[mirror_ubuntu-bionic-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 <linux/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;
			}
			/* check if partition is aligned to blocksize */
			if (p.start & (bdev_logical_block_size(bdev) - 1))
				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];
	struct address_space *mapping;
	uint64_t start, end, 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];
	end = start + len - 1;

	if (start & 511)
		return -EINVAL;
	if (len & 511)
		return -EINVAL;
	if (end >= (uint64_t)i_size_read(bdev->bd_inode))
		return -EINVAL;
	if (end < start)
		return -EINVAL;

	/* Invalidate the page cache, including dirty pages */
	mapping = bdev->bd_inode->i_mapping;
	truncate_inode_pages_range(mapping, start, end);

	return blkdev_issue_zeroout(bdev, start >> 9, len >> 9, 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)
{
	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 BLKREPORTZONE:
		return blkdev_report_zones_ioctl(bdev, mode, cmd, arg);
	case BLKRESETZONE:
		return blkdev_reset_zones_ioctl(bdev, mode, cmd, arg);
	case HDIO_GETGEO:
		return blkdev_getgeo(bdev, argp);
	case BLKRAGET:
	case BLKFRAGET:
		if (!arg)
			return -EINVAL;
		return put_long(arg, (bdev->bd_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;
		bdev->bd_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>
7c0f6ba6	11	#include <linux/uaccess.h>
1da177e4 LT	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	}
633395b6 SH	48	/* check if partition is aligned to blocksize */
	49	if (p.start & (bdev_logical_block_size(bdev) - 1))
	50	return -EINVAL;
88e34126	51
c039e313	52	mutex_lock(&bdev->bd_mutex);
88e34126	53
1da177e4	54	/* overlap? */
e71bf0d0 TH	55	disk_part_iter_init(&piter, disk,
	56	DISK_PITER_INCL_EMPTY);
	57	while ((part = disk_part_iter_next(&piter))) {
	58	if (!(start + length <= part->start_sect \|\|
	59	start >= part->start_sect + part->nr_sects)) {
	60	disk_part_iter_exit(&piter);
c039e313	61	mutex_unlock(&bdev->bd_mutex);
1da177e4 LT	62	return -EBUSY;
	63	}
	64	}
e71bf0d0 TH	65	disk_part_iter_exit(&piter);
e71bf0d0 TH	66
1da177e4	67	/* all seems OK */
ba32929a	68	part = add_partition(disk, partno, start, length,
6d1d8050	69	ADDPART_FLAG_NONE, NULL);
c039e313	70	mutex_unlock(&bdev->bd_mutex);
c7d1ba41	71	return PTR_ERR_OR_ZERO(part);
1da177e4	72	case BLKPG_DEL_PARTITION:
e71bf0d0 TH	73	part = disk_get_part(disk, partno);
e71bf0d0 TH	74	if (!part)
1da177e4	75	return -ENXIO;
e71bf0d0 TH	76
	77	bdevp = bdget(part_devt(part));
	78	disk_put_part(part);
1da177e4 LT	79	if (!bdevp)
1da177e4 LT	80	return -ENOMEM;
e71bf0d0	81
2e7b651d	82	mutex_lock(&bdevp->bd_mutex);
1da177e4	83	if (bdevp->bd_openers) {
c039e313	84	mutex_unlock(&bdevp->bd_mutex);
1da177e4 LT	85	bdput(bdevp);
	86	return -EBUSY;
	87	}
	88	/* all seems OK */
	89	fsync_bdev(bdevp);
