[linux.git] / block / blk-lib.c

// SPDX-License-Identifier: GPL-2.0
/*
 * Functions related to generic helpers functions
 */
#include <linux/kernel.h>
#include <linux/module.h>
#include <linux/bio.h>
#include <linux/blkdev.h>
#include <linux/scatterlist.h>

#include "blk.h"

struct bio *blk_next_bio(struct bio *bio, unsigned int nr_pages, gfp_t gfp)
{
	struct bio *new = bio_alloc(gfp, nr_pages);

	if (bio) {
		bio_chain(bio, new);
		submit_bio(bio);
	}

	return new;
}

int __blkdev_issue_discard(struct block_device *bdev, sector_t sector,
		sector_t nr_sects, gfp_t gfp_mask, int flags,
		struct bio **biop)
{
	struct request_queue *q = bdev_get_queue(bdev);
	struct bio *bio = *biop;
	unsigned int op;
	sector_t bs_mask;

	if (!q)
		return -ENXIO;

	if (bdev_read_only(bdev))
		return -EPERM;

	if (flags & BLKDEV_DISCARD_SECURE) {
		if (!blk_queue_secure_erase(q))
			return -EOPNOTSUPP;
		op = REQ_OP_SECURE_ERASE;
	} else {
		if (!blk_queue_discard(q))
			return -EOPNOTSUPP;
		op = REQ_OP_DISCARD;
	}

	bs_mask = (bdev_logical_block_size(bdev) >> 9) - 1;
	if ((sector | nr_sects) & bs_mask)
		return -EINVAL;

	if (!nr_sects)
		return -EINVAL;

	while (nr_sects) {
		sector_t req_sects = min_t(sector_t, nr_sects,
				bio_allowed_max_sectors(q));

		WARN_ON_ONCE((req_sects << 9) > UINT_MAX);

		bio = blk_next_bio(bio, 0, gfp_mask);
		bio->bi_iter.bi_sector = sector;
		bio_set_dev(bio, bdev);
		bio_set_op_attrs(bio, op, 0);

		bio->bi_iter.bi_size = req_sects << 9;
		sector += req_sects;
		nr_sects -= req_sects;

		/*
		 * We can loop for a long time in here, if someone does
		 * full device discards (like mkfs). Be nice and allow
		 * us to schedule out to avoid softlocking if preempt
		 * is disabled.
		 */
		cond_resched();
	}

	*biop = bio;
	return 0;
}
EXPORT_SYMBOL(__blkdev_issue_discard);

/**
 * blkdev_issue_discard - queue a discard
 * @bdev:	blockdev to issue discard for
 * @sector:	start sector
 * @nr_sects:	number of sectors to discard
 * @gfp_mask:	memory allocation flags (for bio_alloc)
 * @flags:	BLKDEV_DISCARD_* flags to control behaviour
 *
 * Description:
 *    Issue a discard request for the sectors in question.
 */
int blkdev_issue_discard(struct block_device *bdev, sector_t sector,
		sector_t nr_sects, gfp_t gfp_mask, unsigned long flags)
{
	struct bio *bio = NULL;
	struct blk_plug plug;
	int ret;

	blk_start_plug(&plug);
	ret = __blkdev_issue_discard(bdev, sector, nr_sects, gfp_mask, flags,
			&bio);
	if (!ret && bio) {
		ret = submit_bio_wait(bio);
		if (ret == -EOPNOTSUPP)
			ret = 0;
		bio_put(bio);
	}
	blk_finish_plug(&plug);

	return ret;
}
EXPORT_SYMBOL(blkdev_issue_discard);

