[mirror_ubuntu-hirsute-kernel.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"

static struct bio *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;

	while (nr_sects) {
		unsigned int req_sects = nr_sects;
		sector_t end_sect;

		if (!req_sects)
			goto fail;
		if (req_sects > UINT_MAX >> 9)
			req_sects = UINT_MAX >> 9;

		end_sect = sector + req_sects;

		bio = 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;
		nr_sects -= req_sects;
		sector = end_sect;

		/*
		 * 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;

fail:
	if (bio) {
		submit_bio_wait(bio);
		bio_put(bio);
	}
	*biop = NULL;
	return -EOPNOTSUPP;
}
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 = UINT_MAX >> 9;

	while (nr_sects) {
		bio = 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 = 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 = 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
4e49ea4a	13	static struct bio next_bio(struct bio bio, unsigned int nr_pages,
9082e87b	14	gfp_t gfp)
f31e7e40	15	{
9082e87b CH	16	struct bio *new = bio_alloc(gfp, nr_pages);
	17
	18	if (bio) {
	19	bio_chain(bio, new);
4e49ea4a	20	submit_bio(bio);
9082e87b	21	}
5dba3089	22
9082e87b	23	return new;
f31e7e40 DM	24	}
f31e7e40 DM	25
38f25255	26	int __blkdev_issue_discard(struct block_device *bdev, sector_t sector,
288dab8a	27	sector_t nr_sects, gfp_t gfp_mask, int flags,
469e3216	28	struct bio **biop)
f31e7e40	29	{
f31e7e40	30	struct request_queue *q = bdev_get_queue(bdev);
38f25255	31	struct bio bio = biop;
ef295ecf	32	unsigned int op;
28b2be20	33	sector_t bs_mask;
f31e7e40 DM	34
	35	if (!q)
	36	return -ENXIO;
288dab8a	37
a13553c7 ID	38	if (bdev_read_only(bdev))
	39	return -EPERM;
	40
288dab8a CH	41	if (flags & BLKDEV_DISCARD_SECURE) {
	42	if (!blk_queue_secure_erase(q))
	43	return -EOPNOTSUPP;
	44	op = REQ_OP_SECURE_ERASE;
	45	} else {
	46	if (!blk_queue_discard(q))
	47	return -EOPNOTSUPP;
	48	op = REQ_OP_DISCARD;
	49	}
f31e7e40	50
28b2be20 DW	51	bs_mask = (bdev_logical_block_size(bdev) >> 9) - 1;
	52	if ((sector \| nr_sects) & bs_mask)
	53	return -EINVAL;
	54
5dba3089	55	while (nr_sects) {
744889b7 ML	56	unsigned int req_sects = nr_sects;
744889b7 ML	57	sector_t end_sect;
c6e66634	58
b88aef36 MP	59	if (!req_sects)
b88aef36 MP	60	goto fail;
af097f5d JA	61	if (req_sects > UINT_MAX >> 9)
af097f5d JA	62	req_sects = UINT_MAX >> 9;
a22c4d7e	63
c6e66634	64	end_sect = sector + req_sects;
c6e66634	65
f9d03f96	66	bio = next_bio(bio, 0, gfp_mask);
4f024f37	67	bio->bi_iter.bi_sector = sector;
74d46992	68	bio_set_dev(bio, bdev);
288dab8a	69	bio_set_op_attrs(bio, op, 0);
f31e7e40	70
4f024f37	71	bio->bi_iter.bi_size = req_sects << 9;
c6e66634 PB	72	nr_sects -= req_sects;
c6e66634 PB	73	sector = end_sect;
f31e7e40	74
c8123f8c JA	75	/*
	76	* We can loop for a long time in here, if someone does
	77	* full device discards (like mkfs). Be nice and allow
	78	* us to schedule out to avoid softlocking if preempt
	79	* is disabled.