f98393a6	90	invalidate_bdev(bdevp);
1da177e4	91
6d740cd5	92	mutex_lock_nested(&bdev->bd_mutex, 1);
cf771cb5	93	delete_partition(disk, partno);
c039e313 AV	94	mutex_unlock(&bdev->bd_mutex);
c039e313 AV	95	mutex_unlock(&bdevp->bd_mutex);
1da177e4 LT	96	bdput(bdevp);
1da177e4 LT	97
c83f6bf9 VG	98	return 0;
	99	case BLKPG_RESIZE_PARTITION:
	100	start = p.start >> 9;
	101	/* new length of partition in bytes */
	102	length = p.length >> 9;
	103	/* check for fit in a hd_struct */
	104	if (sizeof(sector_t) == sizeof(long) &&
	105	sizeof(long long) > sizeof(long)) {
	106	long pstart = start, plength = length;
	107	if (pstart != start \|\| plength != length
	108	\|\| pstart < 0 \|\| plength < 0)
	109	return -EINVAL;
	110	}
	111	part = disk_get_part(disk, partno);
	112	if (!part)
	113	return -ENXIO;
	114	bdevp = bdget(part_devt(part));
	115	if (!bdevp) {
	116	disk_put_part(part);
	117	return -ENOMEM;
	118	}
	119	mutex_lock(&bdevp->bd_mutex);
	120	mutex_lock_nested(&bdev->bd_mutex, 1);
	121	if (start != part->start_sect) {
	122	mutex_unlock(&bdevp->bd_mutex);
	123	mutex_unlock(&bdev->bd_mutex);
	124	bdput(bdevp);
	125	disk_put_part(part);
	126	return -EINVAL;
	127	}
	128	/* overlap? */
	129	disk_part_iter_init(&piter, disk,
	130	DISK_PITER_INCL_EMPTY);
	131	while ((lpart = disk_part_iter_next(&piter))) {
	132	if (lpart->partno != partno &&
	133	!(start + length <= lpart->start_sect \|\|
	134	start >= lpart->start_sect + lpart->nr_sects)
	135	) {
	136	disk_part_iter_exit(&piter);
	137	mutex_unlock(&bdevp->bd_mutex);
	138	mutex_unlock(&bdev->bd_mutex);
	139	bdput(bdevp);
	140	disk_put_part(part);
	141	return -EBUSY;
	142	}
	143	}
	144	disk_part_iter_exit(&piter);
	145	part_nr_sects_write(part, (sector_t)length);
	146	i_size_write(bdevp->bd_inode, p.length);
	147	mutex_unlock(&bdevp->bd_mutex);
	148	mutex_unlock(&bdev->bd_mutex);
	149	bdput(bdevp);
	150	disk_put_part(part);
1da177e4 LT	151	return 0;
	152	default:
	153	return -EINVAL;
	154	}
	155	}
	156
be324177 JW	157	/*
	158	* This is an exported API for the block driver, and will not
	159	* acquire bd_mutex. This API should be used in case that
	160	* caller has held bd_mutex already.
	161	*/
	162	int __blkdev_reread_part(struct block_device *bdev)
1da177e4 LT	163	{
1da177e4 LT	164	struct gendisk *disk = bdev->bd_disk;
1da177e4	165
d27769ec	166	if (!disk_part_scan_enabled(disk) \|\| bdev != bdev->bd_contains)
1da177e4 LT	167	return -EINVAL;
	168	if (!capable(CAP_SYS_ADMIN))
	169	return -EACCES;
be324177 JW	170
	171	lockdep_assert_held(&bdev->bd_mutex);
	172
	173	return rescan_partitions(disk, bdev);
	174	}
	175	EXPORT_SYMBOL(__blkdev_reread_part);
	176
	177	/*
	178	* This is an exported API for the block driver, and will
	179	* try to acquire bd_mutex. If bd_mutex has been held already
	180	* in current context, please call __blkdev_reread_part().