/**
 * __blkdev_issue_write_same - generate number of bios with same page
 * @bdev:	target blockdev
 * @sector:	start sector
 * @nr_sects:	number of sectors to write
 * @gfp_mask:	memory allocation flags (for bio_alloc)
 * @page:	page containing data to write
 * @biop:	pointer to anchor bio
 *
 * Description:
 *  Generate and issue number of bios(REQ_OP_WRITE_SAME) with same page.
 */
static int __blkdev_issue_write_same(struct block_device *bdev, sector_t sector,
		sector_t nr_sects, gfp_t gfp_mask, struct page *page,
		struct bio **biop)
{
	struct request_queue *q = bdev_get_queue(bdev);
	unsigned int max_write_same_sectors;
	struct bio *bio = *biop;
	sector_t bs_mask;

	if (!q)
		return -ENXIO;

	if (bdev_read_only(bdev))
		return -EPERM;

	bs_mask = (bdev_logical_block_size(bdev) >> 9) - 1;
	if ((sector | nr_sects) & bs_mask)
		return -EINVAL;

	if (!bdev_write_same(bdev))
		return -EOPNOTSUPP;

	/* Ensure that max_write_same_sectors doesn't overflow bi_size */
	max_write_same_sectors = bio_allowed_max_sectors(q);

	while (nr_sects) {
		bio = blk_next_bio(bio, 1, gfp_mask);
		bio->bi_iter.bi_sector = sector;
		bio_set_dev(bio, bdev);
		bio->bi_vcnt = 1;
		bio->bi_io_vec->bv_page = page;
		bio->bi_io_vec->bv_offset = 0;
		bio->bi_io_vec->bv_len = bdev_logical_block_size(bdev);
		bio_set_op_attrs(bio, REQ_OP_WRITE_SAME, 0);

		if (nr_sects > max_write_same_sectors) {
			bio->bi_iter.bi_size = max_write_same_sectors << 9;
			nr_sects -= max_write_same_sectors;
			sector += max_write_same_sectors;
		} else {
			bio->bi_iter.bi_size = nr_sects << 9;
			nr_sects = 0;
		}
		cond_resched();
	}

	*biop = bio;
	return 0;
}

/**
 * blkdev_issue_write_same - queue a write same operation
 * @bdev:	target blockdev
 * @sector:	start sector
 * @nr_sects:	number of sectors to write
 * @gfp_mask:	memory allocation flags (for bio_alloc)
 * @page:	page containing data
 *
 * Description:
 *    Issue a write same request for the sectors in question.
 */
int blkdev_issue_write_same(struct block_device *bdev, sector_t sector,
				sector_t nr_sects, gfp_t gfp_mask,
				struct page *page)
{
	struct bio *bio = NULL;
	struct blk_plug plug;
	int ret;

	blk_start_plug(&plug);
	ret = __blkdev_issue_write_same(bdev, sector, nr_sects, gfp_mask, page,
			&bio);
	if (ret == 0 && bio) {
		ret = submit_bio_wait(bio);
		bio_put(bio);
	}
	blk_finish_plug(&plug);
	return ret;
}
EXPORT_SYMBOL(blkdev_issue_write_same);

static int __blkdev_issue_write_zeroes(struct block_device *bdev,
		sector_t sector, sector_t nr_sects, gfp_t gfp_mask,
		struct bio **biop, unsigned flags)
{
	struct bio *bio = *biop;
	unsigned int max_write_zeroes_sectors;
	struct request_queue *q = bdev_get_queue(bdev);

	if (!q)
		return -ENXIO;

	if (bdev_read_only(bdev))
		return -EPERM;

	/* Ensure that max_write_zeroes_sectors doesn't overflow bi_size */
	max_write_zeroes_sectors = bdev_write_zeroes_sectors(bdev);

	if (max_write_zeroes_sectors == 0)
		return -EOPNOTSUPP;

	while (nr_sects) {
		bio = blk_next_bio(bio, 0, gfp_mask);
		bio->bi_iter.bi_sector = sector;
		bio_set_dev(bio, bdev);
		bio->bi_opf = REQ_OP_WRITE_ZEROES;
		if (flags & BLKDEV_ZERO_NOUNMAP)
			bio->bi_opf |= REQ_NOUNMAP;

		if (nr_sects > max_write_zeroes_sectors) {
			bio->bi_iter.bi_size = max_write_zeroes_sectors << 9;
			nr_sects -= max_write_zeroes_sectors;
			sector += max_write_zeroes_sectors;
		} else {
			bio->bi_iter.bi_size = nr_sects << 9;
			nr_sects = 0;
		}
		cond_resched();
	}

