bio.c 29.7 KB
Newer Older
L
Linus Torvalds 已提交
1
/*
2
 * Copyright (C) 2001 Jens Axboe <axboe@kernel.dk>
L
Linus Torvalds 已提交
3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27
 *
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License version 2 as
 * published by the Free Software Foundation.
 *
 * This program is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 * GNU General Public License for more details.
 *
 * You should have received a copy of the GNU General Public Licens
 * along with this program; if not, write to the Free Software
 * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-
 *
 */
#include <linux/mm.h>
#include <linux/swap.h>
#include <linux/bio.h>
#include <linux/blkdev.h>
#include <linux/slab.h>
#include <linux/init.h>
#include <linux/kernel.h>
#include <linux/module.h>
#include <linux/mempool.h>
#include <linux/workqueue.h>
28
#include <linux/blktrace_api.h>
29
#include <scsi/sg.h>		/* for struct sg_iovec */
L
Linus Torvalds 已提交
30 31 32

#define BIO_POOL_SIZE 256

33
static struct kmem_cache *bio_slab __read_mostly;
L
Linus Torvalds 已提交
34 35 36 37 38 39 40 41

#define BIOVEC_NR_POOLS 6

/*
 * a small number of entries is fine, not going to be performance critical.
 * basically we just need to survive
 */
#define BIO_SPLIT_ENTRIES 8	
42
mempool_t *bio_split_pool __read_mostly;
L
Linus Torvalds 已提交
43 44 45 46

struct biovec_slab {
	int nr_vecs;
	char *name; 
47
	struct kmem_cache *slab;
L
Linus Torvalds 已提交
48 49 50 51 52 53 54 55 56
};

/*
 * if you change this list, also change bvec_alloc or things will
 * break badly! cannot be bigger than what you can fit into an
 * unsigned short
 */

#define BV(x) { .nr_vecs = x, .name = "biovec-"__stringify(x) }
57
static struct biovec_slab bvec_slabs[BIOVEC_NR_POOLS] __read_mostly = {
L
Linus Torvalds 已提交
58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78
	BV(1), BV(4), BV(16), BV(64), BV(128), BV(BIO_MAX_PAGES),
};
#undef BV

/*
 * bio_set is used to allow other portions of the IO system to
 * allocate their own private memory pools for bio and iovec structures.
 * These memory pools in turn all allocate from the bio_slab
 * and the bvec_slabs[].
 */
struct bio_set {
	mempool_t *bio_pool;
	mempool_t *bvec_pools[BIOVEC_NR_POOLS];
};

/*
 * fs_bio_set is the bio_set containing bio and iovec memory pools used by
 * IO code that does not need private memory pools.
 */
static struct bio_set *fs_bio_set;

A
Al Viro 已提交
79
static inline struct bio_vec *bvec_alloc_bs(gfp_t gfp_mask, int nr, unsigned long *idx, struct bio_set *bs)
L
Linus Torvalds 已提交
80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100
{
	struct bio_vec *bvl;

	/*
	 * see comment near bvec_array define!
	 */
	switch (nr) {
		case   1        : *idx = 0; break;
		case   2 ...   4: *idx = 1; break;
		case   5 ...  16: *idx = 2; break;
		case  17 ...  64: *idx = 3; break;
		case  65 ... 128: *idx = 4; break;
		case 129 ... BIO_MAX_PAGES: *idx = 5; break;
		default:
			return NULL;
	}
	/*
	 * idx now points to the pool we want to allocate from
	 */

	bvl = mempool_alloc(bs->bvec_pools[*idx], gfp_mask);
A
Andreas Mohr 已提交
101 102 103
	if (bvl) {
		struct biovec_slab *bp = bvec_slabs + *idx;

L
Linus Torvalds 已提交
104
		memset(bvl, 0, bp->nr_vecs * sizeof(struct bio_vec));
A
Andreas Mohr 已提交
105
	}
L
Linus Torvalds 已提交
106 107 108 109

	return bvl;
}

P
Peter Osterlund 已提交
110
void bio_free(struct bio *bio, struct bio_set *bio_set)
L
Linus Torvalds 已提交
111 112 113 114 115
{
	const int pool_idx = BIO_POOL_IDX(bio);

	BIO_BUG_ON(pool_idx >= BIOVEC_NR_POOLS);

P
Peter Osterlund 已提交
116 117 118 119 120 121 122 123 124 125
	mempool_free(bio->bi_io_vec, bio_set->bvec_pools[pool_idx]);
	mempool_free(bio, bio_set->bio_pool);
}

/*
 * default destructor for a bio allocated with bio_alloc_bioset()
 */
static void bio_fs_destructor(struct bio *bio)
{
	bio_free(bio, fs_bio_set);
L
Linus Torvalds 已提交
126 127
}

128
void bio_init(struct bio *bio)
L
Linus Torvalds 已提交
129 130
{
	bio->bi_next = NULL;
131
	bio->bi_bdev = NULL;
L
Linus Torvalds 已提交
132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150
	bio->bi_flags = 1 << BIO_UPTODATE;
	bio->bi_rw = 0;
	bio->bi_vcnt = 0;
	bio->bi_idx = 0;
	bio->bi_phys_segments = 0;
	bio->bi_hw_segments = 0;
	bio->bi_hw_front_size = 0;
	bio->bi_hw_back_size = 0;
	bio->bi_size = 0;
	bio->bi_max_vecs = 0;
	bio->bi_end_io = NULL;
	atomic_set(&bio->bi_cnt, 1);
	bio->bi_private = NULL;
}