b04a5636 ML	181	*
	182	* Make sure the held locks in current context aren't required
	183	* in open()/close() handler and I/O path for avoiding ABBA deadlock:
	184	* - bd_mutex is held before calling block driver's open/close
	185	* handler
	186	* - reading partition table may submit I/O to the block device
be324177 JW	187	*/
	188	int blkdev_reread_part(struct block_device *bdev)
	189	{
	190	int res;
	191
b04a5636	192	mutex_lock(&bdev->bd_mutex);
be324177	193	res = __blkdev_reread_part(bdev);
c039e313	194	mutex_unlock(&bdev->bd_mutex);
be324177	195
1da177e4 LT	196	return res;
1da177e4 LT	197	}
be324177	198	EXPORT_SYMBOL(blkdev_reread_part);
1da177e4	199
d8e4bb81 CH	200	static int blk_ioctl_discard(struct block_device *bdev, fmode_t mode,
d8e4bb81 CH	201	unsigned long arg, unsigned long flags)
d30a2605	202	{
d8e4bb81 CH	203	uint64_t range[2];
	204	uint64_t start, len;
	205
	206	if (!(mode & FMODE_WRITE))
	207	return -EBADF;
	208
	209	if (copy_from_user(range, (void __user *)arg, sizeof(range)))
	210	return -EFAULT;
	211
	212	start = range[0];
	213	len = range[1];
8d57a98c	214
d30a2605 DW	215	if (start & 511)
	216	return -EINVAL;
	217	if (len & 511)
	218	return -EINVAL;
	219	start >>= 9;
	220	len >>= 9;
	221
77304d2a	222	if (start + len > (i_size_read(bdev->bd_inode) >> 9))
d30a2605	223	return -EINVAL;
8d57a98c	224	return blkdev_issue_discard(bdev, start, len, GFP_KERNEL, flags);
d30a2605 DW	225	}
d30a2605 DW	226
d8e4bb81 CH	227	static int blk_ioctl_zeroout(struct block_device *bdev, fmode_t mode,
d8e4bb81 CH	228	unsigned long arg)
66ba32dc	229	{
d8e4bb81	230	uint64_t range[2];
22dd6d35 DW	231	struct address_space *mapping;
22dd6d35 DW	232	uint64_t start, end, len;
d8e4bb81 CH	233
	234	if (!(mode & FMODE_WRITE))
	235	return -EBADF;
	236
	237	if (copy_from_user(range, (void __user *)arg, sizeof(range)))
	238	return -EFAULT;
	239
	240	start = range[0];
	241	len = range[1];
22dd6d35	242	end = start + len - 1;
d8e4bb81	243
66ba32dc MP	244	if (start & 511)
	245	return -EINVAL;
	246	if (len & 511)
	247	return -EINVAL;
22dd6d35 DW	248	if (end >= (uint64_t)i_size_read(bdev->bd_inode))
	249	return -EINVAL;
	250	if (end < start)
66ba32dc MP	251	return -EINVAL;
66ba32dc MP	252
22dd6d35 DW	253	/* Invalidate the page cache, including dirty pages */
	254	mapping = bdev->bd_inode->i_mapping;
	255	truncate_inode_pages_range(mapping, start, end);
	256
	257	return blkdev_issue_zeroout(bdev, start >> 9, len >> 9, GFP_KERNEL,
	258	false);
66ba32dc MP	259	}
66ba32dc MP	260
1da177e4 LT	261	static int put_ushort(unsigned long arg, unsigned short val)
	262	{
	263	return put_user(val, (unsigned short __user *)arg);
	264	}
	265
	266	static int put_int(unsigned long arg, int val)
	267	{
	268	return put_user(val, (int __user *)arg);
	269	}
	270
ac481c20 MP	271	static int put_uint(unsigned long arg, unsigned int val)
	272	{
	273	return put_user(val, (unsigned int __user *)arg);
	274	}
	275
1da177e4 LT	276	static int put_long(unsigned long arg, long val)
	277	{
	278	return put_user(val, (long __user *)arg);
	279	}
	280
	281	static int put_ulong(unsigned long arg, unsigned long val)
	282	{
	283	return put_user(val, (unsigned long __user *)arg);
	284	}
	285
	286	static int put_u64(unsigned long arg, u64 val)
	287	{
	288	return put_user(val, (u64 __user *)arg);
	289	}
	290
633a08b8 AV	291	int __blkdev_driver_ioctl(struct block_device *bdev, fmode_t mode,
	292	unsigned cmd, unsigned long arg)
	293	{
	294	struct gendisk *disk = bdev->bd_disk;
d4430d62 AV	295
	296	if (disk->fops->ioctl)
	297	return disk->fops->ioctl(bdev, mode, cmd, arg);
633a08b8	298
633a08b8 AV	299	return -ENOTTY;
	300	}
	301	/*
	302	* For the record: _GPL here is only because somebody decided to slap it
	303	* on the previous export. Sheer idiocy, since it wasn't copyrightable
	304	* at all and could be open-coded without any exports by anybody who cares.