	*biop = bio;
	return 0;
}

/*
 * Convert a number of 512B sectors to a number of pages.
 * The result is limited to a number of pages that can fit into a BIO.
 * Also make sure that the result is always at least 1 (page) for the cases
 * where nr_sects is lower than the number of sectors in a page.
 */
static unsigned int __blkdev_sectors_to_bio_pages(sector_t nr_sects)
{
	sector_t pages = DIV_ROUND_UP_SECTOR_T(nr_sects, PAGE_SIZE / 512);

	return min(pages, (sector_t)BIO_MAX_PAGES);
}

static int __blkdev_issue_zero_pages(struct block_device *bdev,
		sector_t sector, sector_t nr_sects, gfp_t gfp_mask,
		struct bio **biop)
{
	struct request_queue *q = bdev_get_queue(bdev);
	struct bio *bio = *biop;
	int bi_size = 0;
	unsigned int sz;

	if (!q)
		return -ENXIO;

	if (bdev_read_only(bdev))
		return -EPERM;

	while (nr_sects != 0) {
		bio = blk_next_bio(bio, __blkdev_sectors_to_bio_pages(nr_sects),
				   gfp_mask);
		bio->bi_iter.bi_sector = sector;
		bio_set_dev(bio, bdev);
		bio_set_op_attrs(bio, REQ_OP_WRITE, 0);

		while (nr_sects != 0) {
			sz = min((sector_t) PAGE_SIZE, nr_sects << 9);
			bi_size = bio_add_page(bio, ZERO_PAGE(0), sz, 0);
			nr_sects -= bi_size >> 9;
			sector += bi_size >> 9;
			if (bi_size < sz)
				break;
		}
		cond_resched();
	}

	*biop = bio;
	return 0;
}

/**
 * __blkdev_issue_zeroout - generate number of zero filed write bios
 * @bdev:	blockdev to issue
 * @sector:	start sector
 * @nr_sects:	number of sectors to write
 * @gfp_mask:	memory allocation flags (for bio_alloc)
 * @biop:	pointer to anchor bio
 * @flags:	controls detailed behavior
 *
 * Description:
 *  Zero-fill a block range, either using hardware offload or by explicitly
 *  writing zeroes to the device.
 *
 *  If a device is using logical block provisioning, the underlying space will
 *  not be released if %flags contains BLKDEV_ZERO_NOUNMAP.
 *
 *  If %flags contains BLKDEV_ZERO_NOFALLBACK, the function will return
 *  -EOPNOTSUPP if no explicit hardware offload for zeroing is provided.
 */
int __blkdev_issue_zeroout(struct block_device *bdev, sector_t sector,
		sector_t nr_sects, gfp_t gfp_mask, struct bio **biop,
		unsigned flags)
{
	int ret;
	sector_t bs_mask;

	bs_mask = (bdev_logical_block_size(bdev) >> 9) - 1;
	if ((sector | nr_sects) & bs_mask)
		return -EINVAL;

	ret = __blkdev_issue_write_zeroes(bdev, sector, nr_sects, gfp_mask,
			biop, flags);
	if (ret != -EOPNOTSUPP || (flags & BLKDEV_ZERO_NOFALLBACK))
		return ret;

	return __blkdev_issue_zero_pages(bdev, sector, nr_sects, gfp_mask,
					 biop);
}
EXPORT_SYMBOL(__blkdev_issue_zeroout);

/**
 * blkdev_issue_zeroout - zero-fill a block range
 * @bdev:	blockdev to write
 * @sector:	start sector
 * @nr_sects:	number of sectors to write
 * @gfp_mask:	memory allocation flags (for bio_alloc)
 * @flags:	controls detailed behavior
 *
 * Description:
 *  Zero-fill a block range, either using hardware offload or by explicitly
 *  writing zeroes to the device.  See __blkdev_issue_zeroout() for the
 *  valid values for %flags.
 */
int blkdev_issue_zeroout(struct block_device *bdev, sector_t sector,
		sector_t nr_sects, gfp_t gfp_mask, unsigned flags)
{
	int ret = 0;
	sector_t bs_mask;
	struct bio *bio;
	struct blk_plug plug;
	bool try_write_zeroes = !!bdev_write_zeroes_sectors(bdev);