/**
 * bio_alloc_bioset - allocate a bio for I/O
 * @gfp_mask:   the GFP_ mask given to the slab allocator
 * @nr_iovecs:	number of iovecs to pre-allocate
151
 * @bs:		the bio_set to allocate from
L
Linus Torvalds 已提交
152 153 154 155 156 157 158 159 160
 *
 * Description:
 *   bio_alloc_bioset will first try it's on mempool to satisfy the allocation.
 *   If %__GFP_WAIT is set then we will block on the internal pool waiting
 *   for a &struct bio to become free.
 *
 *   allocate bio and iovecs from the memory pools specified by the
 *   bio_set structure.
 **/
A
Al Viro 已提交
161
struct bio *bio_alloc_bioset(gfp_t gfp_mask, int nr_iovecs, struct bio_set *bs)
L
Linus Torvalds 已提交
162 163 164 165 166 167 168 169
{
	struct bio *bio = mempool_alloc(bs->bio_pool, gfp_mask);

	if (likely(bio)) {
		struct bio_vec *bvl = NULL;

		bio_init(bio);
		if (likely(nr_iovecs)) {
A
Andreas Mohr 已提交
170
			unsigned long idx = 0; /* shut up gcc */
L
Linus Torvalds 已提交
171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186

			bvl = bvec_alloc_bs(gfp_mask, nr_iovecs, &idx, bs);
			if (unlikely(!bvl)) {
				mempool_free(bio, bs->bio_pool);
				bio = NULL;
				goto out;
			}
			bio->bi_flags |= idx << BIO_POOL_OFFSET;
			bio->bi_max_vecs = bvec_slabs[idx].nr_vecs;
		}
		bio->bi_io_vec = bvl;
	}
out:
	return bio;
}

A
Al Viro 已提交
187
struct bio *bio_alloc(gfp_t gfp_mask, int nr_iovecs)
L
Linus Torvalds 已提交
188
{
P
Peter Osterlund 已提交
189 190 191 192 193 194
	struct bio *bio = bio_alloc_bioset(gfp_mask, nr_iovecs, fs_bio_set);

	if (bio)
		bio->bi_destructor = bio_fs_destructor;

	return bio;
L
Linus Torvalds 已提交
195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257
}

void zero_fill_bio(struct bio *bio)
{
	unsigned long flags;
	struct bio_vec *bv;
	int i;

	bio_for_each_segment(bv, bio, i) {
		char *data = bvec_kmap_irq(bv, &flags);
		memset(data, 0, bv->bv_len);
		flush_dcache_page(bv->bv_page);
		bvec_kunmap_irq(data, &flags);
	}
}
EXPORT_SYMBOL(zero_fill_bio);

/**
 * bio_put - release a reference to a bio
 * @bio:   bio to release reference to
 *
 * Description:
 *   Put a reference to a &struct bio, either one you have gotten with
 *   bio_alloc or bio_get. The last put of a bio will free it.
 **/
void bio_put(struct bio *bio)
{
	BIO_BUG_ON(!atomic_read(&bio->bi_cnt));

	/*
	 * last put frees it
	 */
	if (atomic_dec_and_test(&bio->bi_cnt)) {
		bio->bi_next = NULL;
		bio->bi_destructor(bio);
	}
}

inline int bio_phys_segments(request_queue_t *q, struct bio *bio)
{
	if (unlikely(!bio_flagged(bio, BIO_SEG_VALID)))
		blk_recount_segments(q, bio);

	return bio->bi_phys_segments;
}

inline int bio_hw_segments(request_queue_t *q, struct bio *bio)
{
	if (unlikely(!bio_flagged(bio, BIO_SEG_VALID)))
		blk_recount_segments(q, bio);

	return bio->bi_hw_segments;
}

/**
 * 	__bio_clone	-	clone a bio
 * 	@bio: destination bio
 * 	@bio_src: bio to clone
 *
 *	Clone a &bio. Caller will own the returned bio, but not
 *	the actual data it points to. Reference count of returned
 * 	bio will be one.
 */
258
void __bio_clone(struct bio *bio, struct bio *bio_src)
L
Linus Torvalds 已提交
259 260 261
{
	request_queue_t *q = bdev_get_queue(bio_src->bi_bdev);

262 263
	memcpy(bio->bi_io_vec, bio_src->bi_io_vec,
		bio_src->bi_max_vecs * sizeof(struct bio_vec));
L
Linus Torvalds 已提交
264 265 266 267 268 269 270

	bio->bi_sector = bio_src->bi_sector;
	bio->bi_bdev = bio_src->bi_bdev;
	bio->bi_flags |= 1 << BIO_CLONED;
	bio->bi_rw = bio_src->bi_rw;
	bio->bi_vcnt = bio_src->bi_vcnt;
	bio->bi_size = bio_src->bi_size;
A
Andrew Morton 已提交
271
	bio->bi_idx = bio_src->bi_idx;
L
Linus Torvalds 已提交
272 273 274 275 276 277 278 279 280 281 282
	bio_phys_segments(q, bio);
	bio_hw_segments(q, bio);
}

/**
 *	bio_clone	-	clone a bio
 *	@bio: bio to clone
 *	@gfp_mask: allocation priority
 *
 * 	Like __bio_clone, only also allocates the returned bio
 */
A
Al Viro 已提交
283
struct bio *bio_clone(struct bio *bio, gfp_t gfp_mask)
L
Linus Torvalds 已提交
284 285 286
{
	struct bio *b = bio_alloc_bioset(gfp_mask, bio->bi_max_vecs, fs_bio_set);

