[mirror_ubuntu-zesty-kernel.git] / block / ioctl.c

#include <linux/capability.h>
#include <linux/blkdev.h>
#include <linux/export.h>
#include <linux/gfp.h>
#include <linux/blkpg.h>
#include <linux/hdreg.h>
#include <linux/backing-dev.h>
#include <linux/fs.h>
#include <linux/blktrace_api.h>
#include <linux/pr.h>
#include <asm/uaccess.h>

static int blkpg_ioctl(struct block_device *bdev, struct blkpg_ioctl_arg __user *arg)
{
	struct block_device *bdevp;
	struct gendisk *disk;
	struct hd_struct *part, *lpart;
	struct blkpg_ioctl_arg a;
	struct blkpg_partition p;
	struct disk_part_iter piter;
	long long start, length;
	int partno;

	if (!capable(CAP_SYS_ADMIN))
		return -EACCES;
	if (copy_from_user(&a, arg, sizeof(struct blkpg_ioctl_arg)))
		return -EFAULT;
	if (copy_from_user(&p, a.data, sizeof(struct blkpg_partition)))
		return -EFAULT;
	disk = bdev->bd_disk;
	if (bdev != bdev->bd_contains)
		return -EINVAL;
	partno = p.pno;
	if (partno <= 0)
		return -EINVAL;
	switch (a.op) {
		case BLKPG_ADD_PARTITION:
			start = p.start >> 9;
			length = p.length >> 9;
			/* check for fit in a hd_struct */
			if (sizeof(sector_t) == sizeof(long) &&
			    sizeof(long long) > sizeof(long)) {
				long pstart = start, plength = length;
				if (pstart != start || plength != length
				    || pstart < 0 || plength < 0 || partno > 65535)
					return -EINVAL;
			}

			mutex_lock(&bdev->bd_mutex);

			/* overlap? */
			disk_part_iter_init(&piter, disk,
					    DISK_PITER_INCL_EMPTY);
			while ((part = disk_part_iter_next(&piter))) {
				if (!(start + length <= part->start_sect ||
				      start >= part->start_sect + part->nr_sects)) {
					disk_part_iter_exit(&piter);
					mutex_unlock(&bdev->bd_mutex);
					return -EBUSY;
				}
			}
			disk_part_iter_exit(&piter);

			/* all seems OK */
			part = add_partition(disk, partno, start, length,
					     ADDPART_FLAG_NONE, NULL);
			mutex_unlock(&bdev->bd_mutex);
			return PTR_ERR_OR_ZERO(part);
		case BLKPG_DEL_PARTITION:
			part = disk_get_part(disk, partno);
			if (!part)
				return -ENXIO;

			bdevp = bdget(part_devt(part));
			disk_put_part(part);
			if (!bdevp)
				return -ENOMEM;

			mutex_lock(&bdevp->bd_mutex);
			if (bdevp->bd_openers) {
				mutex_unlock(&bdevp->bd_mutex);
				bdput(bdevp);
				return -EBUSY;
			}
			/* all seems OK */
			fsync_bdev(bdevp);
			invalidate_bdev(bdevp);

			mutex_lock_nested(&bdev->bd_mutex, 1);
			delete_partition(disk, partno);
			mutex_unlock(&bdev->bd_mutex);
			mutex_unlock(&bdevp->bd_mutex);
			bdput(bdevp);

			return 0;
		case BLKPG_RESIZE_PARTITION:
			start = p.start >> 9;
			/* new length of partition in bytes */
			length = p.length >> 9;
			/* check for fit in a hd_struct */
			if (sizeof(sector_t) == sizeof(long) &&
			    sizeof(long long) > sizeof(long)) {
				long pstart = start, plength = length;
				if (pstart != start || plength != length
				    || pstart < 0 || plength < 0)
					return -EINVAL;
			}
			part = disk_get_part(disk, partno);
			if (!part)
				return -ENXIO;
			bdevp = bdget(part_devt(part));
			if (!bdevp) {
				disk_put_part(part);
				return -ENOMEM;
			}
			mutex_lock(&bdevp->bd_mutex);
			mutex_lock_nested(&bdev->bd_mutex, 1);
			if (start != part->start_sect) {
				mutex_unlock(&bdevp->bd_mutex);
				mutex_unlock(&bdev->bd_mutex);
				bdput(bdevp);
				disk_put_part(part);
				return -EINVAL;
			}
			/* overlap? */
			disk_part_iter_init(&piter, disk,
					    DISK_PITER_INCL_EMPTY);
			while ((lpart = disk_part_iter_next(&piter))) {
				if (lpart->partno != partno &&
				   !(start + length <= lpart->start_sect ||
				   start >= lpart->start_sect + lpart->nr_sects)
				   ) {
					disk_part_iter_exit(&piter);
					mutex_unlock(&bdevp->bd_mutex);
					mutex_unlock(&bdev->bd_mutex);
					bdput(bdevp);
					disk_put_part(part);
					return -EBUSY;
				}
			}
			disk_part_iter_exit(&piter);
			part_nr_sects_write(part, (sector_t)length);
			i_size_write(bdevp->bd_inode, p.length);
			mutex_unlock(&bdevp->bd_mutex);
			mutex_unlock(&bdev->bd_mutex);
			bdput(bdevp);
			disk_put_part(part);
			return 0;
		default:
			return -EINVAL;
	}
}

/*
 * This is an exported API for the block driver, and will not
 * acquire bd_mutex. This API should be used in case that
 * caller has held bd_mutex already.
 */
int __blkdev_reread_part(struct block_device *bdev)
{
	struct gendisk *disk = bdev->bd_disk;

	if (!disk_part_scan_enabled(disk) || bdev != bdev->bd_contains)
		return -EINVAL;
	if (!capable(CAP_SYS_ADMIN))
		return -EACCES;

	lockdep_assert_held(&bdev->bd_mutex);

	return rescan_partitions(disk, bdev);
}
EXPORT_SYMBOL(__blkdev_reread_part);

