[linux-2.6-block.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;

	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 = 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;
		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 = 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
5dba3089	54	while (nr_sects) {
744889b7 ML	55	unsigned int req_sects = nr_sects;
744889b7 ML	56	sector_t end_sect;
c6e66634	57
b88aef36 MP	58	if (!req_sects)
b88aef36 MP	59	goto fail;
af097f5d JA	60	if (req_sects > UINT_MAX >> 9)
af097f5d JA	61	req_sects = UINT_MAX >> 9;
a22c4d7e	62
c6e66634	63	end_sect = sector + req_sects;
c6e66634	64
a2d6b3a2	65	bio = blk_next_bio(bio, 0, gfp_mask);
4f024f37	66	bio->bi_iter.bi_sector = sector;
74d46992	67	bio_set_dev(bio, bdev);
288dab8a	68	bio_set_op_attrs(bio, op, 0);
f31e7e40	69
4f024f37	70	bio->bi_iter.bi_size = req_sects << 9;
c6e66634 PB	71	nr_sects -= req_sects;
c6e66634 PB	72	sector = end_sect;
f31e7e40	73
c8123f8c JA	74	/*
	75	* We can loop for a long time in here, if someone does
	76	* full device discards (like mkfs). Be nice and allow
	77	* us to schedule out to avoid softlocking if preempt
	78	* is disabled.
	79	*/
	80	cond_resched();
5dba3089	81	}
38f25255 CH	82
	83	*biop = bio;
	84	return 0;
b88aef36 MP	85
	86	fail:
	87	if (bio) {
	88	submit_bio_wait(bio);
	89	bio_put(bio);
	90	}
	91	*biop = NULL;
	92	return -EOPNOTSUPP;
38f25255 CH	93	}
	94	EXPORT_SYMBOL(__blkdev_issue_discard);
	95
	96	/**
	97	* blkdev_issue_discard - queue a discard
	98	* @bdev: blockdev to issue discard for
	99	* @sector: start sector
	100	* @nr_sects: number of sectors to discard
	101	* @gfp_mask: memory allocation flags (for bio_alloc)
e554911c	102	* @flags: BLKDEV_DISCARD_* flags to control behaviour
38f25255 CH	103	*
	104	* Description:
	105	* Issue a discard request for the sectors in question.
	106	*/
	107	int blkdev_issue_discard(struct block_device *bdev, sector_t sector,
	108	sector_t nr_sects, gfp_t gfp_mask, unsigned long flags)
	109	{
38f25255 CH	110	struct bio *bio = NULL;
	111	struct blk_plug plug;
	112	int ret;
	113
38f25255	114	blk_start_plug(&plug);
288dab8a	115	ret = __blkdev_issue_discard(bdev, sector, nr_sects, gfp_mask, flags,
38f25255	116	&bio);
bbd848e0	117	if (!ret && bio) {
4e49ea4a	118	ret = submit_bio_wait(bio);
48920ff2	119	if (ret == -EOPNOTSUPP)
bbd848e0	120	ret = 0;
05bd92dd	121	bio_put(bio);
bbd848e0	122	}
0cfbcafc	123	blk_finish_plug(&plug);
f31e7e40	124
bbd848e0	125	return ret;
f31e7e40 DM	126	}
f31e7e40 DM	127	EXPORT_SYMBOL(blkdev_issue_discard);
3f14d792	128
4363ac7c	129	/**
e73c23ff	130	* __blkdev_issue_write_same - generate number of bios with same page
4363ac7c MP	131	* @bdev: target blockdev
	132	* @sector: start sector
	133	* @nr_sects: number of sectors to write
	134	* @gfp_mask: memory allocation flags (for bio_alloc)
	135	* @page: page containing data to write
e73c23ff	136	* @biop: pointer to anchor bio
4363ac7c MP	137	*
4363ac7c MP	138	* Description:
e73c23ff	139	* Generate and issue number of bios(REQ_OP_WRITE_SAME) with same page.