P
Peter Osterlund 已提交
287 288
	if (b) {
		b->bi_destructor = bio_fs_destructor;
L
Linus Torvalds 已提交
289
		__bio_clone(b, bio);
P
Peter Osterlund 已提交
290
	}
L
Linus Torvalds 已提交
291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318

	return b;
}

/**
 *	bio_get_nr_vecs		- return approx number of vecs
 *	@bdev:  I/O target
 *
 *	Return the approximate number of pages we can send to this target.
 *	There's no guarantee that you will be able to fit this number of pages
 *	into a bio, it does not account for dynamic restrictions that vary
 *	on offset.
 */
int bio_get_nr_vecs(struct block_device *bdev)
{
	request_queue_t *q = bdev_get_queue(bdev);
	int nr_pages;

	nr_pages = ((q->max_sectors << 9) + PAGE_SIZE - 1) >> PAGE_SHIFT;
	if (nr_pages > q->max_phys_segments)
		nr_pages = q->max_phys_segments;
	if (nr_pages > q->max_hw_segments)
		nr_pages = q->max_hw_segments;

	return nr_pages;
}

static int __bio_add_page(request_queue_t *q, struct bio *bio, struct page
319 320
			  *page, unsigned int len, unsigned int offset,
			  unsigned short max_sectors)
L
Linus Torvalds 已提交
321 322 323 324 325 326 327 328 329 330
{
	int retried_segments = 0;
	struct bio_vec *bvec;

	/*
	 * cloned bio must not modify vec list
	 */
	if (unlikely(bio_flagged(bio, BIO_CLONED)))
		return 0;

331
	if (((bio->bi_size + len) >> 9) > max_sectors)
L
Linus Torvalds 已提交
332 333
		return 0;

334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355
	/*
	 * For filesystems with a blocksize smaller than the pagesize
	 * we will often be called with the same page as last time and
	 * a consecutive offset.  Optimize this special case.
	 */
	if (bio->bi_vcnt > 0) {
		struct bio_vec *prev = &bio->bi_io_vec[bio->bi_vcnt - 1];

		if (page == prev->bv_page &&
		    offset == prev->bv_offset + prev->bv_len) {
			prev->bv_len += len;
			if (q->merge_bvec_fn &&
			    q->merge_bvec_fn(q, bio, prev) < len) {
				prev->bv_len -= len;
				return 0;
			}

			goto done;
		}
	}

	if (bio->bi_vcnt >= bio->bi_max_vecs)
L
Linus Torvalds 已提交
356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408
		return 0;

	/*
	 * we might lose a segment or two here, but rather that than
	 * make this too complex.
	 */

	while (bio->bi_phys_segments >= q->max_phys_segments
	       || bio->bi_hw_segments >= q->max_hw_segments
	       || BIOVEC_VIRT_OVERSIZE(bio->bi_size)) {

		if (retried_segments)
			return 0;

		retried_segments = 1;
		blk_recount_segments(q, bio);
	}

	/*
	 * setup the new entry, we might clear it again later if we
	 * cannot add the page
	 */
	bvec = &bio->bi_io_vec[bio->bi_vcnt];
	bvec->bv_page = page;
	bvec->bv_len = len;
	bvec->bv_offset = offset;

	/*
	 * if queue has other restrictions (eg varying max sector size
	 * depending on offset), it can specify a merge_bvec_fn in the
	 * queue to get further control
	 */
	if (q->merge_bvec_fn) {
		/*
		 * merge_bvec_fn() returns number of bytes it can accept
		 * at this offset
		 */
		if (q->merge_bvec_fn(q, bio, bvec) < len) {
			bvec->bv_page = NULL;
			bvec->bv_len = 0;
			bvec->bv_offset = 0;
			return 0;
		}
	}

	/* If we may be able to merge these biovecs, force a recount */
	if (bio->bi_vcnt && (BIOVEC_PHYS_MERGEABLE(bvec-1, bvec) ||
	    BIOVEC_VIRT_MERGEABLE(bvec-1, bvec)))
		bio->bi_flags &= ~(1 << BIO_SEG_VALID);

	bio->bi_vcnt++;
	bio->bi_phys_segments++;
	bio->bi_hw_segments++;
409
 done:
L
Linus Torvalds 已提交
410 411 412 413
	bio->bi_size += len;
	return len;
}

414 415
/**
 *	bio_add_pc_page	-	attempt to add page to bio
J
Jens Axboe 已提交
416
 *	@q: the target queue
417 418 419 420 421 422 423 424 425 426 427 428 429 430
 *	@bio: destination bio
 *	@page: page to add
 *	@len: vec entry length
 *	@offset: vec entry offset
 *
 *	Attempt to add a page to the bio_vec maplist. This can fail for a
 *	number of reasons, such as the bio being full or target block
 *	device limitations. The target block device must allow bio's
 *      smaller than PAGE_SIZE, so it is always possible to add a single
 *      page to an empty bio. This should only be used by REQ_PC bios.
 */
int bio_add_pc_page(request_queue_t *q, struct bio *bio, struct page *page,
		    unsigned int len, unsigned int offset)
{
431
	return __bio_add_page(q, bio, page, len, offset, q->max_hw_sectors);
432 433
}

