[linux.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;
	struct request_queue *q = bdev_get_queue(bdev);
	struct address_space *mapping = bdev->bd_inode->i_mapping;


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

	if (!blk_queue_discard(q))
		return -EOPNOTSUPP;

	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;

	if (start + len > i_size_read(bdev->bd_inode))
		return -EINVAL;
	truncate_inode_pages_range(mapping, start, start + len);
	return blkdev_issue_discard(bdev, start >> 9, len >> 9,
				    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,
			BLKDEV_ZERO_NOUNMAP);
}

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;

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

	ret = __blkdev_driver_ioctl(bdev, mode, cmd, arg);
	if (!is_unrecognized_ioctl(ret))
		return ret;
	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, 0);
	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];
d8e4bb81 CH	204	uint64_t start, len;
351499a1 DM	205	struct request_queue *q = bdev_get_queue(bdev);
	206	struct address_space *mapping = bdev->bd_inode->i_mapping;
	207
d8e4bb81 CH	208
	209	if (!(mode & FMODE_WRITE))
	210	return -EBADF;
	211
351499a1 DM	212	if (!blk_queue_discard(q))
	213	return -EOPNOTSUPP;
	214
d8e4bb81 CH	215	if (copy_from_user(range, (void __user *)arg, sizeof(range)))
	216	return -EFAULT;
	217
	218	start = range[0];
	219	len = range[1];
8d57a98c	220
d30a2605 DW	221	if (start & 511)
	222	return -EINVAL;
	223	if (len & 511)
	224	return -EINVAL;
d30a2605	225
351499a1	226	if (start + len > i_size_read(bdev->bd_inode))
d30a2605	227	return -EINVAL;
351499a1 DM	228	truncate_inode_pages_range(mapping, start, start + len);
	229	return blkdev_issue_discard(bdev, start >> 9, len >> 9,
	230	GFP_KERNEL, flags);
d30a2605 DW	231	}
d30a2605 DW	232
d8e4bb81 CH	233	static int blk_ioctl_zeroout(struct block_device *bdev, fmode_t mode,
d8e4bb81 CH	234	unsigned long arg)
66ba32dc	235	{
d8e4bb81	236	uint64_t range[2];
22dd6d35 DW	237	struct address_space *mapping;
22dd6d35 DW	238	uint64_t start, end, len;
d8e4bb81 CH	239
	240	if (!(mode & FMODE_WRITE))
	241	return -EBADF;
	242
	243	if (copy_from_user(range, (void __user *)arg, sizeof(range)))
	244	return -EFAULT;
	245
	246	start = range[0];
	247	len = range[1];
22dd6d35	248	end = start + len - 1;
d8e4bb81	249
66ba32dc MP	250	if (start & 511)
	251	return -EINVAL;
	252	if (len & 511)
	253	return -EINVAL;
22dd6d35 DW	254	if (end >= (uint64_t)i_size_read(bdev->bd_inode))
	255	return -EINVAL;
	256	if (end < start)
66ba32dc MP	257	return -EINVAL;
66ba32dc MP	258
22dd6d35 DW	259	/* Invalidate the page cache, including dirty pages */
	260	mapping = bdev->bd_inode->i_mapping;
	261	truncate_inode_pages_range(mapping, start, end);
	262
	263	return blkdev_issue_zeroout(bdev, start >> 9, len >> 9, GFP_KERNEL,
ee472d83	264	BLKDEV_ZERO_NOUNMAP);
66ba32dc MP	265	}
66ba32dc MP	266
1da177e4 LT	267	static int put_ushort(unsigned long arg, unsigned short val)
	268	{
	269	return put_user(val, (unsigned short __user *)arg);
	270	}
	271
	272	static int put_int(unsigned long arg, int val)
	273	{
	274	return put_user(val, (int __user *)arg);
	275	}
	276
ac481c20 MP	277	static int put_uint(unsigned long arg, unsigned int val)
	278	{
	279	return put_user(val, (unsigned int __user *)arg);
	280	}
	281
1da177e4 LT	282	static int put_long(unsigned long arg, long val)
	283	{