	80	*/
	81	cond_resched();
5dba3089	82	}
38f25255 CH	83
	84	*biop = bio;
	85	return 0;
b88aef36 MP	86
	87	fail:
	88	if (bio) {
	89	submit_bio_wait(bio);
	90	bio_put(bio);
	91	}
	92	*biop = NULL;
	93	return -EOPNOTSUPP;
38f25255 CH	94	}
	95	EXPORT_SYMBOL(__blkdev_issue_discard);
	96
	97	/**
	98	* blkdev_issue_discard - queue a discard
	99	* @bdev: blockdev to issue discard for
	100	* @sector: start sector
	101	* @nr_sects: number of sectors to discard
	102	* @gfp_mask: memory allocation flags (for bio_alloc)
e554911c	103	* @flags: BLKDEV_DISCARD_* flags to control behaviour
38f25255 CH	104	*
	105	* Description:
	106	* Issue a discard request for the sectors in question.
	107	*/
	108	int blkdev_issue_discard(struct block_device *bdev, sector_t sector,
	109	sector_t nr_sects, gfp_t gfp_mask, unsigned long flags)
	110	{
38f25255 CH	111	struct bio *bio = NULL;
	112	struct blk_plug plug;
	113	int ret;
	114
38f25255	115	blk_start_plug(&plug);
288dab8a	116	ret = __blkdev_issue_discard(bdev, sector, nr_sects, gfp_mask, flags,
38f25255	117	&bio);
bbd848e0	118	if (!ret && bio) {
4e49ea4a	119	ret = submit_bio_wait(bio);
48920ff2	120	if (ret == -EOPNOTSUPP)
bbd848e0	121	ret = 0;
05bd92dd	122	bio_put(bio);
bbd848e0	123	}
0cfbcafc	124	blk_finish_plug(&plug);
f31e7e40	125
bbd848e0	126	return ret;
f31e7e40 DM	127	}
f31e7e40 DM	128	EXPORT_SYMBOL(blkdev_issue_discard);
3f14d792	129
4363ac7c	130	/**
e73c23ff	131	* __blkdev_issue_write_same - generate number of bios with same page
4363ac7c MP	132	* @bdev: target blockdev
	133	* @sector: start sector
	134	* @nr_sects: number of sectors to write
	135	* @gfp_mask: memory allocation flags (for bio_alloc)
	136	* @page: page containing data to write
e73c23ff	137	* @biop: pointer to anchor bio
4363ac7c MP	138	*
4363ac7c MP	139	* Description:
e73c23ff	140	* Generate and issue number of bios(REQ_OP_WRITE_SAME) with same page.
4363ac7c	141	*/
e73c23ff CK	142	static int __blkdev_issue_write_same(struct block_device *bdev, sector_t sector,
	143	sector_t nr_sects, gfp_t gfp_mask, struct page *page,
	144	struct bio **biop)
4363ac7c	145	{
4363ac7c MP	146	struct request_queue *q = bdev_get_queue(bdev);
4363ac7c MP	147	unsigned int max_write_same_sectors;
e73c23ff	148	struct bio bio = biop;
28b2be20	149	sector_t bs_mask;
4363ac7c MP	150
	151	if (!q)
	152	return -ENXIO;
	153
a13553c7 ID	154	if (bdev_read_only(bdev))
	155	return -EPERM;
	156
28b2be20 DW	157	bs_mask = (bdev_logical_block_size(bdev) >> 9) - 1;
	158	if ((sector \| nr_sects) & bs_mask)
	159	return -EINVAL;
	160
e73c23ff CK	161	if (!bdev_write_same(bdev))
	162	return -EOPNOTSUPP;
	163
b49a0871 ML	164	/* Ensure that max_write_same_sectors doesn't overflow bi_size */
b49a0871 ML	165	max_write_same_sectors = UINT_MAX >> 9;
4363ac7c	166
4363ac7c	167	while (nr_sects) {
4e49ea4a	168	bio = next_bio(bio, 1, gfp_mask);
4f024f37	169	bio->bi_iter.bi_sector = sector;
74d46992	170	bio_set_dev(bio, bdev);
4363ac7c MP	171	bio->bi_vcnt = 1;
	172	bio->bi_io_vec->bv_page = page;
