2661 2662 2663
 *	receive path processing instead of skb_pull unless you know
 *	that the checksum difference is zero (e.g., a valid IP header)
 *	or you are setting ip_summed to CHECKSUM_NONE.
2664 2665 2666 2667 2668 2669 2670 2671 2672
 */
unsigned char *skb_pull_rcsum(struct sk_buff *skb, unsigned int len)
{
	BUG_ON(len > skb->len);
	skb->len -= len;
	BUG_ON(skb->len < skb->data_len);
	skb_postpull_rcsum(skb, skb->data, len);
	return skb->data += len;
}
2673 2674
EXPORT_SYMBOL_GPL(skb_pull_rcsum);

H
Herbert Xu 已提交
2675 2676 2677
/**
 *	skb_segment - Perform protocol segmentation on skb.
 *	@skb: buffer to segment
2678
 *	@features: features for the output path (see dev->features)
H
Herbert Xu 已提交
2679 2680
 *
 *	This function performs segmentation on the given skb.  It returns
2681 2682
 *	a pointer to the first in a list of new skbs for the segments.
 *	In case of error it returns ERR_PTR(err).
H
Herbert Xu 已提交
2683
 */
2684
struct sk_buff *skb_segment(struct sk_buff *skb, netdev_features_t features)
H
Herbert Xu 已提交
2685 2686 2687
{
	struct sk_buff *segs = NULL;
	struct sk_buff *tail = NULL;
2688
	struct sk_buff *fskb = skb_shinfo(skb)->frag_list;
H
Herbert Xu 已提交
2689
	unsigned int mss = skb_shinfo(skb)->gso_size;
2690
	unsigned int doffset = skb->data - skb_mac_header(skb);
H
Herbert Xu 已提交
2691 2692 2693
	unsigned int offset = doffset;
	unsigned int headroom;
	unsigned int len;
2694
	int sg = !!(features & NETIF_F_SG);
H
Herbert Xu 已提交
2695 2696 2697 2698 2699 2700 2701 2702 2703 2704 2705 2706
	int nfrags = skb_shinfo(skb)->nr_frags;
	int err = -ENOMEM;
	int i = 0;
	int pos;

	__skb_push(skb, doffset);
	headroom = skb_headroom(skb);
	pos = skb_headlen(skb);

	do {
		struct sk_buff *nskb;
		skb_frag_t *frag;
2707
		int hsize;
H
Herbert Xu 已提交
2708 2709 2710 2711 2712 2713 2714 2715 2716
		int size;

		len = skb->len - offset;
		if (len > mss)
			len = mss;

		hsize = skb_headlen(skb) - offset;
		if (hsize < 0)
			hsize = 0;
2717 2718
		if (hsize > len || !sg)
			hsize = len;
H
Herbert Xu 已提交
2719

2720 2721 2722 2723 2724 2725 2726 2727 2728 2729
		if (!hsize && i >= nfrags) {
			BUG_ON(fskb->len != len);

			pos += len;
			nskb = skb_clone(fskb, GFP_ATOMIC);
			fskb = fskb->next;

			if (unlikely(!nskb))
				goto err;

2730
			hsize = skb_end_offset(nskb);
2731 2732 2733 2734 2735
			if (skb_cow_head(nskb, doffset + headroom)) {
				kfree_skb(nskb);
				goto err;
			}

2736
			nskb->truesize += skb_end_offset(nskb) - hsize;
2737 2738 2739 2740 2741 2742 2743 2744 2745 2746 2747 2748
			skb_release_head_state(nskb);
			__skb_push(nskb, doffset);
		} else {
			nskb = alloc_skb(hsize + doffset + headroom,
					 GFP_ATOMIC);

			if (unlikely(!nskb))
				goto err;

			skb_reserve(nskb, headroom);
			__skb_put(nskb, doffset);
		}
H
Herbert Xu 已提交
2749 2750 2751 2752 2753 2754 2755

		if (segs)
			tail->next = nskb;
		else
			segs = nskb;
		tail = nskb;

