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