4363ac7c	140	*/
e73c23ff CK	141	static int __blkdev_issue_write_same(struct block_device *bdev, sector_t sector,
	142	sector_t nr_sects, gfp_t gfp_mask, struct page *page,
	143	struct bio **biop)
4363ac7c	144	{
4363ac7c MP	145	struct request_queue *q = bdev_get_queue(bdev);
4363ac7c MP	146	unsigned int max_write_same_sectors;
e73c23ff	147	struct bio bio = biop;
28b2be20	148	sector_t bs_mask;
4363ac7c MP	149
	150	if (!q)
	151	return -ENXIO;
	152
a13553c7 ID	153	if (bdev_read_only(bdev))
	154	return -EPERM;
	155
28b2be20 DW	156	bs_mask = (bdev_logical_block_size(bdev) >> 9) - 1;
	157	if ((sector \| nr_sects) & bs_mask)
	158	return -EINVAL;
	159
e73c23ff CK	160	if (!bdev_write_same(bdev))
	161	return -EOPNOTSUPP;
	162
b49a0871 ML	163	/* Ensure that max_write_same_sectors doesn't overflow bi_size */
b49a0871 ML	164	max_write_same_sectors = UINT_MAX >> 9;
4363ac7c	165
4363ac7c	166	while (nr_sects) {
a2d6b3a2	167	bio = blk_next_bio(bio, 1, gfp_mask);
4f024f37	168	bio->bi_iter.bi_sector = sector;
74d46992	169	bio_set_dev(bio, bdev);
4363ac7c MP	170	bio->bi_vcnt = 1;
	171	bio->bi_io_vec->bv_page = page;
	172	bio->bi_io_vec->bv_offset = 0;
	173	bio->bi_io_vec->bv_len = bdev_logical_block_size(bdev);
95fe6c1a	174	bio_set_op_attrs(bio, REQ_OP_WRITE_SAME, 0);
4363ac7c MP	175
4363ac7c MP	176	if (nr_sects > max_write_same_sectors) {
4f024f37	177	bio->bi_iter.bi_size = max_write_same_sectors << 9;
4363ac7c MP	178	nr_sects -= max_write_same_sectors;
	179	sector += max_write_same_sectors;
	180	} else {
4f024f37	181	bio->bi_iter.bi_size = nr_sects << 9;
4363ac7c MP	182	nr_sects = 0;
4363ac7c MP	183	}
e73c23ff	184	cond_resched();
4363ac7c MP	185	}
4363ac7c MP	186
e73c23ff CK	187	*biop = bio;
	188	return 0;
	189	}
	190
	191	/**
	192	* blkdev_issue_write_same - queue a write same operation
	193	* @bdev: target blockdev
	194	* @sector: start sector
	195	* @nr_sects: number of sectors to write
	196	* @gfp_mask: memory allocation flags (for bio_alloc)
	197	* @page: page containing data
	198	*
	199	* Description:
	200	* Issue a write same request for the sectors in question.
	201	*/
	202	int blkdev_issue_write_same(struct block_device *bdev, sector_t sector,
	203	sector_t nr_sects, gfp_t gfp_mask,
	204	struct page *page)
	205	{
	206	struct bio *bio = NULL;
	207	struct blk_plug plug;
	208	int ret;
	209
	210	blk_start_plug(&plug);
	211	ret = __blkdev_issue_write_same(bdev, sector, nr_sects, gfp_mask, page,
	212	&bio);
	213	if (ret == 0 && bio) {
4e49ea4a	214	ret = submit_bio_wait(bio);
05bd92dd ST	215	bio_put(bio);
05bd92dd ST	216	}
e73c23ff	217	blk_finish_plug(&plug);
3f40bf2c	218	return ret;
4363ac7c MP	219	}
	220	EXPORT_SYMBOL(blkdev_issue_write_same);
	221
a6f0788e CK	222	static int __blkdev_issue_write_zeroes(struct block_device *bdev,
a6f0788e CK	223	sector_t sector, sector_t nr_sects, gfp_t gfp_mask,
d928be9f	224	struct bio **biop, unsigned flags)
a6f0788e CK	225	{
	226	struct bio bio = biop;
	227	unsigned int max_write_zeroes_sectors;
	228	struct request_queue *q = bdev_get_queue(bdev);
	229
	230	if (!q)
	231	return -ENXIO;
	232
a13553c7 ID	233	if (bdev_read_only(bdev))
	234	return -EPERM;
	235
a6f0788e CK	236	/* Ensure that max_write_zeroes_sectors doesn't overflow bi_size */
	237	max_write_zeroes_sectors = bdev_write_zeroes_sectors(bdev);
	238
	239	if (max_write_zeroes_sectors == 0)
	240	return -EOPNOTSUPP;
	241
	242	while (nr_sects) {
a2d6b3a2	243	bio = blk_next_bio(bio, 0, gfp_mask);
a6f0788e	244	bio->bi_iter.bi_sector = sector;
74d46992	245	bio_set_dev(bio, bdev);
d928be9f CH	246	bio->bi_opf = REQ_OP_WRITE_ZEROES;
	247	if (flags & BLKDEV_ZERO_NOUNMAP)
	248	bio->bi_opf \|= REQ_NOUNMAP;
a6f0788e CK	249
	250	if (nr_sects > max_write_zeroes_sectors) {
	251	bio->bi_iter.bi_size = max_write_zeroes_sectors << 9;
	252	nr_sects -= max_write_zeroes_sectors;
	253	sector += max_write_zeroes_sectors;
	254	} else {
	255	bio->bi_iter.bi_size = nr_sects << 9;
	256	nr_sects = 0;
	257	}
	258	cond_resched();
	259	}
	260
	261	*biop = bio;
	262	return 0;
	263	}
	264
615d22a5 DLM	265	/*
	266	* Convert a number of 512B sectors to a number of pages.