/*
 * This is an exported API for the block driver, and will
 * try to acquire bd_mutex. If bd_mutex has been held already
 * in current context, please call __blkdev_reread_part().
 *
 * Make sure the held locks in current context aren't required
 * in open()/close() handler and I/O path for avoiding ABBA deadlock:
 * - bd_mutex is held before calling block driver's open/close
 *   handler
 * - reading partition table may submit I/O to the block device
 */
int blkdev_reread_part(struct block_device *bdev)
{
	int res;

	mutex_lock(&bdev->bd_mutex);
	res = __blkdev_reread_part(bdev);
	mutex_unlock(&bdev->bd_mutex);

	return res;
}
EXPORT_SYMBOL(blkdev_reread_part);

static int blk_ioctl_discard(struct block_device *bdev, fmode_t mode,
		unsigned long arg, unsigned long flags)
{
	uint64_t range[2];
	uint64_t start, len;

	if (!(mode & FMODE_WRITE))
		return -EBADF;

	if (copy_from_user(range, (void __user *)arg, sizeof(range)))
		return -EFAULT;

	start = range[0];
	len = range[1];

	if (start & 511)
		return -EINVAL;
	if (len & 511)
		return -EINVAL;
	start >>= 9;
	len >>= 9;

	if (start + len > (i_size_read(bdev->bd_inode) >> 9))
		return -EINVAL;
	return blkdev_issue_discard(bdev, start, len, GFP_KERNEL, flags);
}

static int blk_ioctl_zeroout(struct block_device *bdev, fmode_t mode,
		unsigned long arg)
{
	uint64_t range[2];
	uint64_t start, len;

	if (!(mode & FMODE_WRITE))
		return -EBADF;

	if (copy_from_user(range, (void __user *)arg, sizeof(range)))
		return -EFAULT;

	start = range[0];
	len = range[1];

	if (start & 511)
		return -EINVAL;
	if (len & 511)
		return -EINVAL;
	start >>= 9;
	len >>= 9;

	if (start + len > (i_size_read(bdev->bd_inode) >> 9))
		return -EINVAL;

	return blkdev_issue_zeroout(bdev, start, len, GFP_KERNEL, false);
}

static int put_ushort(unsigned long arg, unsigned short val)
{
	return put_user(val, (unsigned short __user *)arg);
}

static int put_int(unsigned long arg, int val)
{
	return put_user(val, (int __user *)arg);
}

static int put_uint(unsigned long arg, unsigned int val)
{
	return put_user(val, (unsigned int __user *)arg);
}

static int put_long(unsigned long arg, long val)
{
	return put_user(val, (long __user *)arg);
}

static int put_ulong(unsigned long arg, unsigned long val)
{
	return put_user(val, (unsigned long __user *)arg);
}

static int put_u64(unsigned long arg, u64 val)
{
	return put_user(val, (u64 __user *)arg);
}

int __blkdev_driver_ioctl(struct block_device *bdev, fmode_t mode,
			unsigned cmd, unsigned long arg)
{
	struct gendisk *disk = bdev->bd_disk;

	if (disk->fops->ioctl)
		return disk->fops->ioctl(bdev, mode, cmd, arg);

	return -ENOTTY;
}
/*
 * For the record: _GPL here is only because somebody decided to slap it
 * on the previous export.  Sheer idiocy, since it wasn't copyrightable
 * at all and could be open-coded without any exports by anybody who cares.
 */
EXPORT_SYMBOL_GPL(__blkdev_driver_ioctl);

static int blkdev_pr_register(struct block_device *bdev,
		struct pr_registration __user *arg)
{
	const struct pr_ops *ops = bdev->bd_disk->fops->pr_ops;
	struct pr_registration reg;

	if (!capable(CAP_SYS_ADMIN))
		return -EPERM;
	if (!ops || !ops->pr_register)
		return -EOPNOTSUPP;
	if (copy_from_user(&reg, arg, sizeof(reg)))
		return -EFAULT;

	if (reg.flags & ~PR_FL_IGNORE_KEY)
		return -EOPNOTSUPP;
	return ops->pr_register(bdev, reg.old_key, reg.new_key, reg.flags);
}

static int blkdev_pr_reserve(struct block_device *bdev,
		struct pr_reservation __user *arg)
{
	const struct pr_ops *ops = bdev->bd_disk->fops->pr_ops;
	struct pr_reservation rsv;

	if (!capable(CAP_SYS_ADMIN))
		return -EPERM;
	if (!ops || !ops->pr_reserve)
		return -EOPNOTSUPP;
	if (copy_from_user(&rsv, arg, sizeof(rsv)))
		return -EFAULT;

	if (rsv.flags & ~PR_FL_IGNORE_KEY)
		return -EOPNOTSUPP;
	return ops->pr_reserve(bdev, rsv.key, rsv.type, rsv.flags);
}

static int blkdev_pr_release(struct block_device *bdev,
		struct pr_reservation __user *arg)
{
	const struct pr_ops *ops = bdev->bd_disk->fops->pr_ops;
	struct pr_reservation rsv;

	if (!capable(CAP_SYS_ADMIN))
		return -EPERM;
	if (!ops || !ops->pr_release)
		return -EOPNOTSUPP;
	if (copy_from_user(&rsv, arg, sizeof(rsv)))
		return -EFAULT;

	if (rsv.flags)
		return -EOPNOTSUPP;
	return ops->pr_release(bdev, rsv.key, rsv.type);
}