L
Linus Torvalds 已提交
434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449
/**
 *	bio_add_page	-	attempt to add page to bio
 *	@bio: destination bio
 *	@page: page to add
 *	@len: vec entry length
 *	@offset: vec entry offset
 *
 *	Attempt to add a page to the bio_vec maplist. This can fail for a
 *	number of reasons, such as the bio being full or target block
 *	device limitations. The target block device must allow bio's
 *      smaller than PAGE_SIZE, so it is always possible to add a single
 *      page to an empty bio.
 */
int bio_add_page(struct bio *bio, struct page *page, unsigned int len,
		 unsigned int offset)
{
450 451
	struct request_queue *q = bdev_get_queue(bio->bi_bdev);
	return __bio_add_page(q, bio, page, len, offset, q->max_sectors);
L
Linus Torvalds 已提交
452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562
}

struct bio_map_data {
	struct bio_vec *iovecs;
	void __user *userptr;
};

static void bio_set_map_data(struct bio_map_data *bmd, struct bio *bio)
{
	memcpy(bmd->iovecs, bio->bi_io_vec, sizeof(struct bio_vec) * bio->bi_vcnt);
	bio->bi_private = bmd;
}

static void bio_free_map_data(struct bio_map_data *bmd)
{
	kfree(bmd->iovecs);
	kfree(bmd);
}

static struct bio_map_data *bio_alloc_map_data(int nr_segs)
{
	struct bio_map_data *bmd = kmalloc(sizeof(*bmd), GFP_KERNEL);

	if (!bmd)
		return NULL;

	bmd->iovecs = kmalloc(sizeof(struct bio_vec) * nr_segs, GFP_KERNEL);
	if (bmd->iovecs)
		return bmd;

	kfree(bmd);
	return NULL;
}

/**
 *	bio_uncopy_user	-	finish previously mapped bio
 *	@bio: bio being terminated
 *
 *	Free pages allocated from bio_copy_user() and write back data
 *	to user space in case of a read.
 */
int bio_uncopy_user(struct bio *bio)
{
	struct bio_map_data *bmd = bio->bi_private;
	const int read = bio_data_dir(bio) == READ;
	struct bio_vec *bvec;
	int i, ret = 0;

	__bio_for_each_segment(bvec, bio, i, 0) {
		char *addr = page_address(bvec->bv_page);
		unsigned int len = bmd->iovecs[i].bv_len;

		if (read && !ret && copy_to_user(bmd->userptr, addr, len))
			ret = -EFAULT;

		__free_page(bvec->bv_page);
		bmd->userptr += len;
	}
	bio_free_map_data(bmd);
	bio_put(bio);
	return ret;
}

/**
 *	bio_copy_user	-	copy user data to bio
 *	@q: destination block queue
 *	@uaddr: start of user address
 *	@len: length in bytes
 *	@write_to_vm: bool indicating writing to pages or not
 *
 *	Prepares and returns a bio for indirect user io, bouncing data
 *	to/from kernel pages as necessary. Must be paired with
 *	call bio_uncopy_user() on io completion.
 */
struct bio *bio_copy_user(request_queue_t *q, unsigned long uaddr,
			  unsigned int len, int write_to_vm)
{
	unsigned long end = (uaddr + len + PAGE_SIZE - 1) >> PAGE_SHIFT;
	unsigned long start = uaddr >> PAGE_SHIFT;
	struct bio_map_data *bmd;
	struct bio_vec *bvec;
	struct page *page;
	struct bio *bio;
	int i, ret;

	bmd = bio_alloc_map_data(end - start);
	if (!bmd)
		return ERR_PTR(-ENOMEM);

	bmd->userptr = (void __user *) uaddr;

	ret = -ENOMEM;
	bio = bio_alloc(GFP_KERNEL, end - start);
	if (!bio)
		goto out_bmd;

	bio->bi_rw |= (!write_to_vm << BIO_RW);

	ret = 0;
	while (len) {
		unsigned int bytes = PAGE_SIZE;

		if (bytes > len)
			bytes = len;

		page = alloc_page(q->bounce_gfp | GFP_KERNEL);
		if (!page) {
			ret = -ENOMEM;
			break;
		}

563
		if (bio_add_pc_page(q, bio, page, bytes, 0) < bytes)
L
Linus Torvalds 已提交
564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602
			break;

		len -= bytes;
	}

	if (ret)
		goto cleanup;

	/*
	 * success
	 */
	if (!write_to_vm) {
		char __user *p = (char __user *) uaddr;

		/*
		 * for a write, copy in data to kernel pages
		 */
		ret = -EFAULT;
		bio_for_each_segment(bvec, bio, i) {
			char *addr = page_address(bvec->bv_page);

			if (copy_from_user(addr, p, bvec->bv_len))
				goto cleanup;
			p += bvec->bv_len;
		}
	}

	bio_set_map_data(bmd, bio);
	return bio;
cleanup:
	bio_for_each_segment(bvec, bio, i)
		__free_page(bvec->bv_page);

	bio_put(bio);
out_bmd:
	bio_free_map_data(bmd);
	return ERR_PTR(ret);
}

603 604 605 606
static struct bio *__bio_map_user_iov(request_queue_t *q,
				      struct block_device *bdev,
				      struct sg_iovec *iov, int iov_count,
				      int write_to_vm)
L
Linus Torvalds 已提交
607
{
608 609
	int i, j;
	int nr_pages = 0;
L
Linus Torvalds 已提交
610 611
	struct page **pages;
	struct bio *bio;
612 613
	int cur_page = 0;
	int ret, offset;
L
Linus Torvalds 已提交
614