	173	bio->bi_io_vec->bv_offset = 0;
	174	bio->bi_io_vec->bv_len = bdev_logical_block_size(bdev);
95fe6c1a	175	bio_set_op_attrs(bio, REQ_OP_WRITE_SAME, 0);
4363ac7c MP	176
4363ac7c MP	177	if (nr_sects > max_write_same_sectors) {
4f024f37	178	bio->bi_iter.bi_size = max_write_same_sectors << 9;
4363ac7c MP	179	nr_sects -= max_write_same_sectors;
	180	sector += max_write_same_sectors;
	181	} else {
4f024f37	182	bio->bi_iter.bi_size = nr_sects << 9;
4363ac7c MP	183	nr_sects = 0;
4363ac7c MP	184	}
e73c23ff	185	cond_resched();
4363ac7c MP	186	}
4363ac7c MP	187
e73c23ff CK	188	*biop = bio;
	189	return 0;
	190	}
	191
	192	/**
	193	* blkdev_issue_write_same - queue a write same operation
	194	* @bdev: target blockdev
	195	* @sector: start sector
	196	* @nr_sects: number of sectors to write
	197	* @gfp_mask: memory allocation flags (for bio_alloc)
	198	* @page: page containing data
	199	*
	200	* Description:
	201	* Issue a write same request for the sectors in question.
	202	*/
	203	int blkdev_issue_write_same(struct block_device *bdev, sector_t sector,
	204	sector_t nr_sects, gfp_t gfp_mask,
	205	struct page *page)
	206	{
	207	struct bio *bio = NULL;
	208	struct blk_plug plug;
	209	int ret;
	210
	211	blk_start_plug(&plug);
	212	ret = __blkdev_issue_write_same(bdev, sector, nr_sects, gfp_mask, page,
	213	&bio);
	214	if (ret == 0 && bio) {
4e49ea4a	215	ret = submit_bio_wait(bio);
05bd92dd ST	216	bio_put(bio);
05bd92dd ST	217	}
e73c23ff	218	blk_finish_plug(&plug);
3f40bf2c	219	return ret;
4363ac7c MP	220	}
	221	EXPORT_SYMBOL(blkdev_issue_write_same);
	222
a6f0788e CK	223	static int __blkdev_issue_write_zeroes(struct block_device *bdev,
a6f0788e CK	224	sector_t sector, sector_t nr_sects, gfp_t gfp_mask,
d928be9f	225	struct bio **biop, unsigned flags)
a6f0788e CK	226	{
	227	struct bio bio = biop;
	228	unsigned int max_write_zeroes_sectors;
	229	struct request_queue *q = bdev_get_queue(bdev);
	230
	231	if (!q)
	232	return -ENXIO;
	233
a13553c7 ID	234	if (bdev_read_only(bdev))
	235	return -EPERM;
	236
a6f0788e CK	237	/* Ensure that max_write_zeroes_sectors doesn't overflow bi_size */
	238	max_write_zeroes_sectors = bdev_write_zeroes_sectors(bdev);
	239
	240	if (max_write_zeroes_sectors == 0)
	241	return -EOPNOTSUPP;
	242
	243	while (nr_sects) {
	244	bio = next_bio(bio, 0, gfp_mask);
	245	bio->bi_iter.bi_sector = sector;
74d46992	246	bio_set_dev(bio, bdev);
d928be9f CH	247	bio->bi_opf = REQ_OP_WRITE_ZEROES;
	248	if (flags & BLKDEV_ZERO_NOUNMAP)
	249	bio->bi_opf \|= REQ_NOUNMAP;
a6f0788e CK	250
	251	if (nr_sects > max_write_zeroes_sectors) {
	252	bio->bi_iter.bi_size = max_write_zeroes_sectors << 9;
	253	nr_sects -= max_write_zeroes_sectors;
	254	sector += max_write_zeroes_sectors;
	255	} else {
	256	bio->bi_iter.bi_size = nr_sects << 9;
	257	nr_sects = 0;
	258	}
	259	cond_resched();
	260	}
	261
	262	*biop = bio;
	263	return 0;
	264	}
	265
615d22a5 DLM	266	/*
	267	* Convert a number of 512B sectors to a number of pages.
