[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"

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 granularity;
	unsigned int op;
	int alignment;
	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;

	/* Zero-sector (unknown) and one-sector granularities are the same.  */
	granularity = max(q->limits.discard_granularity >> 9, 1U);
	alignment = (bdev_discard_alignment(bdev) >> 9) % granularity;

	while (nr_sects) {
		unsigned int req_sects;
		sector_t end_sect, tmp;

		/* Make sure bi_size doesn't overflow */
		req_sects = min_t(sector_t, nr_sects, UINT_MAX >> 9);

		/**
		 * If splitting a request, and the next starting sector would be
		 * misaligned, stop the discard at the previous aligned sector.
		 */
		end_sect = sector + req_sects;
		tmp = end_sect;
		if (req_sects < nr_sects &&
		    sector_div(tmp, granularity) != alignment) {
			end_sect = end_sect - alignment;
			sector_div(end_sect, granularity);
			end_sect = end_sect * granularity + alignment;
			req_sects = end_sect - sector;
		}

		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;
}
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;
a22c4d7e	32	unsigned int granularity;
ef295ecf	33	unsigned int op;
a22c4d7e	34	int alignment;
28b2be20	35	sector_t bs_mask;
f31e7e40 DM	36
	37	if (!q)
	38	return -ENXIO;
288dab8a	39
a13553c7 ID	40	if (bdev_read_only(bdev))
	41	return -EPERM;
	42
288dab8a CH	43	if (flags & BLKDEV_DISCARD_SECURE) {
	44	if (!blk_queue_secure_erase(q))
	45	return -EOPNOTSUPP;
	46	op = REQ_OP_SECURE_ERASE;
	47	} else {
	48	if (!blk_queue_discard(q))
	49	return -EOPNOTSUPP;
	50	op = REQ_OP_DISCARD;
	51	}
f31e7e40	52
28b2be20 DW	53	bs_mask = (bdev_logical_block_size(bdev) >> 9) - 1;
	54	if ((sector \| nr_sects) & bs_mask)
	55	return -EINVAL;
	56
a22c4d7e ML	57	/* Zero-sector (unknown) and one-sector granularities are the same. */
	58	granularity = max(q->limits.discard_granularity >> 9, 1U);
	59	alignment = (bdev_discard_alignment(bdev) >> 9) % granularity;
	60
5dba3089	61	while (nr_sects) {
c6e66634	62	unsigned int req_sects;
a22c4d7e	63	sector_t end_sect, tmp;
c6e66634	64
a22c4d7e ML	65	/* Make sure bi_size doesn't overflow */
	66	req_sects = min_t(sector_t, nr_sects, UINT_MAX >> 9);
	67
9082e87b	68	/**
a22c4d7e ML	69	* If splitting a request, and the next starting sector would be
	70	* misaligned, stop the discard at the previous aligned sector.
	71	*/
c6e66634	72	end_sect = sector + req_sects;
a22c4d7e ML	73	tmp = end_sect;
	74	if (req_sects < nr_sects &&
	75	sector_div(tmp, granularity) != alignment) {
	76	end_sect = end_sect - alignment;
	77	sector_div(end_sect, granularity);
	78	end_sect = end_sect * granularity + alignment;
	79	req_sects = end_sect - sector;
	80	}
c6e66634	81
f9d03f96	82	bio = next_bio(bio, 0, gfp_mask);
4f024f37	83	bio->bi_iter.bi_sector = sector;
74d46992	84	bio_set_dev(bio, bdev);
288dab8a	85	bio_set_op_attrs(bio, op, 0);
f31e7e40	86
4f024f37	87	bio->bi_iter.bi_size = req_sects << 9;
c6e66634 PB	88	nr_sects -= req_sects;
c6e66634 PB	89	sector = end_sect;
f31e7e40	90
c8123f8c JA	91	/*
	92	* We can loop for a long time in here, if someone does
	93	* full device discards (like mkfs). Be nice and allow
	94	* us to schedule out to avoid softlocking if preempt
	95	* is disabled.
