[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"

int __blkdev_issue_discard(struct block_device *bdev, sector_t sector,
		sector_t nr_sects, gfp_t gfp_mask, int flags,
		struct bio **biop)
{
	struct request_queue *q = bdev_get_queue(bdev);
	struct bio *bio = *biop;
	unsigned int op;
	sector_t bs_mask, part_offset = 0;

	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;
	}

	/* In case the discard granularity isn't set by buggy device driver */
	if (WARN_ON_ONCE(!q->limits.discard_granularity)) {
		char dev_name[BDEVNAME_SIZE];

		bdevname(bdev, dev_name);
		pr_err_ratelimited("%s: Error: discard_granularity is 0.\n", dev_name);
		return -EOPNOTSUPP;
	}

	bs_mask = (bdev_logical_block_size(bdev) >> 9) - 1;
	if ((sector | nr_sects) & bs_mask)
		return -EINVAL;

	if (!nr_sects)
		return -EINVAL;

	/* In case the discard request is in a partition */
	if (bdev_is_partition(bdev))
		part_offset = bdev->bd_start_sect;

	while (nr_sects) {
		sector_t granularity_aligned_lba, req_sects;
		sector_t sector_mapped = sector + part_offset;

		granularity_aligned_lba = round_up(sector_mapped,
				q->limits.discard_granularity >> SECTOR_SHIFT);

		/*
		 * Check whether the discard bio starts at a discard_granularity
		 * aligned LBA,
		 * - If no: set (granularity_aligned_lba - sector_mapped) to
		 *   bi_size of the first split bio, then the second bio will
		 *   start at a discard_granularity aligned LBA on the device.
		 * - If yes: use bio_aligned_discard_max_sectors() as the max
		 *   possible bi_size of the first split bio. Then when this bio
		 *   is split in device drive, the split ones are very probably
		 *   to be aligned to discard_granularity of the device's queue.
		 */
		if (granularity_aligned_lba == sector_mapped)
			req_sects = min_t(sector_t, nr_sects,
					  bio_aligned_discard_max_sectors(q));
		else
			req_sects = min_t(sector_t, nr_sects,
					  granularity_aligned_lba - sector_mapped);

		WARN_ON_ONCE((req_sects << 9) > UINT_MAX);

		bio = blk_next_bio(bio, bdev, 0, op, gfp_mask);
		bio->bi_iter.bi_sector = sector;
		bio->bi_iter.bi_size = req_sects << 9;
		sector += req_sects;
		nr_sects -= req_sects;

		/*
		 * 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 (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 = bio_allowed_max_sectors(q);

	while (nr_sects) {
		bio = blk_next_bio(bio, bdev, 1, REQ_OP_WRITE_SAME, gfp_mask);
		bio->bi_iter.bi_sector = sector;
		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);

		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;

	if (bdev_read_only(bdev))
		return -EPERM;

	/* Ensure that max_write_zeroes_sectors doesn't overflow bi_size */
	max_write_zeroes_sectors = bdev_write_zeroes_sectors(bdev);

	if (max_write_zeroes_sectors == 0)
		return -EOPNOTSUPP;

	while (nr_sects) {
		bio = blk_next_bio(bio, bdev, 0, REQ_OP_WRITE_ZEROES, gfp_mask);
		bio->bi_iter.bi_sector = sector;
		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_VECS);
}

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 bio *bio = *biop;
	int bi_size = 0;
	unsigned int sz;

	if (bdev_read_only(bdev))
		return -EPERM;

	while (nr_sects != 0) {
		bio = blk_next_bio(bio, bdev, __blkdev_sectors_to_bio_pages(nr_sects),
				   REQ_OP_WRITE, gfp_mask);
		bio->bi_iter.bi_sector = sector;