	284	return put_user(val, (long __user *)arg);
	285	}
	286
	287	static int put_ulong(unsigned long arg, unsigned long val)
	288	{
	289	return put_user(val, (unsigned long __user *)arg);
	290	}
	291
	292	static int put_u64(unsigned long arg, u64 val)
	293	{
	294	return put_user(val, (u64 __user *)arg);
	295	}
	296
633a08b8 AV	297	int __blkdev_driver_ioctl(struct block_device *bdev, fmode_t mode,
	298	unsigned cmd, unsigned long arg)
	299	{
	300	struct gendisk *disk = bdev->bd_disk;
d4430d62 AV	301
	302	if (disk->fops->ioctl)
	303	return disk->fops->ioctl(bdev, mode, cmd, arg);
633a08b8	304
633a08b8 AV	305	return -ENOTTY;
	306	}
	307	/*
	308	* For the record: _GPL here is only because somebody decided to slap it
	309	* on the previous export. Sheer idiocy, since it wasn't copyrightable
	310	* at all and could be open-coded without any exports by anybody who cares.
	311	*/
	312	EXPORT_SYMBOL_GPL(__blkdev_driver_ioctl);
	313
bbd3e064 CH	314	static int blkdev_pr_register(struct block_device *bdev,
	315	struct pr_registration __user *arg)
	316	{
	317	const struct pr_ops *ops = bdev->bd_disk->fops->pr_ops;
	318	struct pr_registration reg;
	319
	320	if (!capable(CAP_SYS_ADMIN))
	321	return -EPERM;
	322	if (!ops \|\| !ops->pr_register)
	323	return -EOPNOTSUPP;
	324	if (copy_from_user(&reg, arg, sizeof(reg)))
	325	return -EFAULT;
	326
	327	if (reg.flags & ~PR_FL_IGNORE_KEY)
	328	return -EOPNOTSUPP;
	329	return ops->pr_register(bdev, reg.old_key, reg.new_key, reg.flags);
	330	}
	331
	332	static int blkdev_pr_reserve(struct block_device *bdev,
	333	struct pr_reservation __user *arg)
	334	{
	335	const struct pr_ops *ops = bdev->bd_disk->fops->pr_ops;
	336	struct pr_reservation rsv;
	337
	338	if (!capable(CAP_SYS_ADMIN))
	339	return -EPERM;
	340	if (!ops \|\| !ops->pr_reserve)
	341	return -EOPNOTSUPP;
	342	if (copy_from_user(&rsv, arg, sizeof(rsv)))
	343	return -EFAULT;
	344
	345	if (rsv.flags & ~PR_FL_IGNORE_KEY)
	346	return -EOPNOTSUPP;
	347	return ops->pr_reserve(bdev, rsv.key, rsv.type, rsv.flags);
	348	}
	349
	350	static int blkdev_pr_release(struct block_device *bdev,
	351	struct pr_reservation __user *arg)
	352	{
	353	const struct pr_ops *ops = bdev->bd_disk->fops->pr_ops;
	354	struct pr_reservation rsv;
	355
	356	if (!capable(CAP_SYS_ADMIN))
	357	return -EPERM;
	358	if (!ops \|\| !ops->pr_release)
	359	return -EOPNOTSUPP;
	360	if (copy_from_user(&rsv, arg, sizeof(rsv)))
	361	return -EFAULT;
	362
	363	if (rsv.flags)
	364	return -EOPNOTSUPP;
	365	return ops->pr_release(bdev, rsv.key, rsv.type);
	366	}
	367
	368	static int blkdev_pr_preempt(struct block_device *bdev,
	369	struct pr_preempt __user *arg, bool abort)
	370	{
	371	const struct pr_ops *ops = bdev->bd_disk->fops->pr_ops;
	372	struct pr_preempt p;
	373
	374	if (!capable(CAP_SYS_ADMIN))
	375	return -EPERM;
	376	if (!ops \|\| !ops->pr_preempt)
	377	return -EOPNOTSUPP;
378	if (copy_from_user(&p, arg, sizeof(p)))
379	return -EFAULT;
380
381	if (p.flags)
382	return -EOPNOTSUPP;
383	return ops->pr_preempt(bdev, p.old_key, p.new_key, p.type, abort);
384	}
385
386	static int blkdev_pr_clear(struct block_device *bdev,
387	struct pr_clear __user *arg)
388	{
389	const struct pr_ops *ops = bdev->bd_disk->fops->pr_ops;
390	struct pr_clear c;
391
392	if (!capable(CAP_SYS_ADMIN))
393	return -EPERM;
394	if (!ops \|\| !ops->pr_clear)
395	return -EOPNOTSUPP;
396	if (copy_from_user(&c, arg, sizeof(c)))
397	return -EFAULT;