615 616 617 618 619 620 621 622
	for (i = 0; i < iov_count; i++) {
		unsigned long uaddr = (unsigned long)iov[i].iov_base;
		unsigned long len = iov[i].iov_len;
		unsigned long end = (uaddr + len + PAGE_SIZE - 1) >> PAGE_SHIFT;
		unsigned long start = uaddr >> PAGE_SHIFT;

		nr_pages += end - start;
		/*
623
		 * buffer must be aligned to at least hardsector size for now
624
		 */
625
		if (uaddr & queue_dma_alignment(q))
626 627 628 629
			return ERR_PTR(-EINVAL);
	}

	if (!nr_pages)
L
Linus Torvalds 已提交
630 631 632 633 634 635 636
		return ERR_PTR(-EINVAL);

	bio = bio_alloc(GFP_KERNEL, nr_pages);
	if (!bio)
		return ERR_PTR(-ENOMEM);

	ret = -ENOMEM;
637
	pages = kcalloc(nr_pages, sizeof(struct page *), GFP_KERNEL);
L
Linus Torvalds 已提交
638 639 640
	if (!pages)
		goto out;

641 642 643 644 645 646 647 648 649 650 651 652 653 654
	for (i = 0; i < iov_count; i++) {
		unsigned long uaddr = (unsigned long)iov[i].iov_base;
		unsigned long len = iov[i].iov_len;
		unsigned long end = (uaddr + len + PAGE_SIZE - 1) >> PAGE_SHIFT;
		unsigned long start = uaddr >> PAGE_SHIFT;
		const int local_nr_pages = end - start;
		const int page_limit = cur_page + local_nr_pages;
		
		down_read(&current->mm->mmap_sem);
		ret = get_user_pages(current, current->mm, uaddr,
				     local_nr_pages,
				     write_to_vm, 0, &pages[cur_page], NULL);
		up_read(&current->mm->mmap_sem);

655 656
		if (ret < local_nr_pages) {
			ret = -EFAULT;
657
			goto out_unmap;
658
		}
659 660 661 662 663 664 665 666 667 668 669 670 671 672

		offset = uaddr & ~PAGE_MASK;
		for (j = cur_page; j < page_limit; j++) {
			unsigned int bytes = PAGE_SIZE - offset;

			if (len <= 0)
				break;
			
			if (bytes > len)
				bytes = len;

			/*
			 * sorry...
			 */
673 674
			if (bio_add_pc_page(q, bio, pages[j], bytes, offset) <
					    bytes)
675 676 677 678 679
				break;

			len -= bytes;
			offset = 0;
		}
L
Linus Torvalds 已提交
680

681
		cur_page = j;
L
Linus Torvalds 已提交
682
		/*
683
		 * release the pages we didn't map into the bio, if any
L
Linus Torvalds 已提交
684
		 */
685 686
		while (j < page_limit)
			page_cache_release(pages[j++]);
L
Linus Torvalds 已提交
687 688 689 690 691 692 693 694 695 696
	}

	kfree(pages);

	/*
	 * set data direction, and check if mapped pages need bouncing
	 */
	if (!write_to_vm)
		bio->bi_rw |= (1 << BIO_RW);

697
	bio->bi_bdev = bdev;
L
Linus Torvalds 已提交
698 699
	bio->bi_flags |= (1 << BIO_USER_MAPPED);
	return bio;
700 701 702 703 704 705 706 707

 out_unmap:
	for (i = 0; i < nr_pages; i++) {
		if(!pages[i])
			break;
		page_cache_release(pages[i]);
	}
 out:
L
Linus Torvalds 已提交
708 709 710 711 712 713 714
	kfree(pages);
	bio_put(bio);
	return ERR_PTR(ret);
}

/**
 *	bio_map_user	-	map user address into bio
715
 *	@q: the request_queue_t for the bio
L
Linus Torvalds 已提交
716 717 718 719 720 721 722 723 724 725
 *	@bdev: destination block device
 *	@uaddr: start of user address
 *	@len: length in bytes
 *	@write_to_vm: bool indicating writing to pages or not
 *
 *	Map the user space address into a bio suitable for io to a block
 *	device. Returns an error pointer in case of error.
 */
struct bio *bio_map_user(request_queue_t *q, struct block_device *bdev,
			 unsigned long uaddr, unsigned int len, int write_to_vm)
726 727 728
{
	struct sg_iovec iov;

729
	iov.iov_base = (void __user *)uaddr;
730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748
	iov.iov_len = len;

	return bio_map_user_iov(q, bdev, &iov, 1, write_to_vm);
}

/**
 *	bio_map_user_iov - map user sg_iovec table into bio
 *	@q: the request_queue_t for the bio
 *	@bdev: destination block device
 *	@iov:	the iovec.
 *	@iov_count: number of elements in the iovec
 *	@write_to_vm: bool indicating writing to pages or not
 *
 *	Map the user space address into a bio suitable for io to a block
 *	device. Returns an error pointer in case of error.
 */
struct bio *bio_map_user_iov(request_queue_t *q, struct block_device *bdev,
			     struct sg_iovec *iov, int iov_count,
			     int write_to_vm)
L
Linus Torvalds 已提交
749 750 751
{
	struct bio *bio;

752
	bio = __bio_map_user_iov(q, bdev, iov, iov_count, write_to_vm);
L
Linus Torvalds 已提交
753 754 755 756 757 758 759 760 761 762 763 764

	if (IS_ERR(bio))
		return bio;

	/*
	 * subtle -- if __bio_map_user() ended up bouncing a bio,
	 * it would normally disappear when its bi_end_io is run.
	 * however, we need it for the unmap, so grab an extra
	 * reference to it
	 */
	bio_get(bio);

765
	return bio;
L
Linus Torvalds 已提交
766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800
}

static void __bio_unmap_user(struct bio *bio)
{
	struct bio_vec *bvec;
	int i;