	267	* The result is limited to a number of pages that can fit into a BIO.
	268	* Also make sure that the result is always at least 1 (page) for the cases
	269	* where nr_sects is lower than the number of sectors in a page.
	270	*/
	271	static unsigned int __blkdev_sectors_to_bio_pages(sector_t nr_sects)
	272	{
09c2c359	273	sector_t pages = DIV_ROUND_UP_SECTOR_T(nr_sects, PAGE_SIZE / 512);
615d22a5	274
09c2c359	275	return min(pages, (sector_t)BIO_MAX_PAGES);
615d22a5 DLM	276	}
615d22a5 DLM	277
425a4dba ID	278	static int __blkdev_issue_zero_pages(struct block_device *bdev,
	279	sector_t sector, sector_t nr_sects, gfp_t gfp_mask,
	280	struct bio **biop)
	281	{
	282	struct request_queue *q = bdev_get_queue(bdev);
	283	struct bio bio = biop;
	284	int bi_size = 0;
	285	unsigned int sz;
	286
	287	if (!q)
	288	return -ENXIO;
	289
a13553c7 ID	290	if (bdev_read_only(bdev))
	291	return -EPERM;
	292
425a4dba	293	while (nr_sects != 0) {
a2d6b3a2 DLM	294	bio = blk_next_bio(bio, __blkdev_sectors_to_bio_pages(nr_sects),
a2d6b3a2 DLM	295	gfp_mask);
425a4dba ID	296	bio->bi_iter.bi_sector = sector;
	297	bio_set_dev(bio, bdev);
	298	bio_set_op_attrs(bio, REQ_OP_WRITE, 0);
	299
	300	while (nr_sects != 0) {
	301	sz = min((sector_t) PAGE_SIZE, nr_sects << 9);
	302	bi_size = bio_add_page(bio, ZERO_PAGE(0), sz, 0);
	303	nr_sects -= bi_size >> 9;
	304	sector += bi_size >> 9;
	305	if (bi_size < sz)
	306	break;
	307	}
	308	cond_resched();
	309	}
	310
	311	*biop = bio;
	312	return 0;
	313	}
	314
3f14d792	315	/**
e73c23ff	316	* __blkdev_issue_zeroout - generate number of zero filed write bios
3f14d792 DM	317	* @bdev: blockdev to issue
	318	* @sector: start sector
	319	* @nr_sects: number of sectors to write
	320	* @gfp_mask: memory allocation flags (for bio_alloc)
e73c23ff	321	* @biop: pointer to anchor bio
ee472d83	322	* @flags: controls detailed behavior
3f14d792 DM	323	*
3f14d792 DM	324	* Description:
ee472d83 CH	325	* Zero-fill a block range, either using hardware offload or by explicitly
	326	* writing zeroes to the device.
	327	*
	328	* If a device is using logical block provisioning, the underlying space will
	329	* not be released if %flags contains BLKDEV_ZERO_NOUNMAP.
cb365b96 CH	330	*
	331	* If %flags contains BLKDEV_ZERO_NOFALLBACK, the function will return
	332	* -EOPNOTSUPP if no explicit hardware offload for zeroing is provided.