	bs_mask = (bdev_logical_block_size(bdev) >> 9) - 1;
	if ((sector | nr_sects) & bs_mask)
		return -EINVAL;

retry:
	bio = NULL;
	blk_start_plug(&plug);
	if (try_write_zeroes) {
		ret = __blkdev_issue_write_zeroes(bdev, sector, nr_sects,
						  gfp_mask, &bio, flags);
	} else if (!(flags & BLKDEV_ZERO_NOFALLBACK)) {
		ret = __blkdev_issue_zero_pages(bdev, sector, nr_sects,
						gfp_mask, &bio);
	} else {
		/* No zeroing offload support */
		ret = -EOPNOTSUPP;
	}
	if (ret == 0 && bio) {
		ret = submit_bio_wait(bio);
		bio_put(bio);
	}
	blk_finish_plug(&plug);
	if (ret && try_write_zeroes) {
		if (!(flags & BLKDEV_ZERO_NOFALLBACK)) {
			try_write_zeroes = false;
			goto retry;
		}
		if (!bdev_write_zeroes_sectors(bdev)) {
			/*
			 * Zeroing offload support was indicated, but the
			 * device reported ILLEGAL REQUEST (for some devices
			 * there is no non-destructive way to verify whether
			 * WRITE ZEROES is actually supported).
			 */
			ret = -EOPNOTSUPP;
		}
	}

