[mirror_ubuntu-zesty-kernel.git] / block / blk-lib.c

/*
 * 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 (flags & BLKDEV_DISCARD_SECURE) {
		if (flags & BLKDEV_DISCARD_ZERO)
			return -EOPNOTSUPP;
		if (!blk_queue_secure_erase(q))
			return -EOPNOTSUPP;
		op = REQ_OP_SECURE_ERASE;
	} else {
		if (!blk_queue_discard(q))
			return -EOPNOTSUPP;
		if ((flags & BLKDEV_DISCARD_ZERO) &&
		    !q->limits.discard_zeroes_data)
			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->bi_bdev = 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_IFL_* 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 && !(flags & BLKDEV_DISCARD_ZERO))
			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;

	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->bi_bdev = 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);

/**
 * __blkdev_issue_write_zeroes - generate number of bios with WRITE ZEROES
 * @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
 *
 * Description:
 *  Generate and issue number of bios(REQ_OP_WRITE_ZEROES) with zerofiled pages.
 */
static int __blkdev_issue_write_zeroes(struct block_device *bdev,
		sector_t sector, sector_t nr_sects, gfp_t gfp_mask,
		struct bio **biop)
{
	struct bio *bio = *biop;
	unsigned int max_write_zeroes_sectors;
	struct request_queue *q = bdev_get_queue(bdev);

	if (!q)
		return -ENXIO;

	/* 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->bi_bdev = bdev;
		bio_set_op_attrs(bio, REQ_OP_WRITE_ZEROES, 0);

		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;
}

/**
 * __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
 * @discard:	discard flag
 *
 * Description:
 *  Generate and issue number of bios with zerofiled pages.
 */
int __blkdev_issue_zeroout(struct block_device *bdev, sector_t sector,
		sector_t nr_sects, gfp_t gfp_mask, struct bio **biop,
		bool discard)
{
	int ret;
	int bi_size = 0;
	struct bio *bio = *biop;
	unsigned int sz;
	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);
	if (ret == 0 || (ret && ret != -EOPNOTSUPP))
		goto out;

	ret = 0;
	while (nr_sects != 0) {
		bio = next_bio(bio, min(nr_sects, (sector_t)BIO_MAX_PAGES),
				gfp_mask);
		bio->bi_iter.bi_sector = sector;
		bio->bi_bdev   = bdev;
		bio_set_op_attrs(bio, REQ_OP_WRITE, 0);

		while (nr_sects != 0) {
			sz = min((sector_t) PAGE_SIZE >> 9 , nr_sects);
			bi_size = bio_add_page(bio, ZERO_PAGE(0), sz << 9, 0);
			nr_sects -= bi_size >> 9;
			sector += bi_size >> 9;
			if (bi_size < (sz << 9))
				break;
		}
		cond_resched();
	}

	*biop = bio;
out:
	return ret;
}
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)
 * @discard:	whether to discard the block range
 *
 * Description:
 *  Zero-fill a block range.  If the discard flag is set and the block
 *  device guarantees that subsequent READ operations to the block range
 *  in question will return zeroes, the blocks will be discarded. Should
 *  the discard request fail, if the discard flag is not set, or if
 *  discard_zeroes_data is not supported, this function will resort to
 *  zeroing the blocks manually, thus provisioning (allocating,
 *  anchoring) them. If the block device supports WRITE ZEROES or WRITE SAME
 *  command(s), blkdev_issue_zeroout() will use it to optimize the process of
 *  clearing the block range. Otherwise the zeroing will be performed
 *  using regular WRITE calls.
 */
int blkdev_issue_zeroout(struct block_device *bdev, sector_t sector,
			 sector_t nr_sects, gfp_t gfp_mask, bool discard)
{
	int ret;
	struct bio *bio = NULL;
	struct blk_plug plug;

	if (discard) {
		if (!blkdev_issue_discard(bdev, sector, nr_sects, gfp_mask,
				BLKDEV_DISCARD_ZERO))
			return 0;
	}

	if (!blkdev_issue_write_same(bdev, sector, nr_sects, gfp_mask,
			ZERO_PAGE(0)))
		return 0;

	blk_start_plug(&plug);
	ret = __blkdev_issue_zeroout(bdev, sector, nr_sects, gfp_mask,
			&bio, discard);
	if (ret == 0 && bio) {
		ret = submit_bio_wait(bio);
		bio_put(bio);
	}
	blk_finish_plug(&plug);

