[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;

		/*
		 * Issue in chunks of the user defined max discard setting,
		 * ensuring that bi_size doesn't overflow
		 */
		req_sects = min_t(sector_t, nr_sects,
					q->limits.max_discard_sectors);
		if (!req_sects)
			goto fail;
		if (req_sects > UINT_MAX >> 9)
			req_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;

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;
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
af097f5d JA	65	/*
	66	* Issue in chunks of the user defined max discard setting,
	67	* ensuring that bi_size doesn't overflow
	68	*/
	69	req_sects = min_t(sector_t, nr_sects,
	70	q->limits.max_discard_sectors);
b88aef36 MP	71	if (!req_sects)
b88aef36 MP	72	goto fail;
af097f5d JA	73	if (req_sects > UINT_MAX >> 9)
af097f5d JA	74	req_sects = UINT_MAX >> 9;
a22c4d7e	75
af097f5d	76	/*
a22c4d7e ML	77	* If splitting a request, and the next starting sector would be
	78	* misaligned, stop the discard at the previous aligned sector.
	79	*/
c6e66634	80	end_sect = sector + req_sects;
a22c4d7e ML	81	tmp = end_sect;
	82	if (req_sects < nr_sects &&
	83	sector_div(tmp, granularity) != alignment) {
	84	end_sect = end_sect - alignment;
	85	sector_div(end_sect, granularity);
	86	end_sect = end_sect * granularity + alignment;
	87	req_sects = end_sect - sector;
	88	}
c6e66634	89
f9d03f96	90	bio = next_bio(bio, 0, gfp_mask);
4f024f37	91	bio->bi_iter.bi_sector = sector;
74d46992	92	bio_set_dev(bio, bdev);
288dab8a	93	bio_set_op_attrs(bio, op, 0);
f31e7e40	94
4f024f37	95	bio->bi_iter.bi_size = req_sects << 9;
c6e66634 PB	96	nr_sects -= req_sects;
c6e66634 PB	97	sector = end_sect;
f31e7e40	98
c8123f8c JA	99	/*
	100	* We can loop for a long time in here, if someone does
	101	* full device discards (like mkfs). Be nice and allow
	102	* us to schedule out to avoid softlocking if preempt
	103	* is disabled.
	104	*/
	105	cond_resched();
5dba3089	106	}
38f25255 CH	107
	108	*biop = bio;
	109	return 0;
b88aef36 MP	110
	111	fail:
	112	if (bio) {
	113	submit_bio_wait(bio);
	114	bio_put(bio);
	115	}
	116	*biop = NULL;
	117	return -EOPNOTSUPP;
38f25255 CH	118	}
	119	EXPORT_SYMBOL(__blkdev_issue_discard);
	120
	121	/**
	122	* blkdev_issue_discard - queue a discard
	123	* @bdev: blockdev to issue discard for
	124	* @sector: start sector
	125	* @nr_sects: number of sectors to discard
	126	* @gfp_mask: memory allocation flags (for bio_alloc)
e554911c	127	* @flags: BLKDEV_DISCARD_* flags to control behaviour
38f25255 CH	128	*
	129	* Description:
	130	* Issue a discard request for the sectors in question.
	131	*/
	132	int blkdev_issue_discard(struct block_device *bdev, sector_t sector,
	133	sector_t nr_sects, gfp_t gfp_mask, unsigned long flags)
	134	{
38f25255 CH	135	struct bio *bio = NULL;
	136	struct blk_plug plug;
	137	int ret;
	138
38f25255	139	blk_start_plug(&plug);
288dab8a	140	ret = __blkdev_issue_discard(bdev, sector, nr_sects, gfp_mask, flags,
38f25255	141	&bio);
bbd848e0	142	if (!ret && bio) {
4e49ea4a	143	ret = submit_bio_wait(bio);
48920ff2	144	if (ret == -EOPNOTSUPP)
bbd848e0	145	ret = 0;
05bd92dd	146	bio_put(bio);
bbd848e0	147	}
0cfbcafc	148	blk_finish_plug(&plug);
f31e7e40	149
bbd848e0	150	return ret;
f31e7e40 DM	151	}
f31e7e40 DM	152	EXPORT_SYMBOL(blkdev_issue_discard);
3f14d792	153
4363ac7c	154	/**
e73c23ff	155	* __blkdev_issue_write_same - generate number of bios with same page
4363ac7c MP	156	* @bdev: target blockdev
	157	* @sector: start sector
	158	* @nr_sects: number of sectors to write
	159	* @gfp_mask: memory allocation flags (for bio_alloc)
	160	* @page: page containing data to write
e73c23ff	161	* @biop: pointer to anchor bio
4363ac7c MP	162	*
4363ac7c MP	163	* Description:
e73c23ff	164	* Generate and issue number of bios(REQ_OP_WRITE_SAME) with same page.