	268	* The result is limited to a number of pages that can fit into a BIO.
	269	* Also make sure that the result is always at least 1 (page) for the cases
	270	* where nr_sects is lower than the number of sectors in a page.
	271	*/
	272	static unsigned int __blkdev_sectors_to_bio_pages(sector_t nr_sects)
	273	{
09c2c359	274	sector_t pages = DIV_ROUND_UP_SECTOR_T(nr_sects, PAGE_SIZE / 512);
615d22a5	275
09c2c359	276	return min(pages, (sector_t)BIO_MAX_PAGES);
615d22a5 DLM	277	}
615d22a5 DLM	278
425a4dba ID	279	static int __blkdev_issue_zero_pages(struct block_device *bdev,
	280	sector_t sector, sector_t nr_sects, gfp_t gfp_mask,
	281	struct bio **biop)
	282	{
	283	struct request_queue *q = bdev_get_queue(bdev);
	284	struct bio bio = biop;
	285	int bi_size = 0;
	286	unsigned int sz;
	287
	288	if (!q)
	289	return -ENXIO;
	290
a13553c7 ID	291	if (bdev_read_only(bdev))
	292	return -EPERM;
	293
425a4dba ID	294	while (nr_sects != 0) {
	295	bio = next_bio(bio, __blkdev_sectors_to_bio_pages(nr_sects),
	296	gfp_mask);
	297	bio->bi_iter.bi_sector = sector;
	298	bio_set_dev(bio, bdev);
	299	bio_set_op_attrs(bio, REQ_OP_WRITE, 0);
	300
	301	while (nr_sects != 0) {
	302	sz = min((sector_t) PAGE_SIZE, nr_sects << 9);
	303	bi_size = bio_add_page(bio, ZERO_PAGE(0), sz, 0);
	304	nr_sects -= bi_size >> 9;
	305	sector += bi_size >> 9;
	306	if (bi_size < sz)
	307	break;
	308	}
	309	cond_resched();
	310	}
	311
	312	*biop = bio;
	313	return 0;
	314	}
	315
3f14d792	316	/**
e73c23ff	317	* __blkdev_issue_zeroout - generate number of zero filed write bios
3f14d792 DM	318	* @bdev: blockdev to issue
	319	* @sector: start sector
	320	* @nr_sects: number of sectors to write
	321	* @gfp_mask: memory allocation flags (for bio_alloc)
e73c23ff	322	* @biop: pointer to anchor bio
ee472d83	323	* @flags: controls detailed behavior
3f14d792 DM	324	*
3f14d792 DM	325	* Description:
ee472d83 CH	326	* Zero-fill a block range, either using hardware offload or by explicitly
	327	* writing zeroes to the device.
	328	*
	329	* If a device is using logical block provisioning, the underlying space will
	330	* not be released if %flags contains BLKDEV_ZERO_NOUNMAP.
cb365b96 CH	331	*
	332	* If %flags contains BLKDEV_ZERO_NOFALLBACK, the function will return
	333	* -EOPNOTSUPP if no explicit hardware offload for zeroing is provided.