	96	*/
	97	cond_resched();
5dba3089	98	}
38f25255 CH	99
	100	*biop = bio;
	101	return 0;
	102	}
	103	EXPORT_SYMBOL(__blkdev_issue_discard);
	104
	105	/**
	106	* blkdev_issue_discard - queue a discard
	107	* @bdev: blockdev to issue discard for
	108	* @sector: start sector
	109	* @nr_sects: number of sectors to discard
	110	* @gfp_mask: memory allocation flags (for bio_alloc)
e554911c	111	* @flags: BLKDEV_DISCARD_* flags to control behaviour
38f25255 CH	112	*
	113	* Description:
	114	* Issue a discard request for the sectors in question.
	115	*/
	116	int blkdev_issue_discard(struct block_device *bdev, sector_t sector,
	117	sector_t nr_sects, gfp_t gfp_mask, unsigned long flags)
	118	{
38f25255 CH	119	struct bio *bio = NULL;
	120	struct blk_plug plug;
	121	int ret;
	122
38f25255	123	blk_start_plug(&plug);
288dab8a	124	ret = __blkdev_issue_discard(bdev, sector, nr_sects, gfp_mask, flags,
38f25255	125	&bio);
bbd848e0	126	if (!ret && bio) {
4e49ea4a	127	ret = submit_bio_wait(bio);
48920ff2	128	if (ret == -EOPNOTSUPP)
bbd848e0	129	ret = 0;
05bd92dd	130	bio_put(bio);
bbd848e0	131	}
0cfbcafc	132	blk_finish_plug(&plug);
f31e7e40	133
bbd848e0	134	return ret;
f31e7e40 DM	135	}
f31e7e40 DM	136	EXPORT_SYMBOL(blkdev_issue_discard);
3f14d792	137
4363ac7c	138	/**
e73c23ff	139	* __blkdev_issue_write_same - generate number of bios with same page
4363ac7c MP	140	* @bdev: target blockdev
	141	* @sector: start sector
	142	* @nr_sects: number of sectors to write
	143	* @gfp_mask: memory allocation flags (for bio_alloc)
	144	* @page: page containing data to write
e73c23ff	145	* @biop: pointer to anchor bio
4363ac7c MP	146	*
4363ac7c MP	147	* Description:
e73c23ff	148	* Generate and issue number of bios(REQ_OP_WRITE_SAME) with same page.
4363ac7c	149	*/
e73c23ff CK	150	static int __blkdev_issue_write_same(struct block_device *bdev, sector_t sector,
	151	sector_t nr_sects, gfp_t gfp_mask, struct page *page,
	152	struct bio **biop)
4363ac7c	153	{
4363ac7c MP	154	struct request_queue *q = bdev_get_queue(bdev);
4363ac7c MP	155	unsigned int max_write_same_sectors;
e73c23ff	156	struct bio bio = biop;
28b2be20	157	sector_t bs_mask;
4363ac7c MP	158
	159	if (!q)
	160	return -ENXIO;
	161
a13553c7 ID	162	if (bdev_read_only(bdev))
	163	return -EPERM;
	164
28b2be20 DW	165	bs_mask = (bdev_logical_block_size(bdev) >> 9) - 1;
	166	if ((sector \| nr_sects) & bs_mask)
	167	return -EINVAL;
	168
e73c23ff CK	169	if (!bdev_write_same(bdev))
	170	return -EOPNOTSUPP;
	171
b49a0871 ML	172	/* Ensure that max_write_same_sectors doesn't overflow bi_size */
b49a0871 ML	173	max_write_same_sectors = UINT_MAX >> 9;
4363ac7c	174
4363ac7c	175	while (nr_sects) {
4e49ea4a	176	bio = next_bio(bio, 1, gfp_mask);
4f024f37	177	bio->bi_iter.bi_sector = sector;
74d46992	178	bio_set_dev(bio, bdev);
4363ac7c MP	179	bio->bi_vcnt = 1;
	180	bio->bi_io_vec->bv_page = page;
	181	bio->bi_io_vec->bv_offset = 0;