	305	*/
	306	EXPORT_SYMBOL_GPL(__blkdev_driver_ioctl);
	307
bbd3e064 CH	308	static int blkdev_pr_register(struct block_device *bdev,
	309	struct pr_registration __user *arg)
	310	{
	311	const struct pr_ops *ops = bdev->bd_disk->fops->pr_ops;
	312	struct pr_registration reg;
	313
	314	if (!capable(CAP_SYS_ADMIN))
	315	return -EPERM;
	316	if (!ops \|\| !ops->pr_register)
	317	return -EOPNOTSUPP;
	318	if (copy_from_user(&reg, arg, sizeof(reg)))
	319	return -EFAULT;
	320
	321	if (reg.flags & ~PR_FL_IGNORE_KEY)
	322	return -EOPNOTSUPP;
	323	return ops->pr_register(bdev, reg.old_key, reg.new_key, reg.flags);
	324	}
	325
	326	static int blkdev_pr_reserve(struct block_device *bdev,
	327	struct pr_reservation __user *arg)
	328	{
	329	const struct pr_ops *ops = bdev->bd_disk->fops->pr_ops;
	330	struct pr_reservation rsv;
	331
	332	if (!capable(CAP_SYS_ADMIN))
	333	return -EPERM;
	334	if (!ops \|\| !ops->pr_reserve)
	335	return -EOPNOTSUPP;
	336	if (copy_from_user(&rsv, arg, sizeof(rsv)))
	337	return -EFAULT;
	338
	339	if (rsv.flags & ~PR_FL_IGNORE_KEY)
	340	return -EOPNOTSUPP;
	341	return ops->pr_reserve(bdev, rsv.key, rsv.type, rsv.flags);
	342	}
	343
	344	static int blkdev_pr_release(struct block_device *bdev,
	345	struct pr_reservation __user *arg)
	346	{
	347	const struct pr_ops *ops = bdev->bd_disk->fops->pr_ops;
	348	struct pr_reservation rsv;
	349
	350	if (!capable(CAP_SYS_ADMIN))
	351	return -EPERM;
	352	if (!ops \|\| !ops->pr_release)
	353	return -EOPNOTSUPP;
	354	if (copy_from_user(&rsv, arg, sizeof(rsv)))
	355	return -EFAULT;
	356
	357	if (rsv.flags)
	358	return -EOPNOTSUPP;
	359	return ops->pr_release(bdev, rsv.key, rsv.type);
	360	}
	361
	362	static int blkdev_pr_preempt(struct block_device *bdev,
	363	struct pr_preempt __user *arg, bool abort)
	364	{
	365	const struct pr_ops *ops = bdev->bd_disk->fops->pr_ops;
	366	struct pr_preempt p;
	367
	368	if (!capable(CAP_SYS_ADMIN))
	369	return -EPERM;
	370	if (!ops \|\| !ops->pr_preempt)
	371	return -EOPNOTSUPP;
372	if (copy_from_user(&p, arg, sizeof(p)))
373	return -EFAULT;
374
375	if (p.flags)
376	return -EOPNOTSUPP;
377	return ops->pr_preempt(bdev, p.old_key, p.new_key, p.type, abort);
378	}
379
380	static int blkdev_pr_clear(struct block_device *bdev,
381	struct pr_clear __user *arg)
382	{
383	const struct pr_ops *ops = bdev->bd_disk->fops->pr_ops;
384	struct pr_clear c;
385
386	if (!capable(CAP_SYS_ADMIN))
387	return -EPERM;
388	if (!ops \|\| !ops->pr_clear)
389	return -EOPNOTSUPP;
390	if (copy_from_user(&c, arg, sizeof(c)))
391	return -EFAULT;
392
393	if (c.flags)
394	return -EOPNOTSUPP;
395	return ops->pr_clear(bdev, c.key);
396	}
397
07d106d0 LT	398	/*
	399	* Is it an unrecognized ioctl? The correct returns are either
	400	* ENOTTY (final) or ENOIOCTLCMD ("I don't know this one, try a
	401	* fallback"). ENOIOCTLCMD gets turned into ENOTTY by the ioctl
	402	* code before returning.
	403	*
	404	* Confused drivers sometimes return EINVAL, which is wrong. It
	405	* means "I understood the ioctl command, but the parameters to
	406	* it were wrong".
	407	*
	408	* We should aim to just fix the broken drivers, the EINVAL case
	409	* should go away.