2756
		__copy_skb_header(nskb, skb);
H
Herbert Xu 已提交
2757 2758
		nskb->mac_len = skb->mac_len;

E
Eric Dumazet 已提交
2759 2760 2761 2762
		/* nskb and skb might have different headroom */
		if (nskb->ip_summed == CHECKSUM_PARTIAL)
			nskb->csum_start += skb_headroom(nskb) - headroom;

2763
		skb_reset_mac_header(nskb);
2764
		skb_set_network_header(nskb, skb->mac_len);
2765 2766
		nskb->transport_header = (nskb->network_header +
					  skb_network_header_len(skb));
2767 2768
		skb_copy_from_linear_data(skb, nskb->data, doffset);

H
Herbert Xu 已提交
2769
		if (fskb != skb_shinfo(skb)->frag_list)
2770 2771
			continue;

H
Herbert Xu 已提交
2772
		if (!sg) {
2773
			nskb->ip_summed = CHECKSUM_NONE;
H
Herbert Xu 已提交
2774 2775 2776 2777 2778 2779 2780 2781
			nskb->csum = skb_copy_and_csum_bits(skb, offset,
							    skb_put(nskb, len),
							    len, 0);
			continue;
		}

		frag = skb_shinfo(nskb)->frags;

2782 2783
		skb_copy_from_linear_data_offset(skb, offset,
						 skb_put(nskb, hsize), hsize);
H
Herbert Xu 已提交
2784

2785
		while (pos < offset + len && i < nfrags) {
H
Herbert Xu 已提交
2786
			*frag = skb_shinfo(skb)->frags[i];
2787
			__skb_frag_ref(frag);
E
Eric Dumazet 已提交
2788
			size = skb_frag_size(frag);
H
Herbert Xu 已提交
2789 2790 2791

			if (pos < offset) {
				frag->page_offset += offset - pos;
E
Eric Dumazet 已提交
2792
				skb_frag_size_sub(frag, offset - pos);
H
Herbert Xu 已提交
2793 2794
			}

2795
			skb_shinfo(nskb)->nr_frags++;
H
Herbert Xu 已提交
2796 2797 2798 2799 2800

			if (pos + size <= offset + len) {
				i++;
				pos += size;
			} else {
E
Eric Dumazet 已提交
2801
				skb_frag_size_sub(frag, pos + size - (offset + len));
2802
				goto skip_fraglist;
H
Herbert Xu 已提交
2803 2804 2805 2806 2807
			}

			frag++;
		}

2808 2809 2810 2811 2812 2813 2814 2815 2816 2817 2818 2819 2820 2821 2822
		if (pos < offset + len) {
			struct sk_buff *fskb2 = fskb;

			BUG_ON(pos + fskb->len != offset + len);

			pos += fskb->len;
			fskb = fskb->next;

			if (fskb2->next) {
				fskb2 = skb_clone(fskb2, GFP_ATOMIC);
				if (!fskb2)
					goto err;
			} else
				skb_get(fskb2);

2823
			SKB_FRAG_ASSERT(nskb);
2824 2825 2826 2827
			skb_shinfo(nskb)->frag_list = fskb2;
		}

skip_fraglist:
H
Herbert Xu 已提交
2828 2829 2830 2831 2832 2833 2834 2835 2836 2837
		nskb->data_len = len - hsize;
		nskb->len += nskb->data_len;
		nskb->truesize += nskb->data_len;
	} while ((offset += len) < skb->len);

	return segs;

err:
	while ((skb = segs)) {
		segs = skb->next;
P
Patrick McHardy 已提交
2838
		kfree_skb(skb);
H
Herbert Xu 已提交
2839 2840 2841 2842 2843
	}
	return ERR_PTR(err);
}
EXPORT_SYMBOL_GPL(skb_segment);