398
399	if (c.flags)
400	return -EOPNOTSUPP;
401	return ops->pr_clear(bdev, c.key);
402	}
403
07d106d0 LT	404	/*
	405	* Is it an unrecognized ioctl? The correct returns are either
	406	* ENOTTY (final) or ENOIOCTLCMD ("I don't know this one, try a
	407	* fallback"). ENOIOCTLCMD gets turned into ENOTTY by the ioctl
	408	* code before returning.
	409	*
	410	* Confused drivers sometimes return EINVAL, which is wrong. It
	411	* means "I understood the ioctl command, but the parameters to
	412	* it were wrong".
	413	*
	414	* We should aim to just fix the broken drivers, the EINVAL case
	415	* should go away.
	416	*/
	417	static inline int is_unrecognized_ioctl(int ret)
	418	{
	419	return ret == -EINVAL \|\|
	420	ret == -ENOTTY \|\|
	421	ret == -ENOIOCTLCMD;
	422	}
	423
d8e4bb81 CH	424	static int blkdev_flushbuf(struct block_device *bdev, fmode_t mode,
d8e4bb81 CH	425	unsigned cmd, unsigned long arg)
bb93e3a5	426	{
d8e4bb81	427	int ret;
bb93e3a5	428
d8e4bb81 CH	429	if (!capable(CAP_SYS_ADMIN))
d8e4bb81 CH	430	return -EACCES;
bb93e3a5	431
d8e4bb81 CH	432	ret = __blkdev_driver_ioctl(bdev, mode, cmd, arg);
	433	if (!is_unrecognized_ioctl(ret))
	434	return ret;
bb93e3a5	435
d8e4bb81 CH	436	fsync_bdev(bdev);
	437	invalidate_bdev(bdev);
	438	return 0;
	439	}
d30a2605	440
d8e4bb81 CH	441	static int blkdev_roset(struct block_device *bdev, fmode_t mode,
	442	unsigned cmd, unsigned long arg)
	443	{
	444	int ret, n;
d30a2605	445
bb749b31 ID	446	if (!capable(CAP_SYS_ADMIN))
	447	return -EACCES;
	448
d8e4bb81 CH	449	ret = __blkdev_driver_ioctl(bdev, mode, cmd, arg);
	450	if (!is_unrecognized_ioctl(ret))
	451	return ret;
d8e4bb81 CH	452	if (get_user(n, (int __user *)arg))
	453	return -EFAULT;
	454	set_device_ro(bdev, n);
	455	return 0;
	456	}
d30a2605	457
d8e4bb81 CH	458	static int blkdev_getgeo(struct block_device *bdev,
	459	struct hd_geometry __user *argp)
	460	{
	461	struct gendisk *disk = bdev->bd_disk;
	462	struct hd_geometry geo;
	463	int ret;
d30a2605	464
d8e4bb81 CH	465	if (!argp)
	466	return -EINVAL;
	467	if (!disk->fops->getgeo)
	468	return -ENOTTY;
	469
	470	/*
	471	* We need to set the startsect first, the driver may
	472	* want to override it.
	473	*/
	474	memset(&geo, 0, sizeof(geo));
	475	geo.start = get_start_sect(bdev);
	476	ret = disk->fops->getgeo(bdev, &geo);
	477	if (ret)
	478	return ret;
	479	if (copy_to_user(argp, &geo, sizeof(geo)))
	480	return -EFAULT;
	481	return 0;
	482	}
66ba32dc	483
d8e4bb81 CH	484	/* set the logical block size */
	485	static int blkdev_bszset(struct block_device *bdev, fmode_t mode,
	486	int __user *argp)
	487	{
	488	int ret, n;
66ba32dc	489
d8e4bb81 CH	490	if (!capable(CAP_SYS_ADMIN))
	491	return -EACCES;
	492	if (!argp)
	493	return -EINVAL;
	494	if (get_user(n, argp))
	495	return -EFAULT;
66ba32dc	496
d8e4bb81 CH	497	if (!(mode & FMODE_EXCL)) {
	498	bdgrab(bdev);
	499	if (blkdev_get(bdev, mode \| FMODE_EXCL, &bdev) < 0)
	500	return -EBUSY;
66ba32dc	501	}
d30a2605	502
d8e4bb81 CH	503	ret = set_blocksize(bdev, n);
	504	if (!(mode & FMODE_EXCL))
	505	blkdev_put(bdev, mode \| FMODE_EXCL);
	506	return ret;
	507	}
a885c8c4	508
d8e4bb81 CH	509	/*
	510	* always keep this in sync with compat_blkdev_ioctl()
	511	*/
	512	int blkdev_ioctl(struct block_device *bdev, fmode_t mode, unsigned cmd,
	513	unsigned long arg)
	514	{
d8e4bb81 CH	515	void __user argp = (void __user )arg;
	516	loff_t size;
	517	unsigned int max_sectors;