	410	*/
	411	static inline int is_unrecognized_ioctl(int ret)
	412	{
	413	return ret == -EINVAL \|\|
	414	ret == -ENOTTY \|\|
	415	ret == -ENOIOCTLCMD;
	416	}
	417
d8e4bb81 CH	418	static int blkdev_flushbuf(struct block_device *bdev, fmode_t mode,
d8e4bb81 CH	419	unsigned cmd, unsigned long arg)
bb93e3a5	420	{
d8e4bb81	421	int ret;
bb93e3a5	422
d8e4bb81 CH	423	if (!capable(CAP_SYS_ADMIN))
d8e4bb81 CH	424	return -EACCES;
bb93e3a5	425
d8e4bb81 CH	426	ret = __blkdev_driver_ioctl(bdev, mode, cmd, arg);
	427	if (!is_unrecognized_ioctl(ret))
	428	return ret;
bb93e3a5	429
d8e4bb81 CH	430	fsync_bdev(bdev);
	431	invalidate_bdev(bdev);
	432	return 0;
	433	}
d30a2605	434
d8e4bb81 CH	435	static int blkdev_roset(struct block_device *bdev, fmode_t mode,
	436	unsigned cmd, unsigned long arg)
	437	{
	438	int ret, n;
d30a2605	439
d8e4bb81 CH	440	ret = __blkdev_driver_ioctl(bdev, mode, cmd, arg);
	441	if (!is_unrecognized_ioctl(ret))
	442	return ret;
	443	if (!capable(CAP_SYS_ADMIN))
	444	return -EACCES;
	445	if (get_user(n, (int __user *)arg))
	446	return -EFAULT;
	447	set_device_ro(bdev, n);
	448	return 0;
	449	}
d30a2605	450
d8e4bb81 CH	451	static int blkdev_getgeo(struct block_device *bdev,
	452	struct hd_geometry __user *argp)
	453	{
	454	struct gendisk *disk = bdev->bd_disk;
	455	struct hd_geometry geo;
	456	int ret;
d30a2605	457
d8e4bb81 CH	458	if (!argp)
	459	return -EINVAL;
	460	if (!disk->fops->getgeo)
	461	return -ENOTTY;
	462
	463	/*
	464	* We need to set the startsect first, the driver may
	465	* want to override it.
	466	*/
	467	memset(&geo, 0, sizeof(geo));
	468	geo.start = get_start_sect(bdev);
	469	ret = disk->fops->getgeo(bdev, &geo);
	470	if (ret)
	471	return ret;
	472	if (copy_to_user(argp, &geo, sizeof(geo)))
	473	return -EFAULT;
	474	return 0;
	475	}
66ba32dc	476
d8e4bb81 CH	477	/* set the logical block size */
	478	static int blkdev_bszset(struct block_device *bdev, fmode_t mode,
	479	int __user *argp)
	480	{
	481	int ret, n;
66ba32dc	482
d8e4bb81 CH	483	if (!capable(CAP_SYS_ADMIN))
	484	return -EACCES;
	485	if (!argp)
	486	return -EINVAL;
	487	if (get_user(n, argp))
	488	return -EFAULT;
66ba32dc	489
d8e4bb81 CH	490	if (!(mode & FMODE_EXCL)) {
	491	bdgrab(bdev);
	492	if (blkdev_get(bdev, mode \| FMODE_EXCL, &bdev) < 0)
	493	return -EBUSY;
66ba32dc	494	}
d30a2605	495
d8e4bb81 CH	496	ret = set_blocksize(bdev, n);
	497	if (!(mode & FMODE_EXCL))
	498	blkdev_put(bdev, mode \| FMODE_EXCL);
	499	return ret;
	500	}
a885c8c4	501
d8e4bb81 CH	502	/*
	503	* always keep this in sync with compat_blkdev_ioctl()
	504	*/
	505	int blkdev_ioctl(struct block_device *bdev, fmode_t mode, unsigned cmd,
	506	unsigned long arg)
	507	{
d8e4bb81 CH	508	void __user argp = (void __user )arg;
	509	loff_t size;
	510	unsigned int max_sectors;
	511
	512	switch (cmd) {
	513	case BLKFLSBUF:
	514	return blkdev_flushbuf(bdev, mode, cmd, arg);
	515	case BLKROSET:
	516	return blkdev_roset(bdev, mode, cmd, arg);
	517	case BLKDISCARD:
	518	return blk_ioctl_discard(bdev, mode, arg, 0);
	519	case BLKSECDISCARD:
	520	return blk_ioctl_discard(bdev, mode, arg,
	521	BLKDEV_DISCARD_SECURE);