H
Herbert Xu 已提交
2844 2845 2846 2847
int skb_gro_receive(struct sk_buff **head, struct sk_buff *skb)
{
	struct sk_buff *p = *head;
	struct sk_buff *nskb;
2848 2849
	struct skb_shared_info *skbinfo = skb_shinfo(skb);
	struct skb_shared_info *pinfo = skb_shinfo(p);
H
Herbert Xu 已提交
2850
	unsigned int headroom;
2851
	unsigned int len = skb_gro_len(skb);
2852 2853
	unsigned int offset = skb_gro_offset(skb);
	unsigned int headlen = skb_headlen(skb);
2854
	unsigned int delta_truesize;
H
Herbert Xu 已提交
2855

2856
	if (p->len + len >= 65536)
H
Herbert Xu 已提交
2857 2858
		return -E2BIG;

2859
	if (pinfo->frag_list)
H
Herbert Xu 已提交
2860
		goto merge;
2861
	else if (headlen <= offset) {
2862
		skb_frag_t *frag;
2863
		skb_frag_t *frag2;
2864 2865
		int i = skbinfo->nr_frags;
		int nr_frags = pinfo->nr_frags + i;
2866 2867

		offset -= headlen;
2868

2869
		if (nr_frags > MAX_SKB_FRAGS)
2870 2871
			return -E2BIG;

2872 2873
		pinfo->nr_frags = nr_frags;
		skbinfo->nr_frags = 0;
2874

2875 2876
		frag = pinfo->frags + nr_frags;
		frag2 = skbinfo->frags + i;
2877 2878 2879
		do {
			*--frag = *--frag2;
		} while (--i);
H
Herbert Xu 已提交
2880

2881
		frag->page_offset += offset;
E
Eric Dumazet 已提交
2882
		skb_frag_size_sub(frag, offset);
2883

2884
		/* all fragments truesize : remove (head size + sk_buff) */
2885 2886
		delta_truesize = skb->truesize -
				 SKB_TRUESIZE(skb_end_offset(skb));
2887

2888 2889 2890 2891
		skb->truesize -= skb->data_len;
		skb->len -= skb->data_len;
		skb->data_len = 0;

2892
		NAPI_GRO_CB(skb)->free = NAPI_GRO_FREE;
H
Herbert Xu 已提交
2893
		goto done;
2894 2895 2896 2897 2898 2899 2900 2901 2902 2903 2904 2905 2906 2907 2908 2909 2910 2911 2912 2913 2914 2915 2916
	} else if (skb->head_frag) {
		int nr_frags = pinfo->nr_frags;
		skb_frag_t *frag = pinfo->frags + nr_frags;
		struct page *page = virt_to_head_page(skb->head);
		unsigned int first_size = headlen - offset;
		unsigned int first_offset;

		if (nr_frags + 1 + skbinfo->nr_frags > MAX_SKB_FRAGS)
			return -E2BIG;

		first_offset = skb->data -
			       (unsigned char *)page_address(page) +
			       offset;

		pinfo->nr_frags = nr_frags + 1 + skbinfo->nr_frags;

		frag->page.p	  = page;
		frag->page_offset = first_offset;
		skb_frag_size_set(frag, first_size);

		memcpy(frag + 1, skbinfo->frags, sizeof(*frag) * skbinfo->nr_frags);
		/* We dont need to clear skbinfo->nr_frags here */

2917
		delta_truesize = skb->truesize - SKB_DATA_ALIGN(sizeof(struct sk_buff));
2918 2919
		NAPI_GRO_CB(skb)->free = NAPI_GRO_FREE_STOLEN_HEAD;
		goto done;
2920 2921
	} else if (skb_gro_len(p) != pinfo->gso_size)
		return -E2BIG;
H
Herbert Xu 已提交
2922 2923

	headroom = skb_headroom(p);
E
Eric Dumazet 已提交
2924
	nskb = alloc_skb(headroom + skb_gro_offset(p), GFP_ATOMIC);
H
Herbert Xu 已提交
2925 2926 2927 2928 2929 2930 2931
	if (unlikely(!nskb))
		return -ENOMEM;

	__copy_skb_header(nskb, p);
	nskb->mac_len = p->mac_len;

	skb_reserve(nskb, headroom);
2932
	__skb_put(nskb, skb_gro_offset(p));
H
Herbert Xu 已提交
2933