	182	bio->bi_io_vec->bv_len = bdev_logical_block_size(bdev);
95fe6c1a	183	bio_set_op_attrs(bio, REQ_OP_WRITE_SAME, 0);
4363ac7c MP	184
4363ac7c MP	185	if (nr_sects > max_write_same_sectors) {
4f024f37	186	bio->bi_iter.bi_size = max_write_same_sectors << 9;
4363ac7c MP	187	nr_sects -= max_write_same_sectors;
	188	sector += max_write_same_sectors;
	189	} else {
4f024f37	190	bio->bi_iter.bi_size = nr_sects << 9;
4363ac7c MP	191	nr_sects = 0;
4363ac7c MP	192	}
e73c23ff	193	cond_resched();
4363ac7c MP	194	}
4363ac7c MP	195
e73c23ff CK	196	*biop = bio;
	197	return 0;
	198	}
	199
	200	/**
	201	* blkdev_issue_write_same - queue a write same operation
	202	* @bdev: target blockdev
	203	* @sector: start sector
	204	* @nr_sects: number of sectors to write
	205	* @gfp_mask: memory allocation flags (for bio_alloc)
	206	* @page: page containing data
	207	*
	208	* Description:
	209	* Issue a write same request for the sectors in question.
	210	*/
	211	int blkdev_issue_write_same(struct block_device *bdev, sector_t sector,
	212	sector_t nr_sects, gfp_t gfp_mask,
	213	struct page *page)
	214	{
	215	struct bio *bio = NULL;
	216	struct blk_plug plug;
	217	int ret;
	218
	219	blk_start_plug(&plug);
	220	ret = __blkdev_issue_write_same(bdev, sector, nr_sects, gfp_mask, page,
	221	&bio);
	222	if (ret == 0 && bio) {
4e49ea4a	223	ret = submit_bio_wait(bio);
05bd92dd ST	224	bio_put(bio);
05bd92dd ST	225	}
e73c23ff	226	blk_finish_plug(&plug);
3f40bf2c	227	return ret;
4363ac7c MP	228	}
	229	EXPORT_SYMBOL(blkdev_issue_write_same);
	230
a6f0788e CK	231	static int __blkdev_issue_write_zeroes(struct block_device *bdev,
a6f0788e CK	232	sector_t sector, sector_t nr_sects, gfp_t gfp_mask,
d928be9f	233	struct bio **biop, unsigned flags)
a6f0788e CK	234	{
	235	struct bio bio = biop;
	236	unsigned int max_write_zeroes_sectors;
	237	struct request_queue *q = bdev_get_queue(bdev);
	238
	239	if (!q)
	240	return -ENXIO;
	241
a13553c7 ID	242	if (bdev_read_only(bdev))
	243	return -EPERM;
	244
a6f0788e CK	245	/* Ensure that max_write_zeroes_sectors doesn't overflow bi_size */
	246	max_write_zeroes_sectors = bdev_write_zeroes_sectors(bdev);
	247
	248	if (max_write_zeroes_sectors == 0)
	249	return -EOPNOTSUPP;
	250
	251	while (nr_sects) {
	252	bio = next_bio(bio, 0, gfp_mask);
	253	bio->bi_iter.bi_sector = sector;
74d46992	254	bio_set_dev(bio, bdev);
d928be9f CH	255	bio->bi_opf = REQ_OP_WRITE_ZEROES;
	256	if (flags & BLKDEV_ZERO_NOUNMAP)
	257	bio->bi_opf \|= REQ_NOUNMAP;
a6f0788e CK	258
	259	if (nr_sects > max_write_zeroes_sectors) {
	260	bio->bi_iter.bi_size = max_write_zeroes_sectors << 9;
	261	nr_sects -= max_write_zeroes_sectors;
	262	sector += max_write_zeroes_sectors;
	263	} else {
	264	bio->bi_iter.bi_size = nr_sects << 9;
	265	nr_sects = 0;
	266	}
	267	cond_resched();
	268	}
	269
	270	*biop = bio;
	271	return 0;
	272	}
	273
615d22a5 DLM	274	/*
	275	* Convert a number of 512B sectors to a number of pages.