	522	case BLKZEROOUT:
	523	return blk_ioctl_zeroout(bdev, mode, arg);
3ed05a98 ST	524	case BLKREPORTZONE:
	525	return blkdev_report_zones_ioctl(bdev, mode, cmd, arg);
	526	case BLKRESETZONE:
	527	return blkdev_reset_zones_ioctl(bdev, mode, cmd, arg);
d8e4bb81 CH	528	case HDIO_GETGEO:
d8e4bb81 CH	529	return blkdev_getgeo(bdev, argp);
45048d09 AV	530	case BLKRAGET:
	531	case BLKFRAGET:
	532	if (!arg)
	533	return -EINVAL;
efa7c9f9	534	return put_long(arg, (bdev->bd_bdi->ra_pages*PAGE_SIZE) / 512);
45048d09 AV	535	case BLKROGET:
45048d09 AV	536	return put_int(arg, bdev_read_only(bdev) != 0);
ac481c20	537	case BLKBSZGET: /* get block device soft block size (cf. BLKSSZGET) */
45048d09	538	return put_int(arg, block_size(bdev));
ac481c20	539	case BLKSSZGET: /* get block device logical block size */
e1defc4f	540	return put_int(arg, bdev_logical_block_size(bdev));
ac481c20 MP	541	case BLKPBSZGET: /* get block device physical block size */
	542	return put_uint(arg, bdev_physical_block_size(bdev));
	543	case BLKIOMIN:
	544	return put_uint(arg, bdev_io_min(bdev));
	545	case BLKIOOPT:
	546	return put_uint(arg, bdev_io_opt(bdev));
	547	case BLKALIGNOFF:
	548	return put_int(arg, bdev_alignment_offset(bdev));
98262f27 MP	549	case BLKDISCARDZEROES:
98262f27 MP	550	return put_uint(arg, bdev_discard_zeroes_data(bdev));
45048d09	551	case BLKSECTGET:
63f26496 AM	552	max_sectors = min_t(unsigned int, USHRT_MAX,
	553	queue_max_sectors(bdev_get_queue(bdev)));
	554	return put_ushort(arg, max_sectors);
ef00f59c MP	555	case BLKROTATIONAL:
ef00f59c MP	556	return put_ushort(arg, !blk_queue_nonrot(bdev_get_queue(bdev)));
45048d09 AV	557	case BLKRASET:
	558	case BLKFRASET:
	559	if(!capable(CAP_SYS_ADMIN))
	560	return -EACCES;
efa7c9f9	561	bdev->bd_bdi->ra_pages = (arg * 512) / PAGE_SIZE;
45048d09 AV	562	return 0;
45048d09 AV	563	case BLKBSZSET:
d8e4bb81	564	return blkdev_bszset(bdev, mode, argp);
45048d09	565	case BLKPG:
d8e4bb81	566	return blkpg_ioctl(bdev, argp);
45048d09	567	case BLKRRPART:
d8e4bb81	568	return blkdev_reread_part(bdev);
45048d09	569	case BLKGETSIZE:
77304d2a	570	size = i_size_read(bdev->bd_inode);
45048d09 AV	571	if ((size >> 9) > ~0UL)
	572	return -EFBIG;
	573	return put_ulong(arg, size >> 9);
	574	case BLKGETSIZE64:
77304d2a	575	return put_u64(arg, i_size_read(bdev->bd_inode));
45048d09 AV	576	case BLKTRACESTART:
	577	case BLKTRACESTOP:
	578	case BLKTRACESETUP:
	579	case BLKTRACETEARDOWN:
d8e4bb81	580	return blk_trace_ioctl(bdev, cmd, argp);
bbd3e064 CH	581	case IOC_PR_REGISTER:
	582	return blkdev_pr_register(bdev, argp);
	583	case IOC_PR_RESERVE:
	584	return blkdev_pr_reserve(bdev, argp);
	585	case IOC_PR_RELEASE:
	586	return blkdev_pr_release(bdev, argp);
	587	case IOC_PR_PREEMPT:
	588	return blkdev_pr_preempt(bdev, argp, false);
	589	case IOC_PR_PREEMPT_ABORT:
	590	return blkdev_pr_preempt(bdev, argp, true);
	591	case IOC_PR_CLEAR:
	592	return blkdev_pr_clear(bdev, argp);
45048d09	593	default:
d8e4bb81	594	return __blkdev_driver_ioctl(bdev, mode, cmd, arg);
45048d09	595	}
1da177e4	596	}
68f66feb	597	EXPORT_SYMBOL_GPL(blkdev_ioctl);