3f14d792	334	*/
e73c23ff CK	335	int __blkdev_issue_zeroout(struct block_device *bdev, sector_t sector,
e73c23ff CK	336	sector_t nr_sects, gfp_t gfp_mask, struct bio **biop,
ee472d83	337	unsigned flags)
3f14d792	338	{
18edc8ea	339	int ret;
28b2be20 DW	340	sector_t bs_mask;
	341
	342	bs_mask = (bdev_logical_block_size(bdev) >> 9) - 1;
	343	if ((sector \| nr_sects) & bs_mask)
	344	return -EINVAL;
3f14d792	345
a6f0788e	346	ret = __blkdev_issue_write_zeroes(bdev, sector, nr_sects, gfp_mask,
d928be9f	347	biop, flags);
cb365b96	348	if (ret != -EOPNOTSUPP \|\| (flags & BLKDEV_ZERO_NOFALLBACK))
425a4dba	349	return ret;
3f14d792	350
425a4dba ID	351	return __blkdev_issue_zero_pages(bdev, sector, nr_sects, gfp_mask,
425a4dba ID	352	biop);
3f14d792	353	}
e73c23ff	354	EXPORT_SYMBOL(__blkdev_issue_zeroout);
579e8f3c MP	355
	356	/**
	357	* blkdev_issue_zeroout - zero-fill a block range
	358	* @bdev: blockdev to write
	359	* @sector: start sector
	360	* @nr_sects: number of sectors to write
	361	* @gfp_mask: memory allocation flags (for bio_alloc)
ee472d83	362	* @flags: controls detailed behavior
579e8f3c MP	363	*
579e8f3c MP	364	* Description:
ee472d83 CH	365	* Zero-fill a block range, either using hardware offload or by explicitly
	366	* writing zeroes to the device. See __blkdev_issue_zeroout() for the
	367	* valid values for %flags.
579e8f3c	368	*/
579e8f3c	369	int blkdev_issue_zeroout(struct block_device *bdev, sector_t sector,
ee472d83	370	sector_t nr_sects, gfp_t gfp_mask, unsigned flags)
579e8f3c	371	{
d5ce4c31 ID	372	int ret = 0;
	373	sector_t bs_mask;
	374	struct bio *bio;
e73c23ff	375	struct blk_plug plug;
d5ce4c31	376	bool try_write_zeroes = !!bdev_write_zeroes_sectors(bdev);
d93ba7a5	377
d5ce4c31 ID	378	bs_mask = (bdev_logical_block_size(bdev) >> 9) - 1;
	379	if ((sector \| nr_sects) & bs_mask)
	380	return -EINVAL;
	381
	382	retry:
	383	bio = NULL;
e73c23ff	384	blk_start_plug(&plug);
d5ce4c31 ID	385	if (try_write_zeroes) {
	386	ret = __blkdev_issue_write_zeroes(bdev, sector, nr_sects,
	387	gfp_mask, &bio, flags);
	388	} else if (!(flags & BLKDEV_ZERO_NOFALLBACK)) {
	389	ret = __blkdev_issue_zero_pages(bdev, sector, nr_sects,
	390	gfp_mask, &bio);
	391	} else {
	392	/* No zeroing offload support */
	393	ret = -EOPNOTSUPP;
	394	}
e73c23ff CK	395	if (ret == 0 && bio) {
	396	ret = submit_bio_wait(bio);
	397	bio_put(bio);
	398	}
	399	blk_finish_plug(&plug);
d5ce4c31 ID	400	if (ret && try_write_zeroes) {
	401	if (!(flags & BLKDEV_ZERO_NOFALLBACK)) {
	402	try_write_zeroes = false;
	403	goto retry;
	404	}
	405	if (!bdev_write_zeroes_sectors(bdev)) {
	406	/*
	407	* Zeroing offload support was indicated, but the
	408	* device reported ILLEGAL REQUEST (for some devices
	409	* there is no non-destructive way to verify whether
	410	* WRITE ZEROES is actually supported).
	411	*/
	412	ret = -EOPNOTSUPP;
	413	}
	414	}
579e8f3c	415
e73c23ff	416	return ret;
579e8f3c	417	}
3f14d792	418	EXPORT_SYMBOL(blkdev_issue_zeroout);