static int blkdev_pr_preempt(struct block_device *bdev,
		struct pr_preempt __user *arg, bool abort)
{
	const struct pr_ops *ops = bdev->bd_disk->fops->pr_ops;
	struct pr_preempt p;

	if (!capable(CAP_SYS_ADMIN))
		return -EPERM;
	if (!ops || !ops->pr_preempt)
		return -EOPNOTSUPP;
	if (copy_from_user(&p, arg, sizeof(p)))
		return -EFAULT;

	if (p.flags)
		return -EOPNOTSUPP;
	return ops->pr_preempt(bdev, p.old_key, p.new_key, p.type, abort);
}

static int blkdev_pr_clear(struct block_device *bdev,
		struct pr_clear __user *arg)
{
	const struct pr_ops *ops = bdev->bd_disk->fops->pr_ops;
	struct pr_clear c;

	if (!capable(CAP_SYS_ADMIN))
		return -EPERM;
	if (!ops || !ops->pr_clear)
		return -EOPNOTSUPP;
	if (copy_from_user(&c, arg, sizeof(c)))
		return -EFAULT;

	if (c.flags)
		return -EOPNOTSUPP;
	return ops->pr_clear(bdev, c.key);
}

/*
 * Is it an unrecognized ioctl? The correct returns are either
 * ENOTTY (final) or ENOIOCTLCMD ("I don't know this one, try a
 * fallback"). ENOIOCTLCMD gets turned into ENOTTY by the ioctl
 * code before returning.
 *
 * Confused drivers sometimes return EINVAL, which is wrong. It
 * means "I understood the ioctl command, but the parameters to
 * it were wrong".
 *
 * We should aim to just fix the broken drivers, the EINVAL case
 * should go away.
 */
static inline int is_unrecognized_ioctl(int ret)
{
	return	ret == -EINVAL ||
		ret == -ENOTTY ||
		ret == -ENOIOCTLCMD;
}

static int blkdev_flushbuf(struct block_device *bdev, fmode_t mode,
		unsigned cmd, unsigned long arg)
{
	int ret;

	if (!capable(CAP_SYS_ADMIN))
		return -EACCES;

	ret = __blkdev_driver_ioctl(bdev, mode, cmd, arg);
	if (!is_unrecognized_ioctl(ret))
		return ret;

	fsync_bdev(bdev);
	invalidate_bdev(bdev);
	return 0;
}

static int blkdev_roset(struct block_device *bdev, fmode_t mode,
		unsigned cmd, unsigned long arg)
{
	int ret, n;

	ret = __blkdev_driver_ioctl(bdev, mode, cmd, arg);
	if (!is_unrecognized_ioctl(ret))
		return ret;
	if (!capable(CAP_SYS_ADMIN))
		return -EACCES;
	if (get_user(n, (int __user *)arg))
		return -EFAULT;
	set_device_ro(bdev, n);
	return 0;
}

static int blkdev_getgeo(struct block_device *bdev,
		struct hd_geometry __user *argp)
{
	struct gendisk *disk = bdev->bd_disk;
	struct hd_geometry geo;
	int ret;

	if (!argp)
		return -EINVAL;
	if (!disk->fops->getgeo)
		return -ENOTTY;

	/*
	 * We need to set the startsect first, the driver may
	 * want to override it.
	 */
	memset(&geo, 0, sizeof(geo));
	geo.start = get_start_sect(bdev);
	ret = disk->fops->getgeo(bdev, &geo);
	if (ret)
		return ret;
	if (copy_to_user(argp, &geo, sizeof(geo)))
		return -EFAULT;
	return 0;
}

/* set the logical block size */
static int blkdev_bszset(struct block_device *bdev, fmode_t mode,
		int __user *argp)
{
	int ret, n;

	if (!capable(CAP_SYS_ADMIN))
		return -EACCES;
	if (!argp)
		return -EINVAL;
	if (get_user(n, argp))
		return -EFAULT;

	if (!(mode & FMODE_EXCL)) {
		bdgrab(bdev);
		if (blkdev_get(bdev, mode | FMODE_EXCL, &bdev) < 0)
			return -EBUSY;
	}

	ret = set_blocksize(bdev, n);
	if (!(mode & FMODE_EXCL))
		blkdev_put(bdev, mode | FMODE_EXCL);
	return ret;
}

/*
 * always keep this in sync with compat_blkdev_ioctl()
 */
int blkdev_ioctl(struct block_device *bdev, fmode_t mode, unsigned cmd,
			unsigned long arg)
{
	struct backing_dev_info *bdi;
	void __user *argp = (void __user *)arg;
	loff_t size;
	unsigned int max_sectors;