	return ret;
}
EXPORT_SYMBOL(blkdev_issue_zeroout);
Commit	Line	Data
f31e7e40 DM	1	/*
	2	* Functions related to generic helpers functions
	3	*/
	4	#include <linux/kernel.h>
	5	#include <linux/module.h>
	6	#include <linux/bio.h>
	7	#include <linux/blkdev.h>
	8	#include <linux/scatterlist.h>
	9
	10	#include "blk.h"
	11
4e49ea4a	12	static struct bio next_bio(struct bio bio, unsigned int nr_pages,
9082e87b	13	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;
a22c4d7e	31	unsigned int granularity;
ef295ecf	32	unsigned int op;
a22c4d7e	33	int alignment;
28b2be20	34	sector_t bs_mask;
f31e7e40 DM	35
	36	if (!q)
	37	return -ENXIO;
288dab8a CH	38
288dab8a CH	39	if (flags & BLKDEV_DISCARD_SECURE) {
e950fdf7 CH	40	if (flags & BLKDEV_DISCARD_ZERO)
e950fdf7 CH	41	return -EOPNOTSUPP;
288dab8a CH	42	if (!blk_queue_secure_erase(q))
	43	return -EOPNOTSUPP;
	44	op = REQ_OP_SECURE_ERASE;
	45	} else {
	46	if (!blk_queue_discard(q))
	47	return -EOPNOTSUPP;
e950fdf7 CH	48	if ((flags & BLKDEV_DISCARD_ZERO) &&
	49	!q->limits.discard_zeroes_data)
	50	return -EOPNOTSUPP;
288dab8a CH	51	op = REQ_OP_DISCARD;
288dab8a CH	52	}
f31e7e40	53
28b2be20 DW	54	bs_mask = (bdev_logical_block_size(bdev) >> 9) - 1;
	55	if ((sector \| nr_sects) & bs_mask)
	56	return -EINVAL;
	57
a22c4d7e ML	58	/* Zero-sector (unknown) and one-sector granularities are the same. */
	59	granularity = max(q->limits.discard_granularity >> 9, 1U);
	60	alignment = (bdev_discard_alignment(bdev) >> 9) % granularity;
	61
5dba3089	62	while (nr_sects) {
c6e66634	63	unsigned int req_sects;
a22c4d7e	64	sector_t end_sect, tmp;
c6e66634	65
a22c4d7e ML	66	/* Make sure bi_size doesn't overflow */
	67	req_sects = min_t(sector_t, nr_sects, UINT_MAX >> 9);
	68
9082e87b	69	/**
a22c4d7e ML	70	* If splitting a request, and the next starting sector would be
	71	* misaligned, stop the discard at the previous aligned sector.
	72	*/
c6e66634	73	end_sect = sector + req_sects;
a22c4d7e ML	74	tmp = end_sect;
	75	if (req_sects < nr_sects &&
	76	sector_div(tmp, granularity) != alignment) {
	77	end_sect = end_sect - alignment;
	78	sector_div(end_sect, granularity);
	79	end_sect = end_sect * granularity + alignment;
	80	req_sects = end_sect - sector;
	81	}
c6e66634	82
f9d03f96	83	bio = next_bio(bio, 0, gfp_mask);
4f024f37	84	bio->bi_iter.bi_sector = sector;
f31e7e40	85	bio->bi_bdev = bdev;
288dab8a	86	bio_set_op_attrs(bio, op, 0);
f31e7e40	87
4f024f37	88	bio->bi_iter.bi_size = req_sects << 9;
c6e66634 PB	89	nr_sects -= req_sects;
c6e66634 PB	90	sector = end_sect;
f31e7e40	91
c8123f8c JA	92	/*
	93	* We can loop for a long time in here, if someone does
	94	* full device discards (like mkfs). Be nice and allow
	95	* us to schedule out to avoid softlocking if preempt
	96	* is disabled.
	97	*/
	98	cond_resched();
5dba3089	99	}
38f25255 CH	100
	101	*biop = bio;
	102	return 0;
	103	}
	104	EXPORT_SYMBOL(__blkdev_issue_discard);
	105
	106	/**
	107	* blkdev_issue_discard - queue a discard
	108	* @bdev: blockdev to issue discard for
	109	* @sector: start sector
	110	* @nr_sects: number of sectors to discard