		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
38f25255	13	int __blkdev_issue_discard(struct block_device *bdev, sector_t sector,
288dab8a	14	sector_t nr_sects, gfp_t gfp_mask, int flags,
469e3216	15	struct bio **biop)
f31e7e40	16	{
f31e7e40	17	struct request_queue *q = bdev_get_queue(bdev);
38f25255	18	struct bio bio = biop;
ef295ecf	19	unsigned int op;
9b15d109	20	sector_t bs_mask, part_offset = 0;
f31e7e40	21
a13553c7 ID	22	if (bdev_read_only(bdev))
	23	return -EPERM;
	24
288dab8a CH	25	if (flags & BLKDEV_DISCARD_SECURE) {
	26	if (!blk_queue_secure_erase(q))
	27	return -EOPNOTSUPP;
	28	op = REQ_OP_SECURE_ERASE;
	29	} else {
	30	if (!blk_queue_discard(q))
	31	return -EOPNOTSUPP;
	32	op = REQ_OP_DISCARD;
	33	}
f31e7e40	34
b35fd742 CL	35	/* In case the discard granularity isn't set by buggy device driver */
	36	if (WARN_ON_ONCE(!q->limits.discard_granularity)) {
	37	char dev_name[BDEVNAME_SIZE];
	38
	39	bdevname(bdev, dev_name);
	40	pr_err_ratelimited("%s: Error: discard_granularity is 0.\n", dev_name);
	41	return -EOPNOTSUPP;
	42	}
	43
28b2be20 DW	44	bs_mask = (bdev_logical_block_size(bdev) >> 9) - 1;
	45	if ((sector \| nr_sects) & bs_mask)
	46	return -EINVAL;
	47
ba5d7385 ML	48	if (!nr_sects)
ba5d7385 ML	49	return -EINVAL;
a22c4d7e	50
9b15d109	51	/* In case the discard request is in a partition */
fa01b1e9	52	if (bdev_is_partition(bdev))
29ff57c6	53	part_offset = bdev->bd_start_sect;
9b15d109	54
ba5d7385	55	while (nr_sects) {
9b15d109 CL	56	sector_t granularity_aligned_lba, req_sects;
	57	sector_t sector_mapped = sector + part_offset;
	58
	59	granularity_aligned_lba = round_up(sector_mapped,
	60	q->limits.discard_granularity >> SECTOR_SHIFT);
	61
	62	/*
	63	* Check whether the discard bio starts at a discard_granularity
	64	* aligned LBA,
	65	* - If no: set (granularity_aligned_lba - sector_mapped) to
	66	* bi_size of the first split bio, then the second bio will
	67	* start at a discard_granularity aligned LBA on the device.
	68	* - If yes: use bio_aligned_discard_max_sectors() as the max
	69	* possible bi_size of the first split bio. Then when this bio
	70	* is split in device drive, the split ones are very probably
	71	* to be aligned to discard_granularity of the device's queue.
	72	*/
	73	if (granularity_aligned_lba == sector_mapped)
	74	req_sects = min_t(sector_t, nr_sects,
	75	bio_aligned_discard_max_sectors(q));
	76	else
	77	req_sects = min_t(sector_t, nr_sects,
	78	granularity_aligned_lba - sector_mapped);
c6e66634	79
4800bf7b DC	80	WARN_ON_ONCE((req_sects << 9) > UINT_MAX);
4800bf7b DC	81
0a3140ea	82	bio = blk_next_bio(bio, bdev, 0, op, gfp_mask);
4f024f37	83	bio->bi_iter.bi_sector = sector;
4f024f37	84	bio->bi_iter.bi_size = req_sects << 9;
ba5d7385	85	sector += req_sects;
c6e66634	86	nr_sects -= req_sects;
f31e7e40	87
c8123f8c JA	88	/*
	89	* We can loop for a long time in here, if someone does
	90	* full device discards (like mkfs). Be nice and allow
	91	* us to schedule out to avoid softlocking if preempt
	92	* is disabled.
	93	*/
	94	cond_resched();
5dba3089	95	}
38f25255 CH	96
	97	*biop = bio;
	98	return 0;
	99	}
	100	EXPORT_SYMBOL(__blkdev_issue_discard);
	101
	102	/**
	103	* blkdev_issue_discard - queue a discard
	104	* @bdev: blockdev to issue discard for
	105	* @sector: start sector
	106	* @nr_sects: number of sectors to discard
	107	* @gfp_mask: memory allocation flags (for bio_alloc)
e554911c	108	* @flags: BLKDEV_DISCARD_* flags to control behaviour