4363ac7c	165	*/
e73c23ff CK	166	static int __blkdev_issue_write_same(struct block_device *bdev, sector_t sector,
	167	sector_t nr_sects, gfp_t gfp_mask, struct page *page,
	168	struct bio **biop)
4363ac7c	169	{
4363ac7c MP	170	struct request_queue *q = bdev_get_queue(bdev);
4363ac7c MP	171	unsigned int max_write_same_sectors;
e73c23ff	172	struct bio bio = biop;
28b2be20	173	sector_t bs_mask;
4363ac7c MP	174
	175	if (!q)
	176	return -ENXIO;
	177
a13553c7 ID	178	if (bdev_read_only(bdev))
	179	return -EPERM;
	180
28b2be20 DW	181	bs_mask = (bdev_logical_block_size(bdev) >> 9) - 1;
	182	if ((sector \| nr_sects) & bs_mask)
	183	return -EINVAL;
	184
e73c23ff CK	185	if (!bdev_write_same(bdev))
	186	return -EOPNOTSUPP;
	187
b49a0871 ML	188	/* Ensure that max_write_same_sectors doesn't overflow bi_size */
b49a0871 ML	189	max_write_same_sectors = UINT_MAX >> 9;
4363ac7c	190
4363ac7c	191	while (nr_sects) {
4e49ea4a	192	bio = next_bio(bio, 1, gfp_mask);
4f024f37	193	bio->bi_iter.bi_sector = sector;
74d46992	194	bio_set_dev(bio, bdev);
4363ac7c MP	195	bio->bi_vcnt = 1;
	196	bio->bi_io_vec->bv_page = page;
	197	bio->bi_io_vec->bv_offset = 0;
	198	bio->bi_io_vec->bv_len = bdev_logical_block_size(bdev);
95fe6c1a	199	bio_set_op_attrs(bio, REQ_OP_WRITE_SAME, 0);
4363ac7c MP	200
4363ac7c MP	201	if (nr_sects > max_write_same_sectors) {
4f024f37	202	bio->bi_iter.bi_size = max_write_same_sectors << 9;
4363ac7c MP	203	nr_sects -= max_write_same_sectors;
	204	sector += max_write_same_sectors;
	205	} else {
4f024f37	206	bio->bi_iter.bi_size = nr_sects << 9;
4363ac7c MP	207	nr_sects = 0;
4363ac7c MP	208	}
e73c23ff	209	cond_resched();
4363ac7c MP	210	}
4363ac7c MP	211
e73c23ff CK	212	*biop = bio;
	213	return 0;
	214	}
	215
	216	/**
	217	* blkdev_issue_write_same - queue a write same operation
	218	* @bdev: target blockdev
	219	* @sector: start sector
	220	* @nr_sects: number of sectors to write
	221	* @gfp_mask: memory allocation flags (for bio_alloc)
	222	* @page: page containing data
	223	*
	224	* Description:
	225	* Issue a write same request for the sectors in question.