	switch (cmd) {
	case BLKFLSBUF:
		return blkdev_flushbuf(bdev, mode, cmd, arg);
	case BLKROSET:
		return blkdev_roset(bdev, mode, cmd, arg);
	case BLKDISCARD:
		return blk_ioctl_discard(bdev, mode, arg, 0);
	case BLKSECDISCARD:
		return blk_ioctl_discard(bdev, mode, arg,
				BLKDEV_DISCARD_SECURE);
	case BLKZEROOUT:
		return blk_ioctl_zeroout(bdev, mode, arg);
	case HDIO_GETGEO:
		return blkdev_getgeo(bdev, argp);
	case BLKRAGET:
	case BLKFRAGET:
		if (!arg)
			return -EINVAL;
		bdi = blk_get_backing_dev_info(bdev);
		return put_long(arg, (bdi->ra_pages * PAGE_SIZE) / 512);
	case BLKROGET:
		return put_int(arg, bdev_read_only(bdev) != 0);
	case BLKBSZGET: /* get block device soft block size (cf. BLKSSZGET) */
		return put_int(arg, block_size(bdev));
	case BLKSSZGET: /* get block device logical block size */
		return put_int(arg, bdev_logical_block_size(bdev));
	case BLKPBSZGET: /* get block device physical block size */
		return put_uint(arg, bdev_physical_block_size(bdev));
	case BLKIOMIN:
		return put_uint(arg, bdev_io_min(bdev));
	case BLKIOOPT:
		return put_uint(arg, bdev_io_opt(bdev));
	case BLKALIGNOFF:
		return put_int(arg, bdev_alignment_offset(bdev));
	case BLKDISCARDZEROES:
		return put_uint(arg, bdev_discard_zeroes_data(bdev));
	case BLKSECTGET:
		max_sectors = min_t(unsigned int, USHRT_MAX,
				    queue_max_sectors(bdev_get_queue(bdev)));
		return put_ushort(arg, max_sectors);
	case BLKROTATIONAL:
		return put_ushort(arg, !blk_queue_nonrot(bdev_get_queue(bdev)));
	case BLKRASET:
	case BLKFRASET:
		if(!capable(CAP_SYS_ADMIN))
			return -EACCES;
		bdi = blk_get_backing_dev_info(bdev);
		bdi->ra_pages = (arg * 512) / PAGE_SIZE;
		return 0;
	case BLKBSZSET:
		return blkdev_bszset(bdev, mode, argp);
	case BLKPG:
		return blkpg_ioctl(bdev, argp);
	case BLKRRPART:
		return blkdev_reread_part(bdev);
	case BLKGETSIZE:
		size = i_size_read(bdev->bd_inode);
		if ((size >> 9) > ~0UL)
			return -EFBIG;
		return put_ulong(arg, size >> 9);
	case BLKGETSIZE64:
		return put_u64(arg, i_size_read(bdev->bd_inode));
	case BLKTRACESTART:
	case BLKTRACESTOP:
	case BLKTRACESETUP:
	case BLKTRACETEARDOWN:
		return blk_trace_ioctl(bdev, cmd, argp);
	case IOC_PR_REGISTER:
		return blkdev_pr_register(bdev, argp);
	case IOC_PR_RESERVE:
		return blkdev_pr_reserve(bdev, argp);
	case IOC_PR_RELEASE:
		return blkdev_pr_release(bdev, argp);
	case IOC_PR_PREEMPT:
		return blkdev_pr_preempt(bdev, argp, false);
	case IOC_PR_PREEMPT_ABORT:
		return blkdev_pr_preempt(bdev, argp, true);
	case IOC_PR_CLEAR:
		return blkdev_pr_clear(bdev, argp);
	default:
		return __blkdev_driver_ioctl(bdev, mode, cmd, arg);
	}
}
EXPORT_SYMBOL_GPL(blkdev_ioctl);
Commit	Line	Data
c59ede7b	1	#include <linux/capability.h>
1da177e4	2	#include <linux/blkdev.h>
d5decd3b	3	#include <linux/export.h>
5a0e3ad6	4	#include <linux/gfp.h>
1da177e4	5	#include <linux/blkpg.h>
a885c8c4	6	#include <linux/hdreg.h>
1da177e4	7	#include <linux/backing-dev.h>
ff01bb48	8	#include <linux/fs.h>
2056a782	9	#include <linux/blktrace_api.h>
bbd3e064	10	#include <linux/pr.h>
1da177e4 LT	11	#include <asm/uaccess.h>
	12
	13	static int blkpg_ioctl(struct block_device bdev, struct blkpg_ioctl_arg __user arg)
	14	{
	15	struct block_device *bdevp;
	16	struct gendisk *disk;
c83f6bf9	17	struct hd_struct part, lpart;
1da177e4 LT	18	struct blkpg_ioctl_arg a;
1da177e4 LT	19	struct blkpg_partition p;
e71bf0d0	20	struct disk_part_iter piter;
1da177e4	21	long long start, length;
cf771cb5	22	int partno;
1da177e4 LT	23
	24	if (!capable(CAP_SYS_ADMIN))
	25	return -EACCES;
	26	if (copy_from_user(&a, arg, sizeof(struct blkpg_ioctl_arg)))
	27	return -EFAULT;
	28	if (copy_from_user(&p, a.data, sizeof(struct blkpg_partition)))
	29	return -EFAULT;
	30	disk = bdev->bd_disk;
	31	if (bdev != bdev->bd_contains)
	32	return -EINVAL;
cf771cb5	33	partno = p.pno;
540eed56	34	if (partno <= 0)
1da177e4 LT	35	return -EINVAL;
	36	switch (a.op) {
	37	case BLKPG_ADD_PARTITION:
	38	start = p.start >> 9;
	39	length = p.length >> 9;
c83f6bf9 VG	40	/* check for fit in a hd_struct */
c83f6bf9 VG	41	if (sizeof(sector_t) == sizeof(long) &&
1da177e4 LT	42	sizeof(long long) > sizeof(long)) {
	43	long pstart = start, plength = length;
	44	if (pstart != start \|\| plength != length