38f25255 CH	109	*
	110	* Description:
	111	* Issue a discard request for the sectors in question.
	112	*/
	113	int blkdev_issue_discard(struct block_device *bdev, sector_t sector,
	114	sector_t nr_sects, gfp_t gfp_mask, unsigned long flags)
	115	{
38f25255 CH	116	struct bio *bio = NULL;
	117	struct blk_plug plug;
	118	int ret;
	119
38f25255	120	blk_start_plug(&plug);
288dab8a	121	ret = __blkdev_issue_discard(bdev, sector, nr_sects, gfp_mask, flags,
38f25255	122	&bio);
bbd848e0	123	if (!ret && bio) {
4e49ea4a	124	ret = submit_bio_wait(bio);
48920ff2	125	if (ret == -EOPNOTSUPP)
bbd848e0	126	ret = 0;
05bd92dd	127	bio_put(bio);
bbd848e0	128	}
0cfbcafc	129	blk_finish_plug(&plug);
f31e7e40	130
bbd848e0	131	return ret;
f31e7e40 DM	132	}
f31e7e40 DM	133	EXPORT_SYMBOL(blkdev_issue_discard);
3f14d792	134
4363ac7c	135	/**
e73c23ff	136	* __blkdev_issue_write_same - generate number of bios with same page
4363ac7c MP	137	* @bdev: target blockdev
	138	* @sector: start sector
	139	* @nr_sects: number of sectors to write
	140	* @gfp_mask: memory allocation flags (for bio_alloc)
	141	* @page: page containing data to write
e73c23ff	142	* @biop: pointer to anchor bio
4363ac7c MP	143	*
4363ac7c MP	144	* Description:
e73c23ff	145	* Generate and issue number of bios(REQ_OP_WRITE_SAME) with same page.
4363ac7c	146	*/
e73c23ff CK	147	static int __blkdev_issue_write_same(struct block_device *bdev, sector_t sector,
	148	sector_t nr_sects, gfp_t gfp_mask, struct page *page,
	149	struct bio **biop)
4363ac7c	150	{
4363ac7c MP	151	struct request_queue *q = bdev_get_queue(bdev);
4363ac7c MP	152	unsigned int max_write_same_sectors;
e73c23ff	153	struct bio bio = biop;
28b2be20	154	sector_t bs_mask;
4363ac7c	155
a13553c7 ID	156	if (bdev_read_only(bdev))
	157	return -EPERM;
	158
28b2be20 DW	159	bs_mask = (bdev_logical_block_size(bdev) >> 9) - 1;
	160	if ((sector \| nr_sects) & bs_mask)
	161	return -EINVAL;
	162
e73c23ff CK	163	if (!bdev_write_same(bdev))
	164	return -EOPNOTSUPP;
	165
b49a0871	166	/* Ensure that max_write_same_sectors doesn't overflow bi_size */
34ffec60	167	max_write_same_sectors = bio_allowed_max_sectors(q);
4363ac7c	168
4363ac7c	169	while (nr_sects) {
0a3140ea	170	bio = blk_next_bio(bio, bdev, 1, REQ_OP_WRITE_SAME, gfp_mask);
4f024f37	171	bio->bi_iter.bi_sector = sector;
4363ac7c MP	172	bio->bi_vcnt = 1;
	173	bio->bi_io_vec->bv_page = page;
	174	bio->bi_io_vec->bv_offset = 0;
	175	bio->bi_io_vec->bv_len = bdev_logical_block_size(bdev);
	176
	177	if (nr_sects > max_write_same_sectors) {
4f024f37	178	bio->bi_iter.bi_size = max_write_same_sectors << 9;
4363ac7c MP	179	nr_sects -= max_write_same_sectors;
	180	sector += max_write_same_sectors;
	181	} else {
4f024f37	182	bio->bi_iter.bi_size = nr_sects << 9;
4363ac7c MP	183	nr_sects = 0;
4363ac7c MP	184	}
e73c23ff	185	cond_resched();
4363ac7c MP	186	}
4363ac7c MP	187
e73c23ff CK	188	*biop = bio;
	189	return 0;
	190	}
	191
	192	/**
	193	* blkdev_issue_write_same - queue a write same operation
	194	* @bdev: target blockdev
	195	* @sector: start sector
	196	* @nr_sects: number of sectors to write
	197	* @gfp_mask: memory allocation flags (for bio_alloc)
	198	* @page: page containing data
	199	*
	200	* Description:
	201	* Issue a write same request for the sectors in question.
	202	*/
	203	int blkdev_issue_write_same(struct block_device *bdev, sector_t sector,
	204	sector_t nr_sects, gfp_t gfp_mask,
	205	struct page *page)
	206	{
	207	struct bio *bio = NULL;