	111	* @gfp_mask: memory allocation flags (for bio_alloc)
	112	* @flags: BLKDEV_IFL_* flags to control behaviour
	113	*
	114	* Description:
	115	* Issue a discard request for the sectors in question.
	116	*/
	117	int blkdev_issue_discard(struct block_device *bdev, sector_t sector,
	118	sector_t nr_sects, gfp_t gfp_mask, unsigned long flags)
	119	{
38f25255 CH	120	struct bio *bio = NULL;
	121	struct blk_plug plug;
	122	int ret;
	123
38f25255	124	blk_start_plug(&plug);
288dab8a	125	ret = __blkdev_issue_discard(bdev, sector, nr_sects, gfp_mask, flags,
38f25255	126	&bio);
bbd848e0	127	if (!ret && bio) {
4e49ea4a	128	ret = submit_bio_wait(bio);
e950fdf7	129	if (ret == -EOPNOTSUPP && !(flags & BLKDEV_DISCARD_ZERO))
bbd848e0	130	ret = 0;
05bd92dd	131	bio_put(bio);
bbd848e0	132	}
0cfbcafc	133	blk_finish_plug(&plug);
f31e7e40	134
bbd848e0	135	return ret;
f31e7e40 DM	136	}
f31e7e40 DM	137	EXPORT_SYMBOL(blkdev_issue_discard);
3f14d792	138
4363ac7c	139	/**
e73c23ff	140	* __blkdev_issue_write_same - generate number of bios with same page
4363ac7c MP	141	* @bdev: target blockdev
	142	* @sector: start sector
	143	* @nr_sects: number of sectors to write
	144	* @gfp_mask: memory allocation flags (for bio_alloc)
	145	* @page: page containing data to write
e73c23ff	146	* @biop: pointer to anchor bio
4363ac7c MP	147	*
4363ac7c MP	148	* Description:
e73c23ff	149	* Generate and issue number of bios(REQ_OP_WRITE_SAME) with same page.
4363ac7c	150	*/
e73c23ff CK	151	static int __blkdev_issue_write_same(struct block_device *bdev, sector_t sector,
	152	sector_t nr_sects, gfp_t gfp_mask, struct page *page,
	153	struct bio **biop)
4363ac7c	154	{
4363ac7c MP	155	struct request_queue *q = bdev_get_queue(bdev);
4363ac7c MP	156	unsigned int max_write_same_sectors;
e73c23ff	157	struct bio bio = biop;
28b2be20	158	sector_t bs_mask;
4363ac7c MP	159
	160	if (!q)
	161	return -ENXIO;
	162
28b2be20 DW	163	bs_mask = (bdev_logical_block_size(bdev) >> 9) - 1;
	164	if ((sector \| nr_sects) & bs_mask)
	165	return -EINVAL;
	166
e73c23ff CK	167	if (!bdev_write_same(bdev))
	168	return -EOPNOTSUPP;
	169
b49a0871 ML	170	/* Ensure that max_write_same_sectors doesn't overflow bi_size */
b49a0871 ML	171	max_write_same_sectors = UINT_MAX >> 9;
4363ac7c	172
4363ac7c	173	while (nr_sects) {
4e49ea4a	174	bio = next_bio(bio, 1, gfp_mask);
4f024f37	175	bio->bi_iter.bi_sector = sector;
4363ac7c	176	bio->bi_bdev = bdev;
4363ac7c MP	177	bio->bi_vcnt = 1;
	178	bio->bi_io_vec->bv_page = page;
	179	bio->bi_io_vec->bv_offset = 0;
	180	bio->bi_io_vec->bv_len = bdev_logical_block_size(bdev);
95fe6c1a	181	bio_set_op_attrs(bio, REQ_OP_WRITE_SAME, 0);
4363ac7c MP	182
4363ac7c MP	183	if (nr_sects > max_write_same_sectors) {
4f024f37	184	bio->bi_iter.bi_size = max_write_same_sectors << 9;
4363ac7c MP	185	nr_sects -= max_write_same_sectors;
	186	sector += max_write_same_sectors;
	187	} else {
4f024f37	188	bio->bi_iter.bi_size = nr_sects << 9;
4363ac7c MP	189	nr_sects = 0;
4363ac7c MP	190	}
e73c23ff	191	cond_resched();
4363ac7c MP	192	}
4363ac7c MP	193
e73c23ff CK	194	*biop = bio;