2934
	skb_set_mac_header(nskb, skb_mac_header(p) - p->data);
H
Herbert Xu 已提交
2935 2936 2937
	skb_set_network_header(nskb, skb_network_offset(p));
	skb_set_transport_header(nskb, skb_transport_offset(p));

2938 2939 2940
	__skb_pull(p, skb_gro_offset(p));
	memcpy(skb_mac_header(nskb), skb_mac_header(p),
	       p->data - skb_mac_header(p));
H
Herbert Xu 已提交
2941 2942 2943

	*NAPI_GRO_CB(nskb) = *NAPI_GRO_CB(p);
	skb_shinfo(nskb)->frag_list = p;
2944
	skb_shinfo(nskb)->gso_size = pinfo->gso_size;
2945
	pinfo->gso_size = 0;
H
Herbert Xu 已提交
2946 2947 2948 2949
	skb_header_release(p);
	nskb->prev = p;

	nskb->data_len += p->len;
E
Eric Dumazet 已提交
2950
	nskb->truesize += p->truesize;
H
Herbert Xu 已提交
2951 2952 2953 2954 2955 2956 2957 2958 2959
	nskb->len += p->len;

	*head = nskb;
	nskb->next = p->next;
	p->next = NULL;

	p = nskb;

merge:
2960
	delta_truesize = skb->truesize;
2961
	if (offset > headlen) {
2962 2963 2964
		unsigned int eat = offset - headlen;

		skbinfo->frags[0].page_offset += eat;
E
Eric Dumazet 已提交
2965
		skb_frag_size_sub(&skbinfo->frags[0], eat);
2966 2967
		skb->data_len -= eat;
		skb->len -= eat;
2968
		offset = headlen;
2969 2970
	}

2971
	__skb_pull(skb, offset);
2972

H
Herbert Xu 已提交
2973 2974 2975 2976
	p->prev->next = skb;
	p->prev = skb;
	skb_header_release(skb);

H
Herbert Xu 已提交
2977 2978
done:
	NAPI_GRO_CB(p)->count++;
H
Herbert Xu 已提交
2979
	p->data_len += len;
2980
	p->truesize += delta_truesize;
H
Herbert Xu 已提交
2981
	p->len += len;
H
Herbert Xu 已提交
2982 2983 2984 2985 2986 2987

	NAPI_GRO_CB(skb)->same_flow = 1;
	return 0;
}
EXPORT_SYMBOL_GPL(skb_gro_receive);

L
Linus Torvalds 已提交
2988 2989 2990 2991 2992
void __init skb_init(void)
{
	skbuff_head_cache = kmem_cache_create("skbuff_head_cache",
					      sizeof(struct sk_buff),
					      0,
A
Alexey Dobriyan 已提交
2993
					      SLAB_HWCACHE_ALIGN|SLAB_PANIC,
2994
					      NULL);
2995 2996 2997 2998
	skbuff_fclone_cache = kmem_cache_create("skbuff_fclone_cache",
						(2*sizeof(struct sk_buff)) +
						sizeof(atomic_t),
						0,
A
Alexey Dobriyan 已提交
2999
						SLAB_HWCACHE_ALIGN|SLAB_PANIC,
3000
						NULL);
L
Linus Torvalds 已提交
3001 3002
}

3003 3004 3005 3006 3007 3008 3009 3010 3011 3012
/**
 *	skb_to_sgvec - Fill a scatter-gather list from a socket buffer
 *	@skb: Socket buffer containing the buffers to be mapped
 *	@sg: The scatter-gather list to map into
 *	@offset: The offset into the buffer's contents to start mapping
 *	@len: Length of buffer space to be mapped
 *
 *	Fill the specified scatter-gather list with mappings/pointers into a
 *	region of the buffer space attached to a socket buffer.
 */
3013 3014
static int
__skb_to_sgvec(struct sk_buff *skb, struct scatterlist *sg, int offset, int len)
3015
{
3016 3017
	int start = skb_headlen(skb);
	int i, copy = start - offset;
3018
	struct sk_buff *frag_iter;
3019 3020 3021 3022 3023
	int elt = 0;