	return ret;
}
EXPORT_SYMBOL(blkdev_issue_zeroout);
Commit	Line	Data
b2441318	1	// SPDX-License-Identifier: GPL-2.0
f31e7e40 DM	2	/*
	3	* Functions related to generic helpers functions
	4	*/
	5	#include <linux/kernel.h>
	6	#include <linux/module.h>
	7	#include <linux/bio.h>
	8	#include <linux/blkdev.h>
	9	#include <linux/scatterlist.h>
	10
	11	#include "blk.h"
	12
a2d6b3a2	13	struct bio blk_next_bio(struct bio bio, unsigned int nr_pages, gfp_t gfp)
f31e7e40	14	{
9082e87b CH	15	struct bio *new = bio_alloc(gfp, nr_pages);
	16
	17	if (bio) {
	18	bio_chain(bio, new);
4e49ea4a	19	submit_bio(bio);
9082e87b	20	}
5dba3089	21
9082e87b	22	return new;
f31e7e40 DM	23	}
f31e7e40 DM	24
38f25255	25	int __blkdev_issue_discard(struct block_device *bdev, sector_t sector,
288dab8a	26	sector_t nr_sects, gfp_t gfp_mask, int flags,
469e3216	27	struct bio **biop)
f31e7e40	28	{
f31e7e40	29	struct request_queue *q = bdev_get_queue(bdev);
38f25255	30	struct bio bio = biop;
ef295ecf	31	unsigned int op;
28b2be20	32	sector_t bs_mask;
f31e7e40 DM	33
	34	if (!q)
	35	return -ENXIO;
288dab8a	36
a13553c7 ID	37	if (bdev_read_only(bdev))
	38	return -EPERM;
	39
288dab8a CH	40	if (flags & BLKDEV_DISCARD_SECURE) {
	41	if (!blk_queue_secure_erase(q))
	42	return -EOPNOTSUPP;
	43	op = REQ_OP_SECURE_ERASE;
	44	} else {
	45	if (!blk_queue_discard(q))
	46	return -EOPNOTSUPP;
	47	op = REQ_OP_DISCARD;
	48	}
f31e7e40	49
28b2be20 DW	50	bs_mask = (bdev_logical_block_size(bdev) >> 9) - 1;
	51	if ((sector \| nr_sects) & bs_mask)
	52	return -EINVAL;
	53
ba5d7385 ML	54	if (!nr_sects)
ba5d7385 ML	55	return -EINVAL;
a22c4d7e	56
ba5d7385	57	while (nr_sects) {
4800bf7b	58	sector_t req_sects = min_t(sector_t, nr_sects,
ba5d7385	59	bio_allowed_max_sectors(q));
c6e66634	60
4800bf7b DC	61	WARN_ON_ONCE((req_sects << 9) > UINT_MAX);
4800bf7b DC	62
a2d6b3a2	63	bio = blk_next_bio(bio, 0, gfp_mask);
4f024f37	64	bio->bi_iter.bi_sector = sector;
74d46992	65	bio_set_dev(bio, bdev);
288dab8a	66	bio_set_op_attrs(bio, op, 0);
f31e7e40	67
4f024f37	68	bio->bi_iter.bi_size = req_sects << 9;
ba5d7385	69	sector += req_sects;
c6e66634	70	nr_sects -= req_sects;
f31e7e40	71
c8123f8c JA	72	/*
	73	* We can loop for a long time in here, if someone does
	74	* full device discards (like mkfs). Be nice and allow
	75	* us to schedule out to avoid softlocking if preempt
	76	* is disabled.
	77	*/
	78	cond_resched();
5dba3089	79	}
38f25255 CH	80
	81	*biop = bio;
	82	return 0;
	83	}
	84	EXPORT_SYMBOL(__blkdev_issue_discard);
	85
	86	/**
	87	* blkdev_issue_discard - queue a discard
	88	* @bdev: blockdev to issue discard for
	89	* @sector: start sector
	90	* @nr_sects: number of sectors to discard
	91	* @gfp_mask: memory allocation flags (for bio_alloc)
e554911c	92	* @flags: BLKDEV_DISCARD_* flags to control behaviour
38f25255 CH	93	*
	94	* Description:
	95	* Issue a discard request for the sectors in question.
	96	*/
	97	int blkdev_issue_discard(struct block_device *bdev, sector_t sector,
	98	sector_t nr_sects, gfp_t gfp_mask, unsigned long flags)
	99	{
38f25255 CH	100	struct bio *bio = NULL;
	101	struct blk_plug plug;
	102	int ret;
	103
38f25255	104	blk_start_plug(&plug);
288dab8a	105	ret = __blkdev_issue_discard(bdev, sector, nr_sects, gfp_mask, flags,
38f25255	106	&bio);
bbd848e0	107	if (!ret && bio) {
4e49ea4a	108	ret = submit_bio_wait(bio);
48920ff2	109	if (ret == -EOPNOTSUPP)
bbd848e0	110	ret = 0;