	195	return 0;
	196	}
	197
	198	/**
	199	* blkdev_issue_write_same - queue a write same operation
	200	* @bdev: target blockdev
	201	* @sector: start sector
	202	* @nr_sects: number of sectors to write
	203	* @gfp_mask: memory allocation flags (for bio_alloc)
	204	* @page: page containing data
	205	*
	206	* Description:
	207	* Issue a write same request for the sectors in question.
	208	*/
	209	int blkdev_issue_write_same(struct block_device *bdev, sector_t sector,
	210	sector_t nr_sects, gfp_t gfp_mask,
	211	struct page *page)
	212	{
	213	struct bio *bio = NULL;
	214	struct blk_plug plug;
	215	int ret;
	216
	217	blk_start_plug(&plug);
	218	ret = __blkdev_issue_write_same(bdev, sector, nr_sects, gfp_mask, page,
	219	&bio);
	220	if (ret == 0 && bio) {
4e49ea4a	221	ret = submit_bio_wait(bio);
05bd92dd ST	222	bio_put(bio);
05bd92dd ST	223	}
e73c23ff	224	blk_finish_plug(&plug);
3f40bf2c	225	return ret;
4363ac7c MP	226	}
	227	EXPORT_SYMBOL(blkdev_issue_write_same);
	228
a6f0788e CK	229	/**
	230	* __blkdev_issue_write_zeroes - generate number of bios with WRITE ZEROES
	231	* @bdev: blockdev to issue
	232	* @sector: start sector
	233	* @nr_sects: number of sectors to write
	234	* @gfp_mask: memory allocation flags (for bio_alloc)
	235	* @biop: pointer to anchor bio
	236	*
	237	* Description:
	238	* Generate and issue number of bios(REQ_OP_WRITE_ZEROES) with zerofiled pages.
	239	*/
	240	static int __blkdev_issue_write_zeroes(struct block_device *bdev,
	241	sector_t sector, sector_t nr_sects, gfp_t gfp_mask,
	242	struct bio **biop)
	243	{
	244	struct bio bio = biop;
	245	unsigned int max_write_zeroes_sectors;
	246	struct request_queue *q = bdev_get_queue(bdev);
	247
	248	if (!q)
	249	return -ENXIO;
	250
	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;
	260	bio->bi_bdev = bdev;
	261	bio_set_op_attrs(bio, REQ_OP_WRITE_ZEROES, 0);
	262
	263	if (nr_sects > max_write_zeroes_sectors) {
	264	bio->bi_iter.bi_size = max_write_zeroes_sectors << 9;
	265	nr_sects -= max_write_zeroes_sectors;
	266	sector += max_write_zeroes_sectors;
	267	} else {
	268	bio->bi_iter.bi_size = nr_sects << 9;
	269	nr_sects = 0;
	270	}
	271	cond_resched();
	272	}
	273
	274	*biop = bio;
	275	return 0;
	276	}
	277
3f14d792	278	/**
e73c23ff	279	* __blkdev_issue_zeroout - generate number of zero filed write bios
3f14d792 DM	280	* @bdev: blockdev to issue
	281	* @sector: start sector
	282	* @nr_sects: number of sectors to write
	283	* @gfp_mask: memory allocation flags (for bio_alloc)
e73c23ff CK	284	* @biop: pointer to anchor bio
e73c23ff CK	285	* @discard: discard flag
3f14d792 DM	286	*
	287	* Description:
	288	* Generate and issue number of bios with zerofiled pages.
3f14d792	289	*/
e73c23ff CK	290	int __blkdev_issue_zeroout(struct block_device *bdev, sector_t sector,
	291	sector_t nr_sects, gfp_t gfp_mask, struct bio **biop,
	292	bool discard)
3f14d792	293	{
18edc8ea	294	int ret;
e73c23ff CK	295	int bi_size = 0;
e73c23ff CK	296	struct bio bio = biop;
0aeea189	297	unsigned int sz;
28b2be20 DW	298	sector_t bs_mask;
	299