	276	* The result is limited to a number of pages that can fit into a BIO.
	277	* Also make sure that the result is always at least 1 (page) for the cases
	278	* where nr_sects is lower than the number of sectors in a page.
	279	*/
	280	static unsigned int __blkdev_sectors_to_bio_pages(sector_t nr_sects)
	281	{
09c2c359	282	sector_t pages = DIV_ROUND_UP_SECTOR_T(nr_sects, PAGE_SIZE / 512);
615d22a5	283
09c2c359	284	return min(pages, (sector_t)BIO_MAX_PAGES);
615d22a5 DLM	285	}
615d22a5 DLM	286
425a4dba ID	287	static int __blkdev_issue_zero_pages(struct block_device *bdev,
	288	sector_t sector, sector_t nr_sects, gfp_t gfp_mask,
	289	struct bio **biop)
	290	{
	291	struct request_queue *q = bdev_get_queue(bdev);
	292	struct bio bio = biop;
	293	int bi_size = 0;
	294	unsigned int sz;
	295
	296	if (!q)
	297	return -ENXIO;
	298
a13553c7 ID	299	if (bdev_read_only(bdev))
	300	return -EPERM;
	301
425a4dba ID	302	while (nr_sects != 0) {
	303	bio = next_bio(bio, __blkdev_sectors_to_bio_pages(nr_sects),
	304	gfp_mask);
	305	bio->bi_iter.bi_sector = sector;
	306	bio_set_dev(bio, bdev);
	307	bio_set_op_attrs(bio, REQ_OP_WRITE, 0);
	308
	309	while (nr_sects != 0) {
	310	sz = min((sector_t) PAGE_SIZE, nr_sects << 9);
	311	bi_size = bio_add_page(bio, ZERO_PAGE(0), sz, 0);
	312	nr_sects -= bi_size >> 9;
	313	sector += bi_size >> 9;
	314	if (bi_size < sz)
	315	break;
	316	}
	317	cond_resched();
	318	}
	319
	320	*biop = bio;
	321	return 0;
	322	}
	323
3f14d792	324	/**
e73c23ff	325	* __blkdev_issue_zeroout - generate number of zero filed write bios
3f14d792 DM	326	* @bdev: blockdev to issue
	327	* @sector: start sector
	328	* @nr_sects: number of sectors to write
	329	* @gfp_mask: memory allocation flags (for bio_alloc)
e73c23ff	330	* @biop: pointer to anchor bio
ee472d83	331	* @flags: controls detailed behavior
3f14d792 DM	332	*
3f14d792 DM	333	* Description:
ee472d83 CH	334	* Zero-fill a block range, either using hardware offload or by explicitly
	335	* writing zeroes to the device.
	336	*
	337	* If a device is using logical block provisioning, the underlying space will
	338	* not be released if %flags contains BLKDEV_ZERO_NOUNMAP.
cb365b96 CH	339	*
	340	* If %flags contains BLKDEV_ZERO_NOFALLBACK, the function will return
	341	* -EOPNOTSUPP if no explicit hardware offload for zeroing is provided.