	226	*/
	227	int blkdev_issue_write_same(struct block_device *bdev, sector_t sector,
	228	sector_t nr_sects, gfp_t gfp_mask,
	229	struct page *page)
	230	{
	231	struct bio *bio = NULL;
	232	struct blk_plug plug;
	233	int ret;
	234
	235	blk_start_plug(&plug);
	236	ret = __blkdev_issue_write_same(bdev, sector, nr_sects, gfp_mask, page,
	237	&bio);
	238	if (ret == 0 && bio) {
4e49ea4a	239	ret = submit_bio_wait(bio);
05bd92dd ST	240	bio_put(bio);
05bd92dd ST	241	}
e73c23ff	242	blk_finish_plug(&plug);
3f40bf2c	243	return ret;
4363ac7c MP	244	}
	245	EXPORT_SYMBOL(blkdev_issue_write_same);
	246
a6f0788e CK	247	static int __blkdev_issue_write_zeroes(struct block_device *bdev,
a6f0788e CK	248	sector_t sector, sector_t nr_sects, gfp_t gfp_mask,
d928be9f	249	struct bio **biop, unsigned flags)
a6f0788e CK	250	{
	251	struct bio bio = biop;
	252	unsigned int max_write_zeroes_sectors;
	253	struct request_queue *q = bdev_get_queue(bdev);
	254
	255	if (!q)
	256	return -ENXIO;
	257
a13553c7 ID	258	if (bdev_read_only(bdev))
	259	return -EPERM;
	260
a6f0788e CK	261	/* Ensure that max_write_zeroes_sectors doesn't overflow bi_size */
	262	max_write_zeroes_sectors = bdev_write_zeroes_sectors(bdev);
	263
	264	if (max_write_zeroes_sectors == 0)
	265	return -EOPNOTSUPP;
	266
	267	while (nr_sects) {
	268	bio = next_bio(bio, 0, gfp_mask);
	269	bio->bi_iter.bi_sector = sector;
74d46992	270	bio_set_dev(bio, bdev);
d928be9f CH	271	bio->bi_opf = REQ_OP_WRITE_ZEROES;
	272	if (flags & BLKDEV_ZERO_NOUNMAP)
	273	bio->bi_opf \|= REQ_NOUNMAP;
a6f0788e CK	274
	275	if (nr_sects > max_write_zeroes_sectors) {
	276	bio->bi_iter.bi_size = max_write_zeroes_sectors << 9;
	277	nr_sects -= max_write_zeroes_sectors;
	278	sector += max_write_zeroes_sectors;
	279	} else {
	280	bio->bi_iter.bi_size = nr_sects << 9;
	281	nr_sects = 0;
	282	}
	283	cond_resched();
	284	}
	285
	286	*biop = bio;
	287	return 0;
	288	}
	289
615d22a5 DLM	290	/*
	291	* Convert a number of 512B sectors to a number of pages.
	292	* The result is limited to a number of pages that can fit into a BIO.
	293	* Also make sure that the result is always at least 1 (page) for the cases
	294	* where nr_sects is lower than the number of sectors in a page.
	295	*/
	296	static unsigned int __blkdev_sectors_to_bio_pages(sector_t nr_sects)
	297	{
09c2c359	298	sector_t pages = DIV_ROUND_UP_SECTOR_T(nr_sects, PAGE_SIZE / 512);
615d22a5	299
09c2c359	300	return min(pages, (sector_t)BIO_MAX_PAGES);
615d22a5 DLM	301	}
615d22a5 DLM	302
425a4dba ID	303	static int __blkdev_issue_zero_pages(struct block_device *bdev,
	304	sector_t sector, sector_t nr_sects, gfp_t gfp_mask,
	305	struct bio **biop)
	306	{
	307	struct request_queue *q = bdev_get_queue(bdev);
	308	struct bio bio = biop;
	309	int bi_size = 0;
	310	unsigned int sz;
	311
	312	if (!q)
	313	return -ENXIO;
	314
a13553c7 ID	315	if (bdev_read_only(bdev))
	316	return -EPERM;
	317
425a4dba ID	318	while (nr_sects != 0) {
	319	bio = next_bio(bio, __blkdev_sectors_to_bio_pages(nr_sects),
	320	gfp_mask);
	321	bio->bi_iter.bi_sector = sector;
	322	bio_set_dev(bio, bdev);
	323	bio_set_op_attrs(bio, REQ_OP_WRITE, 0);
	324
	325	while (nr_sects != 0) {
	326	sz = min((sector_t) PAGE_SIZE, nr_sects << 9);
	327	bi_size = bio_add_page(bio, ZERO_PAGE(0), sz, 0);
	328	nr_sects -= bi_size >> 9;
	329	sector += bi_size >> 9;
	330	if (bi_size < sz)
	331	break;
	332	}
	333	cond_resched();
	334	}
	335
	336	*biop = bio;
	337	return 0;
	338	}
	339
3f14d792	340	/**
e73c23ff	341	* __blkdev_issue_zeroout - generate number of zero filed write bios
3f14d792 DM	342	* @bdev: blockdev to issue
	343	* @sector: start sector
	344	* @nr_sects: number of sectors to write
	345	* @gfp_mask: memory allocation flags (for bio_alloc)
e73c23ff	346	* @biop: pointer to anchor bio
ee472d83	347	* @flags: controls detailed behavior
3f14d792 DM	348	*
3f14d792 DM	349	* Description:
ee472d83 CH	350	* Zero-fill a block range, either using hardware offload or by explicitly
	351	* writing zeroes to the device.