	/*
	 * make sure we dirty pages we wrote to
	 */
	__bio_for_each_segment(bvec, bio, i, 0) {
		if (bio_data_dir(bio) == READ)
			set_page_dirty_lock(bvec->bv_page);

		page_cache_release(bvec->bv_page);
	}

	bio_put(bio);
}

/**
 *	bio_unmap_user	-	unmap a bio
 *	@bio:		the bio being unmapped
 *
 *	Unmap a bio previously mapped by bio_map_user(). Must be called with
 *	a process context.
 *
 *	bio_unmap_user() may sleep.
 */
void bio_unmap_user(struct bio *bio)
{
	__bio_unmap_user(bio);
	bio_put(bio);
}

801 802 803 804 805 806 807 808 809 810
static int bio_map_kern_endio(struct bio *bio, unsigned int bytes_done, int err)
{
	if (bio->bi_size)
		return 1;

	bio_put(bio);
	return 0;
}


M
Mike Christie 已提交
811
static struct bio *__bio_map_kern(request_queue_t *q, void *data,
A
Al Viro 已提交
812
				  unsigned int len, gfp_t gfp_mask)
M
Mike Christie 已提交
813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834
{
	unsigned long kaddr = (unsigned long)data;
	unsigned long end = (kaddr + len + PAGE_SIZE - 1) >> PAGE_SHIFT;
	unsigned long start = kaddr >> PAGE_SHIFT;
	const int nr_pages = end - start;
	int offset, i;
	struct bio *bio;

	bio = bio_alloc(gfp_mask, nr_pages);
	if (!bio)
		return ERR_PTR(-ENOMEM);

	offset = offset_in_page(kaddr);
	for (i = 0; i < nr_pages; i++) {
		unsigned int bytes = PAGE_SIZE - offset;

		if (len <= 0)
			break;

		if (bytes > len)
			bytes = len;

835 836
		if (bio_add_pc_page(q, bio, virt_to_page(data), bytes,
				    offset) < bytes)
M
Mike Christie 已提交
837 838 839 840 841 842 843
			break;

		data += bytes;
		len -= bytes;
		offset = 0;
	}

844
	bio->bi_end_io = bio_map_kern_endio;
M
Mike Christie 已提交
845 846 847 848 849 850 851 852 853 854 855 856 857 858
	return bio;
}

/**
 *	bio_map_kern	-	map kernel address into bio
 *	@q: the request_queue_t for the bio
 *	@data: pointer to buffer to map
 *	@len: length in bytes
 *	@gfp_mask: allocation flags for bio allocation
 *
 *	Map the kernel address into a bio suitable for io to a block
 *	device. Returns an error pointer in case of error.
 */
struct bio *bio_map_kern(request_queue_t *q, void *data, unsigned int len,
A
Al Viro 已提交
859
			 gfp_t gfp_mask)
M
Mike Christie 已提交
860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876
{
	struct bio *bio;

	bio = __bio_map_kern(q, data, len, gfp_mask);
	if (IS_ERR(bio))
		return bio;

	if (bio->bi_size == len)
		return bio;

	/*
	 * Don't support partial mappings.
	 */
	bio_put(bio);
	return ERR_PTR(-EINVAL);
}

L
Linus Torvalds 已提交
877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942
/*
 * bio_set_pages_dirty() and bio_check_pages_dirty() are support functions
 * for performing direct-IO in BIOs.
 *
 * The problem is that we cannot run set_page_dirty() from interrupt context
 * because the required locks are not interrupt-safe.  So what we can do is to
 * mark the pages dirty _before_ performing IO.  And in interrupt context,
 * check that the pages are still dirty.   If so, fine.  If not, redirty them
 * in process context.
 *
 * We special-case compound pages here: normally this means reads into hugetlb
 * pages.  The logic in here doesn't really work right for compound pages
 * because the VM does not uniformly chase down the head page in all cases.
 * But dirtiness of compound pages is pretty meaningless anyway: the VM doesn't
 * handle them at all.  So we skip compound pages here at an early stage.
 *
 * Note that this code is very hard to test under normal circumstances because
 * direct-io pins the pages with get_user_pages().  This makes
 * is_page_cache_freeable return false, and the VM will not clean the pages.
 * But other code (eg, pdflush) could clean the pages if they are mapped
 * pagecache.
 *
 * Simply disabling the call to bio_set_pages_dirty() is a good way to test the
 * deferred bio dirtying paths.
 */

/*
 * bio_set_pages_dirty() will mark all the bio's pages as dirty.
 */
void bio_set_pages_dirty(struct bio *bio)
{
	struct bio_vec *bvec = bio->bi_io_vec;
	int i;

	for (i = 0; i < bio->bi_vcnt; i++) {
		struct page *page = bvec[i].bv_page;

		if (page && !PageCompound(page))
			set_page_dirty_lock(page);
	}
}

static void bio_release_pages(struct bio *bio)
{
	struct bio_vec *bvec = bio->bi_io_vec;
	int i;

	for (i = 0; i < bio->bi_vcnt; i++) {
		struct page *page = bvec[i].bv_page;

		if (page)
			put_page(page);
	}
}

/*
 * bio_check_pages_dirty() will check that all the BIO's pages are still dirty.
 * If they are, then fine.  If, however, some pages are clean then they must
 * have been written out during the direct-IO read.  So we take another ref on
 * the BIO and the offending pages and re-dirty the pages in process context.
 *
 * It is expected that bio_check_pages_dirty() will wholly own the BIO from
 * here on.  It will run one page_cache_release() against each page and will
 * run one bio_put() against the BIO.
 */

943
static void bio_dirty_fn(struct work_struct *work);
L
Linus Torvalds 已提交
944

945
static DECLARE_WORK(bio_dirty_work, bio_dirty_fn);
L
Linus Torvalds 已提交
946 947 948 949 950 951
static DEFINE_SPINLOCK(bio_dirty_lock);
static struct bio *bio_dirty_list;