3f14d792	342	*/
e73c23ff CK	343	int __blkdev_issue_zeroout(struct block_device *bdev, sector_t sector,
e73c23ff CK	344	sector_t nr_sects, gfp_t gfp_mask, struct bio **biop,
ee472d83	345	unsigned flags)
3f14d792	346	{
18edc8ea	347	int ret;
28b2be20 DW	348	sector_t bs_mask;
	349
	350	bs_mask = (bdev_logical_block_size(bdev) >> 9) - 1;
	351	if ((sector \| nr_sects) & bs_mask)
	352	return -EINVAL;
3f14d792	353
a6f0788e	354	ret = __blkdev_issue_write_zeroes(bdev, sector, nr_sects, gfp_mask,
d928be9f	355	biop, flags);
cb365b96	356	if (ret != -EOPNOTSUPP \|\| (flags & BLKDEV_ZERO_NOFALLBACK))
425a4dba	357	return ret;
3f14d792	358
425a4dba ID	359	return __blkdev_issue_zero_pages(bdev, sector, nr_sects, gfp_mask,
425a4dba ID	360	biop);
3f14d792	361	}
e73c23ff	362	EXPORT_SYMBOL(__blkdev_issue_zeroout);
579e8f3c MP	363
	364	/**
	365	* blkdev_issue_zeroout - zero-fill a block range
	366	* @bdev: blockdev to write
	367	* @sector: start sector
	368	* @nr_sects: number of sectors to write
	369	* @gfp_mask: memory allocation flags (for bio_alloc)
ee472d83	370	* @flags: controls detailed behavior
579e8f3c MP	371	*
579e8f3c MP	372	* Description:
ee472d83 CH	373	* Zero-fill a block range, either using hardware offload or by explicitly
	374	* writing zeroes to the device. See __blkdev_issue_zeroout() for the
	375	* valid values for %flags.
579e8f3c	376	*/
579e8f3c	377	int blkdev_issue_zeroout(struct block_device *bdev, sector_t sector,
ee472d83	378	sector_t nr_sects, gfp_t gfp_mask, unsigned flags)
579e8f3c	379	{
d5ce4c31 ID	380	int ret = 0;
	381	sector_t bs_mask;
	382	struct bio *bio;
e73c23ff	383	struct blk_plug plug;
d5ce4c31	384	bool try_write_zeroes = !!bdev_write_zeroes_sectors(bdev);
d93ba7a5	385
d5ce4c31 ID	386	bs_mask = (bdev_logical_block_size(bdev) >> 9) - 1;
	387	if ((sector \| nr_sects) & bs_mask)
	388	return -EINVAL;
	389
	390	retry:
	391	bio = NULL;
e73c23ff	392	blk_start_plug(&plug);
d5ce4c31 ID	393	if (try_write_zeroes) {
	394	ret = __blkdev_issue_write_zeroes(bdev, sector, nr_sects,
	395	gfp_mask, &bio, flags);
	396	} else if (!(flags & BLKDEV_ZERO_NOFALLBACK)) {
	397	ret = __blkdev_issue_zero_pages(bdev, sector, nr_sects,
	398	gfp_mask, &bio);
	399	} else {
	400	/* No zeroing offload support */
	401	ret = -EOPNOTSUPP;
	402	}
e73c23ff CK	403	if (ret == 0 && bio) {
	404	ret = submit_bio_wait(bio);
	405	bio_put(bio);
	406	}
	407	blk_finish_plug(&plug);
d5ce4c31 ID	408	if (ret && try_write_zeroes) {
	409	if (!(flags & BLKDEV_ZERO_NOFALLBACK)) {
	410	try_write_zeroes = false;
	411	goto retry;
	412	}
	413	if (!bdev_write_zeroes_sectors(bdev)) {
	414	/*
	415	* Zeroing offload support was indicated, but the
	416	* device reported ILLEGAL REQUEST (for some devices
	417	* there is no non-destructive way to verify whether
	418	* WRITE ZEROES is actually supported).
	419	*/
	420	ret = -EOPNOTSUPP;
	421	}
	422	}
579e8f3c	423
e73c23ff	424	return ret;
579e8f3c	425	}
3f14d792	426	EXPORT_SYMBOL(blkdev_issue_zeroout);