2bd6efad	45	\|\| pstart < 0 \|\| plength < 0 \|\| partno > 65535)
1da177e4 LT	46	return -EINVAL;
1da177e4 LT	47	}
88e34126	48
c039e313	49	mutex_lock(&bdev->bd_mutex);
88e34126	50
1da177e4	51	/* overlap? */
e71bf0d0 TH	52	disk_part_iter_init(&piter, disk,
	53	DISK_PITER_INCL_EMPTY);
	54	while ((part = disk_part_iter_next(&piter))) {
	55	if (!(start + length <= part->start_sect \|\|
	56	start >= part->start_sect + part->nr_sects)) {
	57	disk_part_iter_exit(&piter);
c039e313	58	mutex_unlock(&bdev->bd_mutex);
1da177e4 LT	59	return -EBUSY;
	60	}
	61	}
e71bf0d0 TH	62	disk_part_iter_exit(&piter);
e71bf0d0 TH	63
1da177e4	64	/* all seems OK */
ba32929a	65	part = add_partition(disk, partno, start, length,
6d1d8050	66	ADDPART_FLAG_NONE, NULL);
c039e313	67	mutex_unlock(&bdev->bd_mutex);
c7d1ba41	68	return PTR_ERR_OR_ZERO(part);
1da177e4	69	case BLKPG_DEL_PARTITION:
e71bf0d0 TH	70	part = disk_get_part(disk, partno);
e71bf0d0 TH	71	if (!part)
1da177e4	72	return -ENXIO;
e71bf0d0 TH	73
	74	bdevp = bdget(part_devt(part));
	75	disk_put_part(part);
1da177e4 LT	76	if (!bdevp)
1da177e4 LT	77	return -ENOMEM;
e71bf0d0	78
2e7b651d	79	mutex_lock(&bdevp->bd_mutex);
1da177e4	80	if (bdevp->bd_openers) {
c039e313	81	mutex_unlock(&bdevp->bd_mutex);
1da177e4 LT	82	bdput(bdevp);
	83	return -EBUSY;
	84	}
	85	/* all seems OK */
	86	fsync_bdev(bdevp);
f98393a6	87	invalidate_bdev(bdevp);
1da177e4	88
6d740cd5	89	mutex_lock_nested(&bdev->bd_mutex, 1);
cf771cb5	90	delete_partition(disk, partno);
c039e313 AV	91	mutex_unlock(&bdev->bd_mutex);
c039e313 AV	92	mutex_unlock(&bdevp->bd_mutex);
1da177e4 LT	93	bdput(bdevp);
1da177e4 LT	94
c83f6bf9 VG	95	return 0;
	96	case BLKPG_RESIZE_PARTITION:
	97	start = p.start >> 9;
	98	/* new length of partition in bytes */
	99	length = p.length >> 9;
	100	/* check for fit in a hd_struct */
	101	if (sizeof(sector_t) == sizeof(long) &&
	102	sizeof(long long) > sizeof(long)) {
	103	long pstart = start, plength = length;
	104	if (pstart != start \|\| plength != length
	105	\|\| pstart < 0 \|\| plength < 0)
	106	return -EINVAL;
	107	}
	108	part = disk_get_part(disk, partno);
	109	if (!part)
	110	return -ENXIO;
	111	bdevp = bdget(part_devt(part));
	112	if (!bdevp) {
	113	disk_put_part(part);
	114	return -ENOMEM;
	115	}
	116	mutex_lock(&bdevp->bd_mutex);
	117	mutex_lock_nested(&bdev->bd_mutex, 1);
	118	if (start != part->start_sect) {
	119	mutex_unlock(&bdevp->bd_mutex);
	120	mutex_unlock(&bdev->bd_mutex);
	121	bdput(bdevp);
	122	disk_put_part(part);
	123	return -EINVAL;
	124	}
	125	/* overlap? */
	126	disk_part_iter_init(&piter, disk,
	127	DISK_PITER_INCL_EMPTY);
	128	while ((lpart = disk_part_iter_next(&piter))) {
	129	if (lpart->partno != partno &&
	130	!(start + length <= lpart->start_sect \|\|
	131	start >= lpart->start_sect + lpart->nr_sects)
	132	) {
	133	disk_part_iter_exit(&piter);
	134	mutex_unlock(&bdevp->bd_mutex);
	135	mutex_unlock(&bdev->bd_mutex);
	136	bdput(bdevp);
	137	disk_put_part(part);
	138	return -EBUSY;
	139	}
	140	}
	141	disk_part_iter_exit(&piter);
	142	part_nr_sects_write(part, (sector_t)length);
	143	i_size_write(bdevp->bd_inode, p.length);
	144	mutex_unlock(&bdevp->bd_mutex);
	145	mutex_unlock(&bdev->bd_mutex);
	146	bdput(bdevp);
	147	disk_put_part(part);
1da177e4 LT	148	return 0;
	149	default:
	150	return -EINVAL;
	151	}
	152	}
	153
be324177 JW	154	/*
	155	* This is an exported API for the block driver, and will not
	156	* acquire bd_mutex. This API should be used in case that
	157	* caller has held bd_mutex already.
	158	*/
	159	int __blkdev_reread_part(struct block_device *bdev)
1da177e4 LT	160	{
1da177e4 LT	161	struct gendisk *disk = bdev->bd_disk;
1da177e4	162
d27769ec	163	if (!disk_part_scan_enabled(disk) \|\| bdev != bdev->bd_contains)
1da177e4 LT	164	return -EINVAL;
	165	if (!capable(CAP_SYS_ADMIN))
	166	return -EACCES;
be324177 JW	167
	168	lockdep_assert_held(&bdev->bd_mutex);
	169
	170	return rescan_partitions(disk, bdev);
	171	}
	172	EXPORT_SYMBOL(__blkdev_reread_part);
	173
	174	/*
	175	* This is an exported API for the block driver, and will
	176	* try to acquire bd_mutex. If bd_mutex has been held already
	177	* in current context, please call __blkdev_reread_part().
b04a5636 ML	178	*
	179	* Make sure the held locks in current context aren't required