3f14d792	333	*/
e73c23ff CK	334	int __blkdev_issue_zeroout(struct block_device *bdev, sector_t sector,
e73c23ff CK	335	sector_t nr_sects, gfp_t gfp_mask, struct bio **biop,
ee472d83	336	unsigned flags)
3f14d792	337	{
18edc8ea	338	int ret;
28b2be20 DW	339	sector_t bs_mask;
	340
	341	bs_mask = (bdev_logical_block_size(bdev) >> 9) - 1;
	342	if ((sector \| nr_sects) & bs_mask)
	343	return -EINVAL;
3f14d792	344
a6f0788e	345	ret = __blkdev_issue_write_zeroes(bdev, sector, nr_sects, gfp_mask,
d928be9f	346	biop, flags);
cb365b96	347	if (ret != -EOPNOTSUPP \|\| (flags & BLKDEV_ZERO_NOFALLBACK))
425a4dba	348	return ret;
3f14d792	349
425a4dba ID	350	return __blkdev_issue_zero_pages(bdev, sector, nr_sects, gfp_mask,
425a4dba ID	351	biop);
3f14d792	352	}
e73c23ff	353	EXPORT_SYMBOL(__blkdev_issue_zeroout);
579e8f3c MP	354
	355	/**
	356	* blkdev_issue_zeroout - zero-fill a block range
	357	* @bdev: blockdev to write
	358	* @sector: start sector
	359	* @nr_sects: number of sectors to write
	360	* @gfp_mask: memory allocation flags (for bio_alloc)
ee472d83	361	* @flags: controls detailed behavior
579e8f3c MP	362	*
579e8f3c MP	363	* Description:
ee472d83 CH	364	* Zero-fill a block range, either using hardware offload or by explicitly
	365	* writing zeroes to the device. See __blkdev_issue_zeroout() for the
	366	* valid values for %flags.
579e8f3c	367	*/
579e8f3c	368	int blkdev_issue_zeroout(struct block_device *bdev, sector_t sector,
ee472d83	369	sector_t nr_sects, gfp_t gfp_mask, unsigned flags)
579e8f3c	370	{
d5ce4c31 ID	371	int ret = 0;
	372	sector_t bs_mask;
	373	struct bio *bio;
e73c23ff	374	struct blk_plug plug;
d5ce4c31	375	bool try_write_zeroes = !!bdev_write_zeroes_sectors(bdev);
d93ba7a5	376
d5ce4c31 ID	377	bs_mask = (bdev_logical_block_size(bdev) >> 9) - 1;
	378	if ((sector \| nr_sects) & bs_mask)
	379	return -EINVAL;
	380
	381	retry:
	382	bio = NULL;
e73c23ff	383	blk_start_plug(&plug);
d5ce4c31 ID	384	if (try_write_zeroes) {
	385	ret = __blkdev_issue_write_zeroes(bdev, sector, nr_sects,
	386	gfp_mask, &bio, flags);
	387	} else if (!(flags & BLKDEV_ZERO_NOFALLBACK)) {
	388	ret = __blkdev_issue_zero_pages(bdev, sector, nr_sects,
	389	gfp_mask, &bio);
	390	} else {
	391	/* No zeroing offload support */
	392	ret = -EOPNOTSUPP;
	393	}
e73c23ff CK	394	if (ret == 0 && bio) {
	395	ret = submit_bio_wait(bio);
	396	bio_put(bio);
	397	}
	398	blk_finish_plug(&plug);
d5ce4c31 ID	399	if (ret && try_write_zeroes) {
	400	if (!(flags & BLKDEV_ZERO_NOFALLBACK)) {
	401	try_write_zeroes = false;
	402	goto retry;
	403	}
	404	if (!bdev_write_zeroes_sectors(bdev)) {
	405	/*
	406	* Zeroing offload support was indicated, but the
	407	* device reported ILLEGAL REQUEST (for some devices
	408	* there is no non-destructive way to verify whether
	409	* WRITE ZEROES is actually supported).
	410	*/
	411	ret = -EOPNOTSUPP;
	412	}
	413	}
579e8f3c	414
e73c23ff	415	return ret;
579e8f3c	416	}
3f14d792	417	EXPORT_SYMBOL(blkdev_issue_zeroout);