/*
 * This runs in process context
 */
952
static void bio_dirty_fn(struct work_struct *work)
L
Linus Torvalds 已提交
953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083
{
	unsigned long flags;
	struct bio *bio;

	spin_lock_irqsave(&bio_dirty_lock, flags);
	bio = bio_dirty_list;
	bio_dirty_list = NULL;
	spin_unlock_irqrestore(&bio_dirty_lock, flags);

	while (bio) {
		struct bio *next = bio->bi_private;

		bio_set_pages_dirty(bio);
		bio_release_pages(bio);
		bio_put(bio);
		bio = next;
	}
}

void bio_check_pages_dirty(struct bio *bio)
{
	struct bio_vec *bvec = bio->bi_io_vec;
	int nr_clean_pages = 0;
	int i;

	for (i = 0; i < bio->bi_vcnt; i++) {
		struct page *page = bvec[i].bv_page;

		if (PageDirty(page) || PageCompound(page)) {
			page_cache_release(page);
			bvec[i].bv_page = NULL;
		} else {
			nr_clean_pages++;
		}
	}

	if (nr_clean_pages) {
		unsigned long flags;

		spin_lock_irqsave(&bio_dirty_lock, flags);
		bio->bi_private = bio_dirty_list;
		bio_dirty_list = bio;
		spin_unlock_irqrestore(&bio_dirty_lock, flags);
		schedule_work(&bio_dirty_work);
	} else {
		bio_put(bio);
	}
}

/**
 * bio_endio - end I/O on a bio
 * @bio:	bio
 * @bytes_done:	number of bytes completed
 * @error:	error, if any
 *
 * Description:
 *   bio_endio() will end I/O on @bytes_done number of bytes. This may be
 *   just a partial part of the bio, or it may be the whole bio. bio_endio()
 *   is the preferred way to end I/O on a bio, it takes care of decrementing
 *   bi_size and clearing BIO_UPTODATE on error. @error is 0 on success, and
 *   and one of the established -Exxxx (-EIO, for instance) error values in
 *   case something went wrong. Noone should call bi_end_io() directly on
 *   a bio unless they own it and thus know that it has an end_io function.
 **/
void bio_endio(struct bio *bio, unsigned int bytes_done, int error)
{
	if (error)
		clear_bit(BIO_UPTODATE, &bio->bi_flags);

	if (unlikely(bytes_done > bio->bi_size)) {
		printk("%s: want %u bytes done, only %u left\n", __FUNCTION__,
						bytes_done, bio->bi_size);
		bytes_done = bio->bi_size;
	}

	bio->bi_size -= bytes_done;
	bio->bi_sector += (bytes_done >> 9);

	if (bio->bi_end_io)
		bio->bi_end_io(bio, bytes_done, error);
}

void bio_pair_release(struct bio_pair *bp)
{
	if (atomic_dec_and_test(&bp->cnt)) {
		struct bio *master = bp->bio1.bi_private;

		bio_endio(master, master->bi_size, bp->error);
		mempool_free(bp, bp->bio2.bi_private);
	}
}

static int bio_pair_end_1(struct bio * bi, unsigned int done, int err)
{
	struct bio_pair *bp = container_of(bi, struct bio_pair, bio1);

	if (err)
		bp->error = err;

	if (bi->bi_size)
		return 1;

	bio_pair_release(bp);
	return 0;
}

static int bio_pair_end_2(struct bio * bi, unsigned int done, int err)
{
	struct bio_pair *bp = container_of(bi, struct bio_pair, bio2);

	if (err)
		bp->error = err;

	if (bi->bi_size)
		return 1;

	bio_pair_release(bp);
	return 0;
}

/*
 * split a bio - only worry about a bio with a single page
 * in it's iovec
 */
struct bio_pair *bio_split(struct bio *bi, mempool_t *pool, int first_sectors)
{
	struct bio_pair *bp = mempool_alloc(pool, GFP_NOIO);

	if (!bp)
		return bp;

1084 1085 1086
	blk_add_trace_pdu_int(bdev_get_queue(bi->bi_bdev), BLK_TA_SPLIT, bi,
				bi->bi_sector + first_sectors);

L
Linus Torvalds 已提交
1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105
	BUG_ON(bi->bi_vcnt != 1);
	BUG_ON(bi->bi_idx != 0);
	atomic_set(&bp->cnt, 3);
	bp->error = 0;
	bp->bio1 = *bi;
	bp->bio2 = *bi;
	bp->bio2.bi_sector += first_sectors;
	bp->bio2.bi_size -= first_sectors << 9;
	bp->bio1.bi_size = first_sectors << 9;

	bp->bv1 = bi->bi_io_vec[0];
	bp->bv2 = bi->bi_io_vec[0];
	bp->bv2.bv_offset += first_sectors << 9;
	bp->bv2.bv_len -= first_sectors << 9;
	bp->bv1.bv_len = first_sectors << 9;

	bp->bio1.bi_io_vec = &bp->bv1;
	bp->bio2.bi_io_vec = &bp->bv2;

1106 1107 1108
	bp->bio1.bi_max_vecs = 1;
	bp->bio2.bi_max_vecs = 1;