	208	struct blk_plug plug;
	209	int ret;
	210
	211	blk_start_plug(&plug);
	212	ret = __blkdev_issue_write_same(bdev, sector, nr_sects, gfp_mask, page,
	213	&bio);
	214	if (ret == 0 && bio) {
4e49ea4a	215	ret = submit_bio_wait(bio);
05bd92dd ST	216	bio_put(bio);
05bd92dd ST	217	}
e73c23ff	218	blk_finish_plug(&plug);
3f40bf2c	219	return ret;
4363ac7c MP	220	}
	221	EXPORT_SYMBOL(blkdev_issue_write_same);
	222
a6f0788e CK	223	static int __blkdev_issue_write_zeroes(struct block_device *bdev,
a6f0788e CK	224	sector_t sector, sector_t nr_sects, gfp_t gfp_mask,
d928be9f	225	struct bio **biop, unsigned flags)
a6f0788e CK	226	{
	227	struct bio bio = biop;
	228	unsigned int max_write_zeroes_sectors;
a6f0788e	229
a13553c7 ID	230	if (bdev_read_only(bdev))
	231	return -EPERM;
	232
a6f0788e CK	233	/* Ensure that max_write_zeroes_sectors doesn't overflow bi_size */
	234	max_write_zeroes_sectors = bdev_write_zeroes_sectors(bdev);
	235
	236	if (max_write_zeroes_sectors == 0)
	237	return -EOPNOTSUPP;
	238
	239	while (nr_sects) {
0a3140ea	240	bio = blk_next_bio(bio, bdev, 0, REQ_OP_WRITE_ZEROES, gfp_mask);
a6f0788e	241	bio->bi_iter.bi_sector = sector;
d928be9f CH	242	if (flags & BLKDEV_ZERO_NOUNMAP)
d928be9f CH	243	bio->bi_opf \|= REQ_NOUNMAP;
a6f0788e CK	244
	245	if (nr_sects > max_write_zeroes_sectors) {
	246	bio->bi_iter.bi_size = max_write_zeroes_sectors << 9;
	247	nr_sects -= max_write_zeroes_sectors;
	248	sector += max_write_zeroes_sectors;
	249	} else {
	250	bio->bi_iter.bi_size = nr_sects << 9;
	251	nr_sects = 0;
	252	}
	253	cond_resched();
	254	}
	255
	256	*biop = bio;
	257	return 0;
	258	}
	259
615d22a5 DLM	260	/*
	261	* Convert a number of 512B sectors to a number of pages.
	262	* The result is limited to a number of pages that can fit into a BIO.
	263	* Also make sure that the result is always at least 1 (page) for the cases
	264	* where nr_sects is lower than the number of sectors in a page.
	265	*/
	266	static unsigned int __blkdev_sectors_to_bio_pages(sector_t nr_sects)
	267	{
09c2c359	268	sector_t pages = DIV_ROUND_UP_SECTOR_T(nr_sects, PAGE_SIZE / 512);
615d22a5	269
a8affc03	270	return min(pages, (sector_t)BIO_MAX_VECS);
615d22a5 DLM	271	}
615d22a5 DLM	272
425a4dba ID	273	static int __blkdev_issue_zero_pages(struct block_device *bdev,
	274	sector_t sector, sector_t nr_sects, gfp_t gfp_mask,
	275	struct bio **biop)
	276	{
425a4dba ID	277	struct bio bio = biop;
	278	int bi_size = 0;
	279	unsigned int sz;
	280
a13553c7 ID	281	if (bdev_read_only(bdev))
	282	return -EPERM;
	283
425a4dba	284	while (nr_sects != 0) {
0a3140ea CK	285	bio = blk_next_bio(bio, bdev, __blkdev_sectors_to_bio_pages(nr_sects),
0a3140ea CK	286	REQ_OP_WRITE, gfp_mask);
425a4dba	287	bio->bi_iter.bi_sector = sector;
425a4dba ID	288
	289	while (nr_sects != 0) {
	290	sz = min((sector_t) PAGE_SIZE, nr_sects << 9);
	291	bi_size = bio_add_page(bio, ZERO_PAGE(0), sz, 0);
	292	nr_sects -= bi_size >> 9;
	293	sector += bi_size >> 9;
	294	if (bi_size < sz)
	295	break;
	296	}
	297	cond_resched();
	298	}
	299
	300	*biop = bio;
	301	return 0;
	302	}
	303
3f14d792	304	/**
e73c23ff	305	* __blkdev_issue_zeroout - generate number of zero filed write bios
3f14d792 DM	306	* @bdev: blockdev to issue
	307	* @sector: start sector
	308	* @nr_sects: number of sectors to write
	309	* @gfp_mask: memory allocation flags (for bio_alloc)
e73c23ff	310	* @biop: pointer to anchor bio
ee472d83	311	* @flags: controls detailed behavior