	180	* in open()/close() handler and I/O path for avoiding ABBA deadlock:
	181	* - bd_mutex is held before calling block driver's open/close
	182	* handler
	183	* - reading partition table may submit I/O to the block device
be324177 JW	184	*/
	185	int blkdev_reread_part(struct block_device *bdev)
	186	{
	187	int res;
	188
b04a5636	189	mutex_lock(&bdev->bd_mutex);
be324177	190	res = __blkdev_reread_part(bdev);
c039e313	191	mutex_unlock(&bdev->bd_mutex);
be324177	192
1da177e4 LT	193	return res;
1da177e4 LT	194	}
be324177	195	EXPORT_SYMBOL(blkdev_reread_part);
1da177e4	196
d8e4bb81 CH	197	static int blk_ioctl_discard(struct block_device *bdev, fmode_t mode,
d8e4bb81 CH	198	unsigned long arg, unsigned long flags)
d30a2605	199	{
d8e4bb81 CH	200	uint64_t range[2];
	201	uint64_t start, len;
	202
	203	if (!(mode & FMODE_WRITE))
	204	return -EBADF;
	205
	206	if (copy_from_user(range, (void __user *)arg, sizeof(range)))
	207	return -EFAULT;
	208
	209	start = range[0];
	210	len = range[1];
8d57a98c	211
d30a2605 DW	212	if (start & 511)
	213	return -EINVAL;
	214	if (len & 511)
	215	return -EINVAL;
	216	start >>= 9;
	217	len >>= 9;
	218
77304d2a	219	if (start + len > (i_size_read(bdev->bd_inode) >> 9))
d30a2605	220	return -EINVAL;
8d57a98c	221	return blkdev_issue_discard(bdev, start, len, GFP_KERNEL, flags);
d30a2605 DW	222	}
d30a2605 DW	223
d8e4bb81 CH	224	static int blk_ioctl_zeroout(struct block_device *bdev, fmode_t mode,
d8e4bb81 CH	225	unsigned long arg)
66ba32dc	226	{
d8e4bb81 CH	227	uint64_t range[2];
	228	uint64_t start, len;
	229
	230	if (!(mode & FMODE_WRITE))
	231	return -EBADF;
	232
	233	if (copy_from_user(range, (void __user *)arg, sizeof(range)))
	234	return -EFAULT;
	235
	236	start = range[0];
	237	len = range[1];
	238
66ba32dc MP	239	if (start & 511)
	240	return -EINVAL;
	241	if (len & 511)
	242	return -EINVAL;
	243	start >>= 9;
	244	len >>= 9;
	245
	246	if (start + len > (i_size_read(bdev->bd_inode) >> 9))
	247	return -EINVAL;
	248
d93ba7a5	249	return blkdev_issue_zeroout(bdev, start, len, GFP_KERNEL, false);
66ba32dc MP	250	}
66ba32dc MP	251
1da177e4 LT	252	static int put_ushort(unsigned long arg, unsigned short val)
	253	{
	254	return put_user(val, (unsigned short __user *)arg);
	255	}
	256
	257	static int put_int(unsigned long arg, int val)
	258	{
	259	return put_user(val, (int __user *)arg);
	260	}
	261
ac481c20 MP	262	static int put_uint(unsigned long arg, unsigned int val)
	263	{
	264	return put_user(val, (unsigned int __user *)arg);
	265	}
	266
1da177e4 LT	267	static int put_long(unsigned long arg, long val)
	268	{
	269	return put_user(val, (long __user *)arg);
	270	}
	271
	272	static int put_ulong(unsigned long arg, unsigned long val)
	273	{
	274	return put_user(val, (unsigned long __user *)arg);
	275	}
	276
	277	static int put_u64(unsigned long arg, u64 val)
	278	{
	279	return put_user(val, (u64 __user *)arg);
	280	}
	281
633a08b8 AV	282	int __blkdev_driver_ioctl(struct block_device *bdev, fmode_t mode,
	283	unsigned cmd, unsigned long arg)
	284	{
	285	struct gendisk *disk = bdev->bd_disk;
d4430d62 AV	286
	287	if (disk->fops->ioctl)
	288	return disk->fops->ioctl(bdev, mode, cmd, arg);
633a08b8	289
633a08b8 AV	290	return -ENOTTY;
	291	}
	292	/*
	293	* For the record: _GPL here is only because somebody decided to slap it
	294	* on the previous export. Sheer idiocy, since it wasn't copyrightable
	295	* at all and could be open-coded without any exports by anybody who cares.
	296	*/
	297	EXPORT_SYMBOL_GPL(__blkdev_driver_ioctl);
	298
bbd3e064 CH	299	static int blkdev_pr_register(struct block_device *bdev,
	300	struct pr_registration __user *arg)
	301	{
	302	const struct pr_ops *ops = bdev->bd_disk->fops->pr_ops;
	303	struct pr_registration reg;
	304
	305	if (!capable(CAP_SYS_ADMIN))
	306	return -EPERM;
	307	if (!ops \|\| !ops->pr_register)
	308	return -EOPNOTSUPP;
	309	if (copy_from_user(&reg, arg, sizeof(reg)))
	310	return -EFAULT;
	311
	312	if (reg.flags & ~PR_FL_IGNORE_KEY)
	313	return -EOPNOTSUPP;
	314	return ops->pr_register(bdev, reg.old_key, reg.new_key, reg.flags);
	315	}
	316
	317	static int blkdev_pr_reserve(struct block_device *bdev,
	318	struct pr_reservation __user *arg)
	319	{