L
Linus Torvalds 已提交
1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130
	bp->bio1.bi_end_io = bio_pair_end_1;
	bp->bio2.bi_end_io = bio_pair_end_2;

	bp->bio1.bi_private = bi;
	bp->bio2.bi_private = pool;

	return bp;
}


/*
 * create memory pools for biovec's in a bio_set.
 * use the global biovec slabs created for general use.
 */
static int biovec_create_pools(struct bio_set *bs, int pool_entries, int scale)
{
	int i;

	for (i = 0; i < BIOVEC_NR_POOLS; i++) {
		struct biovec_slab *bp = bvec_slabs + i;
		mempool_t **bvp = bs->bvec_pools + i;

1131
		if (pool_entries > 1 && i >= scale)
L
Linus Torvalds 已提交
1132 1133
			pool_entries >>= 1;

1134
		*bvp = mempool_create_slab_pool(pool_entries, bp->slab);
L
Linus Torvalds 已提交
1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165
		if (!*bvp)
			return -ENOMEM;
	}
	return 0;
}

static void biovec_free_pools(struct bio_set *bs)
{
	int i;

	for (i = 0; i < BIOVEC_NR_POOLS; i++) {
		mempool_t *bvp = bs->bvec_pools[i];

		if (bvp)
			mempool_destroy(bvp);
	}

}

void bioset_free(struct bio_set *bs)
{
	if (bs->bio_pool)
		mempool_destroy(bs->bio_pool);

	biovec_free_pools(bs);

	kfree(bs);
}

struct bio_set *bioset_create(int bio_pool_size, int bvec_pool_size, int scale)
{
1166
	struct bio_set *bs = kzalloc(sizeof(*bs), GFP_KERNEL);
L
Linus Torvalds 已提交
1167 1168 1169 1170

	if (!bs)
		return NULL;

1171
	bs->bio_pool = mempool_create_slab_pool(bio_pool_size, bio_slab);
L
Linus Torvalds 已提交
1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223
	if (!bs->bio_pool)
		goto bad;

	if (!biovec_create_pools(bs, bvec_pool_size, scale))
		return bs;

bad:
	bioset_free(bs);
	return NULL;
}

static void __init biovec_init_slabs(void)
{
	int i;

	for (i = 0; i < BIOVEC_NR_POOLS; i++) {
		int size;
		struct biovec_slab *bvs = bvec_slabs + i;

		size = bvs->nr_vecs * sizeof(struct bio_vec);
		bvs->slab = kmem_cache_create(bvs->name, size, 0,
                                SLAB_HWCACHE_ALIGN|SLAB_PANIC, NULL, NULL);
	}
}

static int __init init_bio(void)
{
	int megabytes, bvec_pool_entries;
	int scale = BIOVEC_NR_POOLS;

	bio_slab = kmem_cache_create("bio", sizeof(struct bio), 0,
				SLAB_HWCACHE_ALIGN|SLAB_PANIC, NULL, NULL);

	biovec_init_slabs();

	megabytes = nr_free_pages() >> (20 - PAGE_SHIFT);

	/*
	 * find out where to start scaling
	 */
	if (megabytes <= 16)
		scale = 0;
	else if (megabytes <= 32)
		scale = 1;
	else if (megabytes <= 64)
		scale = 2;
	else if (megabytes <= 96)
		scale = 3;
	else if (megabytes <= 128)
		scale = 4;

	/*
1224 1225 1226
	 * Limit number of entries reserved -- mempools are only used when
	 * the system is completely unable to allocate memory, so we only
	 * need enough to make progress.
L
Linus Torvalds 已提交
1227
	 */
1228
	bvec_pool_entries = 1 + scale;
L
Linus Torvalds 已提交
1229 1230 1231 1232 1233

	fs_bio_set = bioset_create(BIO_POOL_SIZE, bvec_pool_entries, scale);
	if (!fs_bio_set)
		panic("bio: can't allocate bios\n");

1234 1235
	bio_split_pool = mempool_create_kmalloc_pool(BIO_SPLIT_ENTRIES,
						     sizeof(struct bio_pair));
L
Linus Torvalds 已提交
1236 1237 1238 1239 1240 1241 1242 1243 1244 1245
	if (!bio_split_pool)
		panic("bio: can't create split pool\n");

	return 0;
}

subsys_initcall(init_bio);

EXPORT_SYMBOL(bio_alloc);
EXPORT_SYMBOL(bio_put);
P
Peter Osterlund 已提交
1246
EXPORT_SYMBOL(bio_free);
L
Linus Torvalds 已提交
1247 1248 1249 1250 1251 1252 1253
EXPORT_SYMBOL(bio_endio);
EXPORT_SYMBOL(bio_init);
EXPORT_SYMBOL(__bio_clone);
EXPORT_SYMBOL(bio_clone);
EXPORT_SYMBOL(bio_phys_segments);
EXPORT_SYMBOL(bio_hw_segments);
EXPORT_SYMBOL(bio_add_page);
1254
EXPORT_SYMBOL(bio_add_pc_page);
L
Linus Torvalds 已提交
1255 1256 1257
EXPORT_SYMBOL(bio_get_nr_vecs);
EXPORT_SYMBOL(bio_map_user);
EXPORT_SYMBOL(bio_unmap_user);
M
Mike Christie 已提交
1258
EXPORT_SYMBOL(bio_map_kern);
L
Linus Torvalds 已提交
1259 1260 1261 1262 1263 1264 1265 1266
EXPORT_SYMBOL(bio_pair_release);
EXPORT_SYMBOL(bio_split);
EXPORT_SYMBOL(bio_split_pool);
EXPORT_SYMBOL(bio_copy_user);
EXPORT_SYMBOL(bio_uncopy_user);
EXPORT_SYMBOL(bioset_create);
EXPORT_SYMBOL(bioset_free);
EXPORT_SYMBOL(bio_alloc_bioset);