	if (copy > 0) {
		if (copy > len)
			copy = len;
3024
		sg_set_buf(sg, skb->data + offset, copy);
3025 3026 3027 3028 3029 3030 3031
		elt++;
		if ((len -= copy) == 0)
			return elt;
		offset += copy;
	}

	for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
3032
		int end;
3033

3034
		WARN_ON(start > offset + len);
3035

E
Eric Dumazet 已提交
3036
		end = start + skb_frag_size(&skb_shinfo(skb)->frags[i]);
3037 3038 3039 3040 3041
		if ((copy = end - offset) > 0) {
			skb_frag_t *frag = &skb_shinfo(skb)->frags[i];

			if (copy > len)
				copy = len;
3042
			sg_set_page(&sg[elt], skb_frag_page(frag), copy,
3043
					frag->page_offset+offset-start);
3044 3045 3046 3047 3048
			elt++;
			if (!(len -= copy))
				return elt;
			offset += copy;
		}
3049
		start = end;
3050 3051
	}

3052 3053
	skb_walk_frags(skb, frag_iter) {
		int end;
3054

3055
		WARN_ON(start > offset + len);
3056

3057 3058 3059 3060 3061 3062 3063 3064 3065
		end = start + frag_iter->len;
		if ((copy = end - offset) > 0) {
			if (copy > len)
				copy = len;
			elt += __skb_to_sgvec(frag_iter, sg+elt, offset - start,
					      copy);
			if ((len -= copy) == 0)
				return elt;
			offset += copy;
3066
		}
3067
		start = end;
3068 3069 3070 3071 3072
	}
	BUG_ON(len);
	return elt;
}

3073 3074 3075 3076
int skb_to_sgvec(struct sk_buff *skb, struct scatterlist *sg, int offset, int len)
{
	int nsg = __skb_to_sgvec(skb, sg, offset, len);

J
Jens Axboe 已提交
3077
	sg_mark_end(&sg[nsg - 1]);
3078 3079 3080

	return nsg;
}
3081
EXPORT_SYMBOL_GPL(skb_to_sgvec);
3082

3083 3084 3085 3086 3087 3088 3089 3090 3091 3092 3093 3094 3095 3096 3097 3098 3099 3100 3101 3102 3103 3104 3105 3106 3107 3108 3109 3110 3111 3112 3113 3114
/**
 *	skb_cow_data - Check that a socket buffer's data buffers are writable
 *	@skb: The socket buffer to check.
 *	@tailbits: Amount of trailing space to be added
 *	@trailer: Returned pointer to the skb where the @tailbits space begins
 *
 *	Make sure that the data buffers attached to a socket buffer are
 *	writable. If they are not, private copies are made of the data buffers
 *	and the socket buffer is set to use these instead.
 *
 *	If @tailbits is given, make sure that there is space to write @tailbits
 *	bytes of data beyond current end of socket buffer.  @trailer will be
 *	set to point to the skb in which this space begins.
 *
 *	The number of scatterlist elements required to completely map the
 *	COW'd and extended socket buffer will be returned.
 */
int skb_cow_data(struct sk_buff *skb, int tailbits, struct sk_buff **trailer)
{
	int copyflag;
	int elt;
	struct sk_buff *skb1, **skb_p;

	/* If skb is cloned or its head is paged, reallocate
	 * head pulling out all the pages (pages are considered not writable
	 * at the moment even if they are anonymous).
	 */
	if ((skb_cloned(skb) || skb_shinfo(skb)->nr_frags) &&
	    __pskb_pull_tail(skb, skb_pagelen(skb)-skb_headlen(skb)) == NULL)
		return -ENOMEM;