	352	*
	353	* If a device is using logical block provisioning, the underlying space will
	354	* not be released if %flags contains BLKDEV_ZERO_NOUNMAP.
cb365b96 CH	355	*
	356	* If %flags contains BLKDEV_ZERO_NOFALLBACK, the function will return
	357	* -EOPNOTSUPP if no explicit hardware offload for zeroing is provided.
3f14d792	358	*/
e73c23ff CK	359	int __blkdev_issue_zeroout(struct block_device *bdev, sector_t sector,
e73c23ff CK	360	sector_t nr_sects, gfp_t gfp_mask, struct bio **biop,
ee472d83	361	unsigned flags)
3f14d792	362	{
18edc8ea	363	int ret;
28b2be20 DW	364	sector_t bs_mask;
	365
	366	bs_mask = (bdev_logical_block_size(bdev) >> 9) - 1;
	367	if ((sector \| nr_sects) & bs_mask)
	368	return -EINVAL;
3f14d792	369
a6f0788e	370	ret = __blkdev_issue_write_zeroes(bdev, sector, nr_sects, gfp_mask,
d928be9f	371	biop, flags);
cb365b96	372	if (ret != -EOPNOTSUPP \|\| (flags & BLKDEV_ZERO_NOFALLBACK))
425a4dba	373	return ret;
3f14d792	374
425a4dba ID	375	return __blkdev_issue_zero_pages(bdev, sector, nr_sects, gfp_mask,
425a4dba ID	376	biop);
3f14d792	377	}
e73c23ff	378	EXPORT_SYMBOL(__blkdev_issue_zeroout);
579e8f3c MP	379
	380	/**
	381	* blkdev_issue_zeroout - zero-fill a block range
	382	* @bdev: blockdev to write
	383	* @sector: start sector
	384	* @nr_sects: number of sectors to write
	385	* @gfp_mask: memory allocation flags (for bio_alloc)
ee472d83	386	* @flags: controls detailed behavior
579e8f3c MP	387	*
579e8f3c MP	388	* Description:
ee472d83 CH	389	* Zero-fill a block range, either using hardware offload or by explicitly
	390	* writing zeroes to the device. See __blkdev_issue_zeroout() for the
	391	* valid values for %flags.
579e8f3c	392	*/
579e8f3c	393	int blkdev_issue_zeroout(struct block_device *bdev, sector_t sector,
ee472d83	394	sector_t nr_sects, gfp_t gfp_mask, unsigned flags)
579e8f3c	395	{
d5ce4c31 ID	396	int ret = 0;
	397	sector_t bs_mask;
	398	struct bio *bio;
e73c23ff	399	struct blk_plug plug;
d5ce4c31	400	bool try_write_zeroes = !!bdev_write_zeroes_sectors(bdev);
d93ba7a5	401
d5ce4c31 ID	402	bs_mask = (bdev_logical_block_size(bdev) >> 9) - 1;
	403	if ((sector \| nr_sects) & bs_mask)
	404	return -EINVAL;
	405
	406	retry:
	407	bio = NULL;
e73c23ff	408	blk_start_plug(&plug);
d5ce4c31 ID	409	if (try_write_zeroes) {
	410	ret = __blkdev_issue_write_zeroes(bdev, sector, nr_sects,
	411	gfp_mask, &bio, flags);
	412	} else if (!(flags & BLKDEV_ZERO_NOFALLBACK)) {
	413	ret = __blkdev_issue_zero_pages(bdev, sector, nr_sects,
	414	gfp_mask, &bio);
	415	} else {
	416	/* No zeroing offload support */
	417	ret = -EOPNOTSUPP;
	418	}
e73c23ff CK	419	if (ret == 0 && bio) {
	420	ret = submit_bio_wait(bio);
	421	bio_put(bio);
	422	}
	423	blk_finish_plug(&plug);
d5ce4c31 ID	424	if (ret && try_write_zeroes) {
	425	if (!(flags & BLKDEV_ZERO_NOFALLBACK)) {
	426	try_write_zeroes = false;
	427	goto retry;
	428	}
	429	if (!bdev_write_zeroes_sectors(bdev)) {
	430	/*
	431	* Zeroing offload support was indicated, but the
	432	* device reported ILLEGAL REQUEST (for some devices
	433	* there is no non-destructive way to verify whether
	434	* WRITE ZEROES is actually supported).
	435	*/
	436	ret = -EOPNOTSUPP;
	437	}
	438	}
579e8f3c	439
e73c23ff	440	return ret;
579e8f3c	441	}
3f14d792	442	EXPORT_SYMBOL(blkdev_issue_zeroout);