	518
	519	switch (cmd) {
	520	case BLKFLSBUF:
	521	return blkdev_flushbuf(bdev, mode, cmd, arg);
	522	case BLKROSET:
	523	return blkdev_roset(bdev, mode, cmd, arg);
	524	case BLKDISCARD:
	525	return blk_ioctl_discard(bdev, mode, arg, 0);
	526	case BLKSECDISCARD:
	527	return blk_ioctl_discard(bdev, mode, arg,
	528	BLKDEV_DISCARD_SECURE);
	529	case BLKZEROOUT:
	530	return blk_ioctl_zeroout(bdev, mode, arg);
3ed05a98 ST	531	case BLKREPORTZONE:
	532	return blkdev_report_zones_ioctl(bdev, mode, cmd, arg);
	533	case BLKRESETZONE:
	534	return blkdev_reset_zones_ioctl(bdev, mode, cmd, arg);
d8e4bb81 CH	535	case HDIO_GETGEO:
d8e4bb81 CH	536	return blkdev_getgeo(bdev, argp);
45048d09 AV	537	case BLKRAGET:
	538	case BLKFRAGET:
	539	if (!arg)
	540	return -EINVAL;
efa7c9f9	541	return put_long(arg, (bdev->bd_bdi->ra_pages*PAGE_SIZE) / 512);
45048d09 AV	542	case BLKROGET:
45048d09 AV	543	return put_int(arg, bdev_read_only(bdev) != 0);
ac481c20	544	case BLKBSZGET: /* get block device soft block size (cf. BLKSSZGET) */
45048d09	545	return put_int(arg, block_size(bdev));
ac481c20	546	case BLKSSZGET: /* get block device logical block size */
e1defc4f	547	return put_int(arg, bdev_logical_block_size(bdev));
ac481c20 MP	548	case BLKPBSZGET: /* get block device physical block size */
	549	return put_uint(arg, bdev_physical_block_size(bdev));
	550	case BLKIOMIN:
	551	return put_uint(arg, bdev_io_min(bdev));
	552	case BLKIOOPT:
	553	return put_uint(arg, bdev_io_opt(bdev));
	554	case BLKALIGNOFF:
	555	return put_int(arg, bdev_alignment_offset(bdev));
98262f27	556	case BLKDISCARDZEROES:
48920ff2	557	return put_uint(arg, 0);
45048d09	558	case BLKSECTGET:
63f26496 AM	559	max_sectors = min_t(unsigned int, USHRT_MAX,
	560	queue_max_sectors(bdev_get_queue(bdev)));
	561	return put_ushort(arg, max_sectors);
ef00f59c MP	562	case BLKROTATIONAL:
ef00f59c MP	563	return put_ushort(arg, !blk_queue_nonrot(bdev_get_queue(bdev)));
45048d09 AV	564	case BLKRASET:
	565	case BLKFRASET:
	566	if(!capable(CAP_SYS_ADMIN))
	567	return -EACCES;
efa7c9f9	568	bdev->bd_bdi->ra_pages = (arg * 512) / PAGE_SIZE;
45048d09 AV	569	return 0;
45048d09 AV	570	case BLKBSZSET:
d8e4bb81	571	return blkdev_bszset(bdev, mode, argp);
45048d09	572	case BLKPG:
d8e4bb81	573	return blkpg_ioctl(bdev, argp);
45048d09	574	case BLKRRPART:
d8e4bb81	575	return blkdev_reread_part(bdev);
45048d09	576	case BLKGETSIZE:
77304d2a	577	size = i_size_read(bdev->bd_inode);
45048d09 AV	578	if ((size >> 9) > ~0UL)
	579	return -EFBIG;
	580	return put_ulong(arg, size >> 9);
	581	case BLKGETSIZE64:
77304d2a	582	return put_u64(arg, i_size_read(bdev->bd_inode));
45048d09 AV	583	case BLKTRACESTART:
	584	case BLKTRACESTOP:
	585	case BLKTRACESETUP:
	586	case BLKTRACETEARDOWN:
d8e4bb81	587	return blk_trace_ioctl(bdev, cmd, argp);
bbd3e064 CH	588	case IOC_PR_REGISTER:
	589	return blkdev_pr_register(bdev, argp);
	590	case IOC_PR_RESERVE:
	591	return blkdev_pr_reserve(bdev, argp);
	592	case IOC_PR_RELEASE:
	593	return blkdev_pr_release(bdev, argp);
	594	case IOC_PR_PREEMPT:
	595	return blkdev_pr_preempt(bdev, argp, false);
	596	case IOC_PR_PREEMPT_ABORT:
	597	return blkdev_pr_preempt(bdev, argp, true);
	598	case IOC_PR_CLEAR:
	599	return blkdev_pr_clear(bdev, argp);
45048d09	600	default:
d8e4bb81	601	return __blkdev_driver_ioctl(bdev, mode, cmd, arg);
45048d09	602	}
1da177e4	603	}
68f66feb	604	EXPORT_SYMBOL_GPL(blkdev_ioctl);