05bd92dd	111	bio_put(bio);
bbd848e0	112	}
0cfbcafc	113	blk_finish_plug(&plug);
f31e7e40	114
bbd848e0	115	return ret;
f31e7e40 DM	116	}
f31e7e40 DM	117	EXPORT_SYMBOL(blkdev_issue_discard);
3f14d792	118
4363ac7c	119	/**
e73c23ff	120	* __blkdev_issue_write_same - generate number of bios with same page
4363ac7c MP	121	* @bdev: target blockdev
	122	* @sector: start sector
	123	* @nr_sects: number of sectors to write
	124	* @gfp_mask: memory allocation flags (for bio_alloc)
	125	* @page: page containing data to write
e73c23ff	126	* @biop: pointer to anchor bio
4363ac7c MP	127	*
4363ac7c MP	128	* Description:
e73c23ff	129	* Generate and issue number of bios(REQ_OP_WRITE_SAME) with same page.
4363ac7c	130	*/
e73c23ff CK	131	static int __blkdev_issue_write_same(struct block_device *bdev, sector_t sector,
	132	sector_t nr_sects, gfp_t gfp_mask, struct page *page,
	133	struct bio **biop)
4363ac7c	134	{
4363ac7c MP	135	struct request_queue *q = bdev_get_queue(bdev);
4363ac7c MP	136	unsigned int max_write_same_sectors;
e73c23ff	137	struct bio bio = biop;
28b2be20	138	sector_t bs_mask;
4363ac7c MP	139
	140	if (!q)
	141	return -ENXIO;
	142
a13553c7 ID	143	if (bdev_read_only(bdev))
	144	return -EPERM;
	145
28b2be20 DW	146	bs_mask = (bdev_logical_block_size(bdev) >> 9) - 1;
	147	if ((sector \| nr_sects) & bs_mask)
	148	return -EINVAL;
	149
e73c23ff CK	150	if (!bdev_write_same(bdev))
	151	return -EOPNOTSUPP;
	152
b49a0871	153	/* Ensure that max_write_same_sectors doesn't overflow bi_size */
34ffec60	154	max_write_same_sectors = bio_allowed_max_sectors(q);
4363ac7c	155
4363ac7c	156	while (nr_sects) {
a2d6b3a2	157	bio = blk_next_bio(bio, 1, gfp_mask);
4f024f37	158	bio->bi_iter.bi_sector = sector;
74d46992	159	bio_set_dev(bio, bdev);
4363ac7c MP	160	bio->bi_vcnt = 1;
	161	bio->bi_io_vec->bv_page = page;
	162	bio->bi_io_vec->bv_offset = 0;
	163	bio->bi_io_vec->bv_len = bdev_logical_block_size(bdev);
95fe6c1a	164	bio_set_op_attrs(bio, REQ_OP_WRITE_SAME, 0);
4363ac7c MP	165
4363ac7c MP	166	if (nr_sects > max_write_same_sectors) {
4f024f37	167	bio->bi_iter.bi_size = max_write_same_sectors << 9;
4363ac7c MP	168	nr_sects -= max_write_same_sectors;
	169	sector += max_write_same_sectors;
	170	} else {
4f024f37	171	bio->bi_iter.bi_size = nr_sects << 9;
4363ac7c MP	172	nr_sects = 0;
4363ac7c MP	173	}
e73c23ff	174	cond_resched();
4363ac7c MP	175	}
4363ac7c MP	176
e73c23ff CK	177	*biop = bio;
	178	return 0;
	179	}
	180
	181	/**
	182	* blkdev_issue_write_same - queue a write same operation
	183	* @bdev: target blockdev
	184	* @sector: start sector
	185	* @nr_sects: number of sectors to write
	186	* @gfp_mask: memory allocation flags (for bio_alloc)
	187	* @page: page containing data
	188	*
	189	* Description:
	190	* Issue a write same request for the sectors in question.
	191	*/
	192	int blkdev_issue_write_same(struct block_device *bdev, sector_t sector,
	193	sector_t nr_sects, gfp_t gfp_mask,
	194	struct page *page)
	195	{
	196	struct bio *bio = NULL;
	197	struct blk_plug plug;
	198	int ret;
	199
	200	blk_start_plug(&plug);
	201	ret = __blkdev_issue_write_same(bdev, sector, nr_sects, gfp_mask, page,
	202	&bio);
	203	if (ret == 0 && bio) {
4e49ea4a	204	ret = submit_bio_wait(bio);
05bd92dd ST	205	bio_put(bio);
05bd92dd ST	206	}
e73c23ff	207	blk_finish_plug(&plug);
3f40bf2c	208	return ret;
4363ac7c MP	209	}
	210	EXPORT_SYMBOL(blkdev_issue_write_same);