	300	bs_mask = (bdev_logical_block_size(bdev) >> 9) - 1;
	301	if ((sector \| nr_sects) & bs_mask)
	302	return -EINVAL;
3f14d792	303
a6f0788e CK	304	ret = __blkdev_issue_write_zeroes(bdev, sector, nr_sects, gfp_mask,
	305	biop);
	306	if (ret == 0 \|\| (ret && ret != -EOPNOTSUPP))
	307	goto out;
	308
e73c23ff	309	ret = 0;
3f14d792	310	while (nr_sects != 0) {
4e49ea4a	311	bio = next_bio(bio, min(nr_sects, (sector_t)BIO_MAX_PAGES),
9082e87b	312	gfp_mask);
4f024f37	313	bio->bi_iter.bi_sector = sector;
3f14d792	314	bio->bi_bdev = bdev;
95fe6c1a	315	bio_set_op_attrs(bio, REQ_OP_WRITE, 0);
3f14d792	316
0341aafb JA	317	while (nr_sects != 0) {
0341aafb JA	318	sz = min((sector_t) PAGE_SIZE >> 9 , nr_sects);
e73c23ff CK	319	bi_size = bio_add_page(bio, ZERO_PAGE(0), sz << 9, 0);
	320	nr_sects -= bi_size >> 9;
	321	sector += bi_size >> 9;
	322	if (bi_size < (sz << 9))
3f14d792 DM	323	break;
3f14d792 DM	324	}
e73c23ff	325	cond_resched();
3f14d792	326	}
3f14d792	327
e73c23ff CK	328	*biop = bio;
	329	out:
	330	return ret;
3f14d792	331	}
e73c23ff	332	EXPORT_SYMBOL(__blkdev_issue_zeroout);
579e8f3c MP	333
	334	/**
	335	* blkdev_issue_zeroout - zero-fill a block range
	336	* @bdev: blockdev to write
	337	* @sector: start sector
	338	* @nr_sects: number of sectors to write
	339	* @gfp_mask: memory allocation flags (for bio_alloc)
d93ba7a5	340	* @discard: whether to discard the block range
579e8f3c MP	341	*
579e8f3c MP	342	* Description:
d93ba7a5 MP	343	* Zero-fill a block range. If the discard flag is set and the block
	344	* device guarantees that subsequent READ operations to the block range
	345	* in question will return zeroes, the blocks will be discarded. Should
	346	* the discard request fail, if the discard flag is not set, or if
	347	* discard_zeroes_data is not supported, this function will resort to
	348	* zeroing the blocks manually, thus provisioning (allocating,
a6f0788e CK	349	* anchoring) them. If the block device supports WRITE ZEROES or WRITE SAME
a6f0788e CK	350	* command(s), blkdev_issue_zeroout() will use it to optimize the process of
d93ba7a5 MP	351	* clearing the block range. Otherwise the zeroing will be performed
d93ba7a5 MP	352	* using regular WRITE calls.
579e8f3c	353	*/
579e8f3c	354	int blkdev_issue_zeroout(struct block_device *bdev, sector_t sector,
d93ba7a5	355	sector_t nr_sects, gfp_t gfp_mask, bool discard)
579e8f3c	356	{
e73c23ff CK	357	int ret;
	358	struct bio *bio = NULL;
	359	struct blk_plug plug;
d93ba7a5	360
bef13315 CH	361	if (discard) {
	362	if (!blkdev_issue_discard(bdev, sector, nr_sects, gfp_mask,
	363	BLKDEV_DISCARD_ZERO))
	364	return 0;
	365	}
	366
eeeefd41 CH	367	if (!blkdev_issue_write_same(bdev, sector, nr_sects, gfp_mask,
	368	ZERO_PAGE(0)))
	369	return 0;
	370
e73c23ff CK	371	blk_start_plug(&plug);
	372	ret = __blkdev_issue_zeroout(bdev, sector, nr_sects, gfp_mask,
	373	&bio, discard);
	374	if (ret == 0 && bio) {
	375	ret = submit_bio_wait(bio);
	376	bio_put(bio);
	377	}
	378	blk_finish_plug(&plug);
579e8f3c	379
e73c23ff	380	return ret;
579e8f3c	381	}
3f14d792	382	EXPORT_SYMBOL(blkdev_issue_zeroout);