3f14d792 DM	312	*
3f14d792 DM	313	* Description:
ee472d83 CH	314	* Zero-fill a block range, either using hardware offload or by explicitly
	315	* writing zeroes to the device.
	316	*
	317	* If a device is using logical block provisioning, the underlying space will
	318	* not be released if %flags contains BLKDEV_ZERO_NOUNMAP.
cb365b96 CH	319	*
	320	* If %flags contains BLKDEV_ZERO_NOFALLBACK, the function will return
	321	* -EOPNOTSUPP if no explicit hardware offload for zeroing is provided.
3f14d792	322	*/
e73c23ff CK	323	int __blkdev_issue_zeroout(struct block_device *bdev, sector_t sector,
e73c23ff CK	324	sector_t nr_sects, gfp_t gfp_mask, struct bio **biop,
ee472d83	325	unsigned flags)
3f14d792	326	{
18edc8ea	327	int ret;
28b2be20 DW	328	sector_t bs_mask;
	329
	330	bs_mask = (bdev_logical_block_size(bdev) >> 9) - 1;
	331	if ((sector \| nr_sects) & bs_mask)
	332	return -EINVAL;
3f14d792	333
a6f0788e	334	ret = __blkdev_issue_write_zeroes(bdev, sector, nr_sects, gfp_mask,
d928be9f	335	biop, flags);
cb365b96	336	if (ret != -EOPNOTSUPP \|\| (flags & BLKDEV_ZERO_NOFALLBACK))
425a4dba	337	return ret;
3f14d792	338
425a4dba ID	339	return __blkdev_issue_zero_pages(bdev, sector, nr_sects, gfp_mask,
425a4dba ID	340	biop);
3f14d792	341	}
e73c23ff	342	EXPORT_SYMBOL(__blkdev_issue_zeroout);
579e8f3c MP	343
	344	/**
	345	* blkdev_issue_zeroout - zero-fill a block range
	346	* @bdev: blockdev to write
	347	* @sector: start sector
	348	* @nr_sects: number of sectors to write
	349	* @gfp_mask: memory allocation flags (for bio_alloc)
ee472d83	350	* @flags: controls detailed behavior
579e8f3c MP	351	*
579e8f3c MP	352	* Description:
ee472d83 CH	353	* Zero-fill a block range, either using hardware offload or by explicitly
	354	* writing zeroes to the device. See __blkdev_issue_zeroout() for the
	355	* valid values for %flags.
579e8f3c	356	*/
579e8f3c	357	int blkdev_issue_zeroout(struct block_device *bdev, sector_t sector,
ee472d83	358	sector_t nr_sects, gfp_t gfp_mask, unsigned flags)
579e8f3c	359	{
d5ce4c31 ID	360	int ret = 0;
	361	sector_t bs_mask;
	362	struct bio *bio;
e73c23ff	363	struct blk_plug plug;
d5ce4c31	364	bool try_write_zeroes = !!bdev_write_zeroes_sectors(bdev);
d93ba7a5	365
d5ce4c31 ID	366	bs_mask = (bdev_logical_block_size(bdev) >> 9) - 1;
	367	if ((sector \| nr_sects) & bs_mask)
	368	return -EINVAL;
	369
	370	retry:
	371	bio = NULL;
e73c23ff	372	blk_start_plug(&plug);
d5ce4c31 ID	373	if (try_write_zeroes) {
	374	ret = __blkdev_issue_write_zeroes(bdev, sector, nr_sects,
	375	gfp_mask, &bio, flags);
	376	} else if (!(flags & BLKDEV_ZERO_NOFALLBACK)) {
	377	ret = __blkdev_issue_zero_pages(bdev, sector, nr_sects,
	378	gfp_mask, &bio);
	379	} else {
	380	/* No zeroing offload support */
	381	ret = -EOPNOTSUPP;
	382	}
e73c23ff CK	383	if (ret == 0 && bio) {
	384	ret = submit_bio_wait(bio);
	385	bio_put(bio);
	386	}
	387	blk_finish_plug(&plug);
d5ce4c31 ID	388	if (ret && try_write_zeroes) {
	389	if (!(flags & BLKDEV_ZERO_NOFALLBACK)) {
	390	try_write_zeroes = false;
	391	goto retry;
	392	}
	393	if (!bdev_write_zeroes_sectors(bdev)) {
	394	/*
	395	* Zeroing offload support was indicated, but the
	396	* device reported ILLEGAL REQUEST (for some devices
	397	* there is no non-destructive way to verify whether
	398	* WRITE ZEROES is actually supported).
	399	*/
	400	ret = -EOPNOTSUPP;
	401	}
	402	}
579e8f3c	403
e73c23ff	404	return ret;
579e8f3c	405	}
3f14d792	406	EXPORT_SYMBOL(blkdev_issue_zeroout);