	211
a6f0788e CK	212	static int __blkdev_issue_write_zeroes(struct block_device *bdev,
a6f0788e CK	213	sector_t sector, sector_t nr_sects, gfp_t gfp_mask,
d928be9f	214	struct bio **biop, unsigned flags)
a6f0788e CK	215	{
	216	struct bio bio = biop;
	217	unsigned int max_write_zeroes_sectors;
	218	struct request_queue *q = bdev_get_queue(bdev);
	219
	220	if (!q)
	221	return -ENXIO;
	222
a13553c7 ID	223	if (bdev_read_only(bdev))
	224	return -EPERM;
	225
a6f0788e CK	226	/* Ensure that max_write_zeroes_sectors doesn't overflow bi_size */
	227	max_write_zeroes_sectors = bdev_write_zeroes_sectors(bdev);
	228
	229	if (max_write_zeroes_sectors == 0)
	230	return -EOPNOTSUPP;
	231
	232	while (nr_sects) {
a2d6b3a2	233	bio = blk_next_bio(bio, 0, gfp_mask);
a6f0788e	234	bio->bi_iter.bi_sector = sector;
74d46992	235	bio_set_dev(bio, bdev);
d928be9f CH	236	bio->bi_opf = REQ_OP_WRITE_ZEROES;
	237	if (flags & BLKDEV_ZERO_NOUNMAP)
	238	bio->bi_opf \|= REQ_NOUNMAP;
a6f0788e CK	239
	240	if (nr_sects > max_write_zeroes_sectors) {
	241	bio->bi_iter.bi_size = max_write_zeroes_sectors << 9;
	242	nr_sects -= max_write_zeroes_sectors;
	243	sector += max_write_zeroes_sectors;
	244	} else {
	245	bio->bi_iter.bi_size = nr_sects << 9;
	246	nr_sects = 0;
	247	}
	248	cond_resched();
	249	}
	250
	251	*biop = bio;
	252	return 0;
	253	}
	254
615d22a5 DLM	255	/*
	256	* Convert a number of 512B sectors to a number of pages.
	257	* The result is limited to a number of pages that can fit into a BIO.
	258	* Also make sure that the result is always at least 1 (page) for the cases
	259	* where nr_sects is lower than the number of sectors in a page.
	260	*/
	261	static unsigned int __blkdev_sectors_to_bio_pages(sector_t nr_sects)
	262	{
09c2c359	263	sector_t pages = DIV_ROUND_UP_SECTOR_T(nr_sects, PAGE_SIZE / 512);
615d22a5	264
09c2c359	265	return min(pages, (sector_t)BIO_MAX_PAGES);
615d22a5 DLM	266	}
615d22a5 DLM	267
425a4dba ID	268	static int __blkdev_issue_zero_pages(struct block_device *bdev,
	269	sector_t sector, sector_t nr_sects, gfp_t gfp_mask,
	270	struct bio **biop)
	271	{
	272	struct request_queue *q = bdev_get_queue(bdev);
	273	struct bio bio = biop;
	274	int bi_size = 0;
	275	unsigned int sz;
	276
	277	if (!q)
	278	return -ENXIO;
	279
a13553c7 ID	280	if (bdev_read_only(bdev))
	281	return -EPERM;
	282
425a4dba	283	while (nr_sects != 0) {
a2d6b3a2 DLM	284	bio = blk_next_bio(bio, __blkdev_sectors_to_bio_pages(nr_sects),
a2d6b3a2 DLM	285	gfp_mask);
425a4dba ID	286	bio->bi_iter.bi_sector = sector;
	287	bio_set_dev(bio, bdev);
	288	bio_set_op_attrs(bio, REQ_OP_WRITE, 0);
	289
	290	while (nr_sects != 0) {
	291	sz = min((sector_t) PAGE_SIZE, nr_sects << 9);
	292	bi_size = bio_add_page(bio, ZERO_PAGE(0), sz, 0);
	293	nr_sects -= bi_size >> 9;
	294	sector += bi_size >> 9;
	295	if (bi_size < sz)
	296	break;
	297	}
	298	cond_resched();
	299	}
	300
	301	*biop = bio;
	302	return 0;
	303	}
	304
3f14d792	305	/**
e73c23ff	306	* __blkdev_issue_zeroout - generate number of zero filed write bios
3f14d792 DM	307	* @bdev: blockdev to issue
	308	* @sector: start sector
	309	* @nr_sects: number of sectors to write
	310	* @gfp_mask: memory allocation flags (for bio_alloc)
e73c23ff	311	* @biop: pointer to anchor bio
ee472d83	312	* @flags: controls detailed behavior
3f14d792 DM	313	*
3f14d792 DM	314	* Description:
ee472d83 CH	315	* Zero-fill a block range, either using hardware offload or by explicitly
	316	* writing zeroes to the device.
	317	*
	318	* If a device is using logical block provisioning, the underlying space will
	319	* not be released if %flags contains BLKDEV_ZERO_NOUNMAP.
cb365b96 CH	320	*
	321	* If %flags contains BLKDEV_ZERO_NOFALLBACK, the function will return
	322	* -EOPNOTSUPP if no explicit hardware offload for zeroing is provided.
3f14d792	323	*/
e73c23ff CK	324	int __blkdev_issue_zeroout(struct block_device *bdev, sector_t sector,
e73c23ff CK	325	sector_t nr_sects, gfp_t gfp_mask, struct bio **biop,
ee472d83	326	unsigned flags)
3f14d792	327	{
18edc8ea	328	int ret;
28b2be20 DW	329	sector_t bs_mask;
	330
	331	bs_mask = (bdev_logical_block_size(bdev) >> 9) - 1;
	332	if ((sector \| nr_sects) & bs_mask)
	333	return -EINVAL;
3f14d792	334
a6f0788e	335	ret = __blkdev_issue_write_zeroes(bdev, sector, nr_sects, gfp_mask,
d928be9f	336	biop, flags);
cb365b96	337	if (ret != -EOPNOTSUPP \|\| (flags & BLKDEV_ZERO_NOFALLBACK))
425a4dba	338	return ret;
3f14d792	339
425a4dba ID	340	return __blkdev_issue_zero_pages(bdev, sector, nr_sects, gfp_mask,
425a4dba ID	341	biop);
3f14d792	342	}
e73c23ff	343	EXPORT_SYMBOL(__blkdev_issue_zeroout);
579e8f3c MP	344
	345	/**
	346	* blkdev_issue_zeroout - zero-fill a block range
	347	* @bdev: blockdev to write
	348	* @sector: start sector
	349	* @nr_sects: number of sectors to write
	350	* @gfp_mask: memory allocation flags (for bio_alloc)
ee472d83	351	* @flags: controls detailed behavior
579e8f3c MP	352	*
579e8f3c MP	353	* Description:
ee472d83 CH	354	* Zero-fill a block range, either using hardware offload or by explicitly
	355	* writing zeroes to the device. See __blkdev_issue_zeroout() for the
	356	* valid values for %flags.
579e8f3c	357	*/
579e8f3c	358	int blkdev_issue_zeroout(struct block_device *bdev, sector_t sector,
ee472d83	359	sector_t nr_sects, gfp_t gfp_mask, unsigned flags)
579e8f3c	360	{
d5ce4c31 ID	361	int ret = 0;
	362	sector_t bs_mask;
	363	struct bio *bio;
e73c23ff	364	struct blk_plug plug;
d5ce4c31	365	bool try_write_zeroes = !!bdev_write_zeroes_sectors(bdev);
d93ba7a5	366
d5ce4c31 ID	367	bs_mask = (bdev_logical_block_size(bdev) >> 9) - 1;
	368	if ((sector \| nr_sects) & bs_mask)
	369	return -EINVAL;
	370
	371	retry:
	372	bio = NULL;
e73c23ff	373	blk_start_plug(&plug);
d5ce4c31 ID	374	if (try_write_zeroes) {
	375	ret = __blkdev_issue_write_zeroes(bdev, sector, nr_sects,
	376	gfp_mask, &bio, flags);
	377	} else if (!(flags & BLKDEV_ZERO_NOFALLBACK)) {
	378	ret = __blkdev_issue_zero_pages(bdev, sector, nr_sects,
	379	gfp_mask, &bio);
	380	} else {
	381	/* No zeroing offload support */
	382	ret = -EOPNOTSUPP;
	383	}
e73c23ff CK	384	if (ret == 0 && bio) {
	385	ret = submit_bio_wait(bio);
	386	bio_put(bio);
	387	}
	388	blk_finish_plug(&plug);
d5ce4c31 ID	389	if (ret && try_write_zeroes) {
	390	if (!(flags & BLKDEV_ZERO_NOFALLBACK)) {
	391	try_write_zeroes = false;
	392	goto retry;
	393	}
	394	if (!bdev_write_zeroes_sectors(bdev)) {
	395	/*
	396	* Zeroing offload support was indicated, but the
	397	* device reported ILLEGAL REQUEST (for some devices
	398	* there is no non-destructive way to verify whether
	399	* WRITE ZEROES is actually supported).
	400	*/
	401	ret = -EOPNOTSUPP;
	402	}
	403	}
579e8f3c	404
e73c23ff	405	return ret;
579e8f3c	406	}
3f14d792	407	EXPORT_SYMBOL(blkdev_issue_zeroout);