	/* Easy case. Most of packets will go this way. */
3115
	if (!skb_has_frag_list(skb)) {
3116 3117 3118 3119 3120 3121 3122 3123 3124 3125 3126 3127 3128 3129 3130 3131 3132 3133 3134 3135 3136 3137 3138 3139 3140 3141 3142 3143 3144 3145 3146 3147 3148 3149
		/* A little of trouble, not enough of space for trailer.
		 * This should not happen, when stack is tuned to generate
		 * good frames. OK, on miss we reallocate and reserve even more
		 * space, 128 bytes is fair. */

		if (skb_tailroom(skb) < tailbits &&
		    pskb_expand_head(skb, 0, tailbits-skb_tailroom(skb)+128, GFP_ATOMIC))
			return -ENOMEM;

		/* Voila! */
		*trailer = skb;
		return 1;
	}

	/* Misery. We are in troubles, going to mincer fragments... */

	elt = 1;
	skb_p = &skb_shinfo(skb)->frag_list;
	copyflag = 0;

	while ((skb1 = *skb_p) != NULL) {
		int ntail = 0;

		/* The fragment is partially pulled by someone,
		 * this can happen on input. Copy it and everything
		 * after it. */

		if (skb_shared(skb1))
			copyflag = 1;

		/* If the skb is the last, worry about trailer. */

		if (skb1->next == NULL && tailbits) {
			if (skb_shinfo(skb1)->nr_frags ||
3150
			    skb_has_frag_list(skb1) ||
3151 3152 3153 3154 3155 3156 3157 3158
			    skb_tailroom(skb1) < tailbits)
				ntail = tailbits + 128;
		}

		if (copyflag ||
		    skb_cloned(skb1) ||
		    ntail ||
		    skb_shinfo(skb1)->nr_frags ||
3159
		    skb_has_frag_list(skb1)) {
3160 3161 3162 3163 3164 3165 3166 3167 3168 3169 3170 3171 3172 3173 3174 3175 3176 3177 3178 3179 3180 3181 3182 3183 3184 3185 3186 3187 3188 3189 3190
			struct sk_buff *skb2;

			/* Fuck, we are miserable poor guys... */
			if (ntail == 0)
				skb2 = skb_copy(skb1, GFP_ATOMIC);
			else
				skb2 = skb_copy_expand(skb1,
						       skb_headroom(skb1),
						       ntail,
						       GFP_ATOMIC);
			if (unlikely(skb2 == NULL))
				return -ENOMEM;

			if (skb1->sk)
				skb_set_owner_w(skb2, skb1->sk);

			/* Looking around. Are we still alive?
			 * OK, link new skb, drop old one */

			skb2->next = skb1->next;
			*skb_p = skb2;
			kfree_skb(skb1);
			skb1 = skb2;
		}
		elt++;
		*trailer = skb1;
		skb_p = &skb1->next;
	}

	return elt;
}
3191
EXPORT_SYMBOL_GPL(skb_cow_data);
3192

3193 3194 3195 3196 3197 3198 3199 3200 3201 3202 3203 3204
static void sock_rmem_free(struct sk_buff *skb)
{
	struct sock *sk = skb->sk;

	atomic_sub(skb->truesize, &sk->sk_rmem_alloc);
}

/*
 * Note: We dont mem charge error packets (no sk_forward_alloc changes)
 */
int sock_queue_err_skb(struct sock *sk, struct sk_buff *skb)
{
3205 3206
	int len = skb->len;

3207
	if (atomic_read(&sk->sk_rmem_alloc) + skb->truesize >=
3208
	    (unsigned int)sk->sk_rcvbuf)
3209 3210 3211 3212 3213 3214 3215
		return -ENOMEM;

	skb_orphan(skb);
	skb->sk = sk;
	skb->destructor = sock_rmem_free;
	atomic_add(skb->truesize, &sk->sk_rmem_alloc);

3216 3217 3218
	/* before exiting rcu section, make sure dst is refcounted */
	skb_dst_force(skb);

3219 3220
	skb_queue_tail(&sk->sk_error_queue, skb);
	if (!sock_flag(sk, SOCK_DEAD))
3221
		sk->sk_data_ready(sk, len);
3222 3223 3224 3225
	return 0;
}
EXPORT_SYMBOL(sock_queue_err_skb);