	320	const struct pr_ops *ops = bdev->bd_disk->fops->pr_ops;
	321	struct pr_reservation rsv;
	322
	323	if (!capable(CAP_SYS_ADMIN))
	324	return -EPERM;
	325	if (!ops \|\| !ops->pr_reserve)
	326	return -EOPNOTSUPP;
	327	if (copy_from_user(&rsv, arg, sizeof(rsv)))
	328	return -EFAULT;
	329
	330	if (rsv.flags & ~PR_FL_IGNORE_KEY)
	331	return -EOPNOTSUPP;
	332	return ops->pr_reserve(bdev, rsv.key, rsv.type, rsv.flags);
	333	}
	334
	335	static int blkdev_pr_release(struct block_device *bdev,
	336	struct pr_reservation __user *arg)
	337	{
	338	const struct pr_ops *ops = bdev->bd_disk->fops->pr_ops;
	339	struct pr_reservation rsv;
	340
	341	if (!capable(CAP_SYS_ADMIN))
	342	return -EPERM;
	343	if (!ops \|\| !ops->pr_release)
	344	return -EOPNOTSUPP;
	345	if (copy_from_user(&rsv, arg, sizeof(rsv)))
	346	return -EFAULT;
	347
	348	if (rsv.flags)
	349	return -EOPNOTSUPP;
	350	return ops->pr_release(bdev, rsv.key, rsv.type);
	351	}
	352
	353	static int blkdev_pr_preempt(struct block_device *bdev,
	354	struct pr_preempt __user *arg, bool abort)
	355	{
	356	const struct pr_ops *ops = bdev->bd_disk->fops->pr_ops;
	357	struct pr_preempt p;
	358
	359	if (!capable(CAP_SYS_ADMIN))
	360	return -EPERM;
	361	if (!ops \|\| !ops->pr_preempt)
	362	return -EOPNOTSUPP;
363	if (copy_from_user(&p, arg, sizeof(p)))
364	return -EFAULT;
365
366	if (p.flags)
367	return -EOPNOTSUPP;
368	return ops->pr_preempt(bdev, p.old_key, p.new_key, p.type, abort);
369	}
370
371	static int blkdev_pr_clear(struct block_device *bdev,
372	struct pr_clear __user *arg)
373	{
374	const struct pr_ops *ops = bdev->bd_disk->fops->pr_ops;
375	struct pr_clear c;
376
377	if (!capable(CAP_SYS_ADMIN))
378	return -EPERM;
379	if (!ops \|\| !ops->pr_clear)
380	return -EOPNOTSUPP;
381	if (copy_from_user(&c, arg, sizeof(c)))
382	return -EFAULT;
383
384	if (c.flags)
385	return -EOPNOTSUPP;
386	return ops->pr_clear(bdev, c.key);
387	}
388
07d106d0 LT	389	/*
	390	* Is it an unrecognized ioctl? The correct returns are either
	391	* ENOTTY (final) or ENOIOCTLCMD ("I don't know this one, try a
	392	* fallback"). ENOIOCTLCMD gets turned into ENOTTY by the ioctl
	393	* code before returning.
	394	*
	395	* Confused drivers sometimes return EINVAL, which is wrong. It
	396	* means "I understood the ioctl command, but the parameters to
	397	* it were wrong".
	398	*
	399	* We should aim to just fix the broken drivers, the EINVAL case
	400	* should go away.
	401	*/
	402	static inline int is_unrecognized_ioctl(int ret)
	403	{
	404	return ret == -EINVAL \|\|
	405	ret == -ENOTTY \|\|
	406	ret == -ENOIOCTLCMD;
	407	}
	408
d8e4bb81 CH	409	static int blkdev_flushbuf(struct block_device *bdev, fmode_t mode,
d8e4bb81 CH	410	unsigned cmd, unsigned long arg)
bb93e3a5	411	{
d8e4bb81	412	int ret;
bb93e3a5	413
d8e4bb81 CH	414	if (!capable(CAP_SYS_ADMIN))
d8e4bb81 CH	415	return -EACCES;
bb93e3a5	416
d8e4bb81 CH	417	ret = __blkdev_driver_ioctl(bdev, mode, cmd, arg);
	418	if (!is_unrecognized_ioctl(ret))
	419	return ret;
bb93e3a5	420
d8e4bb81 CH	421	fsync_bdev(bdev);
	422	invalidate_bdev(bdev);
	423	return 0;
	424	}
d30a2605	425
d8e4bb81 CH	426	static int blkdev_roset(struct block_device *bdev, fmode_t mode,
	427	unsigned cmd, unsigned long arg)
	428	{
	429	int ret, n;
d30a2605	430
d8e4bb81 CH	431	ret = __blkdev_driver_ioctl(bdev, mode, cmd, arg);
	432	if (!is_unrecognized_ioctl(ret))
	433	return ret;
	434	if (!capable(CAP_SYS_ADMIN))
	435	return -EACCES;
	436	if (get_user(n, (int __user *)arg))
	437	return -EFAULT;
	438	set_device_ro(bdev, n);
	439	return 0;
	440	}
d30a2605	441
d8e4bb81 CH	442	static int blkdev_getgeo(struct block_device *bdev,
	443	struct hd_geometry __user *argp)
	444	{
	445	struct gendisk *disk = bdev->bd_disk;
	446	struct hd_geometry geo;
	447	int ret;
d30a2605	448
d8e4bb81 CH	449	if (!argp)
	450	return -EINVAL;
	451	if (!disk->fops->getgeo)
	452	return -ENOTTY;
	453
	454	/*
	455	* We need to set the startsect first, the driver may
	456	* want to override it.
	457	*/
	458	memset(&geo, 0, sizeof(geo));
	459	geo.start = get_start_sect(bdev);
	460	ret = disk->fops->getgeo(bdev, &geo);
	461	if (ret)
	462	return ret;
	463	if (copy_to_user(argp, &geo, sizeof(geo)))
	464	return -EFAULT;
	465	return 0;
	466	}
66ba32dc	467
d8e4bb81 CH	468	/* set the logical block size */