3226 3227 3228 3229 3230 3231 3232 3233 3234 3235 3236 3237 3238 3239 3240 3241 3242 3243 3244 3245 3246
void skb_tstamp_tx(struct sk_buff *orig_skb,
		struct skb_shared_hwtstamps *hwtstamps)
{
	struct sock *sk = orig_skb->sk;
	struct sock_exterr_skb *serr;
	struct sk_buff *skb;
	int err;

	if (!sk)
		return;

	skb = skb_clone(orig_skb, GFP_ATOMIC);
	if (!skb)
		return;

	if (hwtstamps) {
		*skb_hwtstamps(skb) =
			*hwtstamps;
	} else {
		/*
		 * no hardware time stamps available,
3247
		 * so keep the shared tx_flags and only
3248 3249 3250 3251 3252 3253 3254 3255 3256
		 * store software time stamp
		 */
		skb->tstamp = ktime_get_real();
	}

	serr = SKB_EXT_ERR(skb);
	memset(serr, 0, sizeof(*serr));
	serr->ee.ee_errno = ENOMSG;
	serr->ee.ee_origin = SO_EE_ORIGIN_TIMESTAMPING;
3257

3258
	err = sock_queue_err_skb(sk, skb);
3259

3260 3261 3262 3263 3264
	if (err)
		kfree_skb(skb);
}
EXPORT_SYMBOL_GPL(skb_tstamp_tx);

3265 3266 3267 3268 3269 3270 3271 3272 3273 3274 3275 3276 3277 3278 3279 3280 3281 3282 3283 3284
void skb_complete_wifi_ack(struct sk_buff *skb, bool acked)
{
	struct sock *sk = skb->sk;
	struct sock_exterr_skb *serr;
	int err;

	skb->wifi_acked_valid = 1;
	skb->wifi_acked = acked;

	serr = SKB_EXT_ERR(skb);
	memset(serr, 0, sizeof(*serr));
	serr->ee.ee_errno = ENOMSG;
	serr->ee.ee_origin = SO_EE_ORIGIN_TXSTATUS;

	err = sock_queue_err_skb(sk, skb);
	if (err)
		kfree_skb(skb);
}
EXPORT_SYMBOL_GPL(skb_complete_wifi_ack);

3285

3286 3287 3288 3289 3290 3291 3292 3293 3294 3295 3296 3297 3298 3299
/**
 * skb_partial_csum_set - set up and verify partial csum values for packet
 * @skb: the skb to set
 * @start: the number of bytes after skb->data to start checksumming.
 * @off: the offset from start to place the checksum.
 *
 * For untrusted partially-checksummed packets, we need to make sure the values
 * for skb->csum_start and skb->csum_offset are valid so we don't oops.
 *
 * This function checks and sets those values and skb->ip_summed: if this
 * returns false you should drop the packet.
 */
bool skb_partial_csum_set(struct sk_buff *skb, u16 start, u16 off)
{
3300 3301
	if (unlikely(start > skb_headlen(skb)) ||
	    unlikely((int)start + off > skb_headlen(skb) - 2)) {
3302 3303 3304
		if (net_ratelimit())
			printk(KERN_WARNING
			       "bad partial csum: csum=%u/%u len=%u\n",
3305
			       start, off, skb_headlen(skb));
3306 3307 3308 3309 3310 3311 3312
		return false;
	}
	skb->ip_summed = CHECKSUM_PARTIAL;
	skb->csum_start = skb_headroom(skb) + start;
	skb->csum_offset = off;
	return true;
}
3313
EXPORT_SYMBOL_GPL(skb_partial_csum_set);
3314

3315 3316 3317 3318 3319 3320 3321
void __skb_warn_lro_forwarding(const struct sk_buff *skb)
{
	if (net_ratelimit())
		pr_warning("%s: received packets cannot be forwarded"
			   " while LRO is enabled\n", skb->dev->name);
}
EXPORT_SYMBOL(__skb_warn_lro_forwarding);