	469	static int blkdev_bszset(struct block_device *bdev, fmode_t mode,
	470	int __user *argp)
	471	{
	472	int ret, n;
66ba32dc	473
d8e4bb81 CH	474	if (!capable(CAP_SYS_ADMIN))
	475	return -EACCES;
	476	if (!argp)
	477	return -EINVAL;
	478	if (get_user(n, argp))
	479	return -EFAULT;
66ba32dc	480
d8e4bb81 CH	481	if (!(mode & FMODE_EXCL)) {
	482	bdgrab(bdev);
	483	if (blkdev_get(bdev, mode \| FMODE_EXCL, &bdev) < 0)
	484	return -EBUSY;
66ba32dc	485	}
d30a2605	486
d8e4bb81 CH	487	ret = set_blocksize(bdev, n);
	488	if (!(mode & FMODE_EXCL))
	489	blkdev_put(bdev, mode \| FMODE_EXCL);
	490	return ret;
	491	}
a885c8c4	492
d8e4bb81 CH	493	/*
	494	* always keep this in sync with compat_blkdev_ioctl()
	495	*/
	496	int blkdev_ioctl(struct block_device *bdev, fmode_t mode, unsigned cmd,
	497	unsigned long arg)
	498	{
	499	struct backing_dev_info *bdi;
	500	void __user argp = (void __user )arg;
	501	loff_t size;
	502	unsigned int max_sectors;
	503
	504	switch (cmd) {
	505	case BLKFLSBUF:
	506	return blkdev_flushbuf(bdev, mode, cmd, arg);
	507	case BLKROSET:
	508	return blkdev_roset(bdev, mode, cmd, arg);
	509	case BLKDISCARD:
	510	return blk_ioctl_discard(bdev, mode, arg, 0);
	511	case BLKSECDISCARD:
	512	return blk_ioctl_discard(bdev, mode, arg,
	513	BLKDEV_DISCARD_SECURE);
	514	case BLKZEROOUT:
	515	return blk_ioctl_zeroout(bdev, mode, arg);
	516	case HDIO_GETGEO:
	517	return blkdev_getgeo(bdev, argp);
45048d09 AV	518	case BLKRAGET:
	519	case BLKFRAGET:
	520	if (!arg)
	521	return -EINVAL;
	522	bdi = blk_get_backing_dev_info(bdev);
09cbfeaf	523	return put_long(arg, (bdi->ra_pages * PAGE_SIZE) / 512);
45048d09 AV	524	case BLKROGET:
45048d09 AV	525	return put_int(arg, bdev_read_only(bdev) != 0);
ac481c20	526	case BLKBSZGET: /* get block device soft block size (cf. BLKSSZGET) */
45048d09	527	return put_int(arg, block_size(bdev));
ac481c20	528	case BLKSSZGET: /* get block device logical block size */
e1defc4f	529	return put_int(arg, bdev_logical_block_size(bdev));
ac481c20 MP	530	case BLKPBSZGET: /* get block device physical block size */
	531	return put_uint(arg, bdev_physical_block_size(bdev));
	532	case BLKIOMIN:
	533	return put_uint(arg, bdev_io_min(bdev));
	534	case BLKIOOPT:
	535	return put_uint(arg, bdev_io_opt(bdev));
	536	case BLKALIGNOFF:
	537	return put_int(arg, bdev_alignment_offset(bdev));
98262f27 MP	538	case BLKDISCARDZEROES:
98262f27 MP	539	return put_uint(arg, bdev_discard_zeroes_data(bdev));
45048d09	540	case BLKSECTGET:
63f26496 AM	541	max_sectors = min_t(unsigned int, USHRT_MAX,
	542	queue_max_sectors(bdev_get_queue(bdev)));
	543	return put_ushort(arg, max_sectors);
ef00f59c MP	544	case BLKROTATIONAL:
ef00f59c MP	545	return put_ushort(arg, !blk_queue_nonrot(bdev_get_queue(bdev)));
45048d09 AV	546	case BLKRASET:
	547	case BLKFRASET:
	548	if(!capable(CAP_SYS_ADMIN))
	549	return -EACCES;
	550	bdi = blk_get_backing_dev_info(bdev);
09cbfeaf	551	bdi->ra_pages = (arg * 512) / PAGE_SIZE;
45048d09 AV	552	return 0;
45048d09 AV	553	case BLKBSZSET:
d8e4bb81	554	return blkdev_bszset(bdev, mode, argp);
45048d09	555	case BLKPG:
d8e4bb81	556	return blkpg_ioctl(bdev, argp);
45048d09	557	case BLKRRPART:
d8e4bb81	558	return blkdev_reread_part(bdev);
45048d09	559	case BLKGETSIZE:
77304d2a	560	size = i_size_read(bdev->bd_inode);
45048d09 AV	561	if ((size >> 9) > ~0UL)
	562	return -EFBIG;
	563	return put_ulong(arg, size >> 9);
	564	case BLKGETSIZE64:
77304d2a	565	return put_u64(arg, i_size_read(bdev->bd_inode));
45048d09 AV	566	case BLKTRACESTART:
	567	case BLKTRACESTOP:
	568	case BLKTRACESETUP:
	569	case BLKTRACETEARDOWN:
d8e4bb81	570	return blk_trace_ioctl(bdev, cmd, argp);
bbd3e064 CH	571	case IOC_PR_REGISTER:
	572	return blkdev_pr_register(bdev, argp);
	573	case IOC_PR_RESERVE:
	574	return blkdev_pr_reserve(bdev, argp);
	575	case IOC_PR_RELEASE:
	576	return blkdev_pr_release(bdev, argp);
	577	case IOC_PR_PREEMPT:
	578	return blkdev_pr_preempt(bdev, argp, false);
	579	case IOC_PR_PREEMPT_ABORT:
	580	return blkdev_pr_preempt(bdev, argp, true);
	581	case IOC_PR_CLEAR:
	582	return blkdev_pr_clear(bdev, argp);
45048d09	583	default:
d8e4bb81	584	return __blkdev_driver_ioctl(bdev, mode, cmd, arg);
45048d09	585	}
1da177e4	586	}
68f66feb	587	EXPORT_SYMBOL_GPL(blkdev_ioctl);