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

// SPDX-License-Identifier: GPL-2.0
#include <linux/capability.h>
#include <linux/compat.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>
#include "blk.h"

static int blkpg_do_ioctl(struct block_device *bdev,
			  struct blkpg_partition __user *upart, int op)
{
	struct blkpg_partition p;
	long long start, length;

	if (!capable(CAP_SYS_ADMIN))
		return -EACCES;
	if (copy_from_user(&p, upart, sizeof(struct blkpg_partition)))
		return -EFAULT;
	if (bdev_is_partition(bdev))
		return -EINVAL;

	if (p.pno <= 0)
		return -EINVAL;

	if (op == BLKPG_DEL_PARTITION)
		return bdev_del_partition(bdev, p.pno);

	start = p.start >> SECTOR_SHIFT;
	length = p.length >> SECTOR_SHIFT;

	switch (op) {
	case BLKPG_ADD_PARTITION:
		/* check if partition is aligned to blocksize */
		if (p.start & (bdev_logical_block_size(bdev) - 1))
			return -EINVAL;
		return bdev_add_partition(bdev, p.pno, start, length);
	case BLKPG_RESIZE_PARTITION:
		return bdev_resize_partition(bdev, p.pno, start, length);
	default:
		return -EINVAL;
	}
}

static int blkpg_ioctl(struct block_device *bdev,
		       struct blkpg_ioctl_arg __user *arg)
{
	struct blkpg_partition __user *udata;
	int op;

	if (get_user(op, &arg->op) || get_user(udata, &arg->data))
		return -EFAULT;

	return blkpg_do_ioctl(bdev, udata, op);
}

#ifdef CONFIG_COMPAT
struct compat_blkpg_ioctl_arg {
	compat_int_t op;
	compat_int_t flags;
	compat_int_t datalen;
	compat_caddr_t data;
};

static int compat_blkpg_ioctl(struct block_device *bdev,
			      struct compat_blkpg_ioctl_arg __user *arg)
{
	compat_caddr_t udata;
	int op;

	if (get_user(op, &arg->op) || get_user(udata, &arg->data))
		return -EFAULT;

	return blkpg_do_ioctl(bdev, compat_ptr(udata), op);
}
#endif

static int blkdev_reread_part(struct block_device *bdev)
{
	int ret;

	if (!disk_part_scan_enabled(bdev->bd_disk) || bdev_is_partition(bdev))
		return -EINVAL;
	if (!capable(CAP_SYS_ADMIN))
		return -EACCES;

	mutex_lock(&bdev->bd_mutex);
	ret = bdev_disk_changed(bdev, false);
	mutex_unlock(&bdev->bd_mutex);

	return ret;
}

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);
	int err;

	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;

	err = truncate_bdev_range(bdev, mode, start, start + len - 1);
	if (err)
		return err;

	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];
	uint64_t start, end, len;
	int err;

	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 */
	err = truncate_bdev_range(bdev, mode, start, end);
	if (err)
		return err;

	return blkdev_issue_zeroout(bdev, start >> 9, len >> 9, GFP_KERNEL,
			BLKDEV_ZERO_NOUNMAP);
}

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

static int put_int(int __user *argp, int val)
{
	return put_user(val, argp);
}

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

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

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

static int put_u64(u64 __user *argp, u64 val)
{
	return put_user(val, argp);
}

#ifdef CONFIG_COMPAT
static int compat_put_long(compat_long_t __user *argp, long val)
{
	return put_user(val, argp);
}

static int compat_put_ulong(compat_ulong_t __user *argp, compat_ulong_t val)
{
	return put_user(val, argp);
}
#endif

#ifdef CONFIG_COMPAT
/*
 * This is the equivalent of compat_ptr_ioctl(), to be used by block
 * drivers that implement only commands that are completely compatible
 * between 32-bit and 64-bit user space
 */
int blkdev_compat_ptr_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,
					 (unsigned long)compat_ptr(arg));

	return -ENOIOCTLCMD;
}
EXPORT_SYMBOL(blkdev_compat_ptr_ioctl);
#endif

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);
}

static int blkdev_flushbuf(struct block_device *bdev, fmode_t mode,
		unsigned cmd, unsigned long arg)
{
	if (!capable(CAP_SYS_ADMIN))
		return -EACCES;
	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;

	if (get_user(n, (int __user *)arg))
		return -EFAULT;
	if (bdev->bd_disk->fops->set_read_only) {
		ret = bdev->bd_disk->fops->set_read_only(bdev, n);
		if (ret)
			return ret;
	}
	bdev->bd_read_only = 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;
}

#ifdef CONFIG_COMPAT
struct compat_hd_geometry {
	unsigned char heads;
	unsigned char sectors;
	unsigned short cylinders;
	u32 start;
};

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

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

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

	ret = copy_to_user(ugeo, &geo, 4);
	ret |= put_user(geo.start, &ugeo->start);
	if (ret)
		ret = -EFAULT;

	return ret;
}
#endif

/* 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)
		return set_blocksize(bdev, n);

	if (IS_ERR(blkdev_get_by_dev(bdev->bd_dev, mode | FMODE_EXCL, &bdev)))
		return -EBUSY;
	ret = set_blocksize(bdev, n);
	blkdev_put(bdev, mode | FMODE_EXCL);

	return ret;
}

/*
 * Common commands that are handled the same way on native and compat
 * user space. Note the separate arg/argp parameters that are needed
 * to deal with the compat_ptr() conversion.
 */
static int blkdev_common_ioctl(struct block_device *bdev, fmode_t mode,
				unsigned cmd, unsigned long arg, void __user *argp)
{
	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:
	case BLKOPENZONE:
	case BLKCLOSEZONE:
	case BLKFINISHZONE:
		return blkdev_zone_mgmt_ioctl(bdev, mode, cmd, arg);
	case BLKGETZONESZ:
		return put_uint(argp, bdev_zone_sectors(bdev));
	case BLKGETNRZONES:
		return put_uint(argp, blkdev_nr_zones(bdev->bd_disk));
	case BLKROGET:
		return put_int(argp, bdev_read_only(bdev) != 0);
	case BLKSSZGET: /* get block device logical block size */
		return put_int(argp, bdev_logical_block_size(bdev));
	case BLKPBSZGET: /* get block device physical block size */
		return put_uint(argp, bdev_physical_block_size(bdev));
	case BLKIOMIN:
		return put_uint(argp, bdev_io_min(bdev));
	case BLKIOOPT:
		return put_uint(argp, bdev_io_opt(bdev));
	case BLKALIGNOFF:
		return put_int(argp, bdev_alignment_offset(bdev));
	case BLKDISCARDZEROES:
		return put_uint(argp, 0);
	case BLKSECTGET:
		max_sectors = min_t(unsigned int, USHRT_MAX,
				    queue_max_sectors(bdev_get_queue(bdev)));
		return put_ushort(argp, max_sectors);
	case BLKROTATIONAL:
		return put_ushort(argp, !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 BLKRRPART:
		return blkdev_reread_part(bdev);
	case BLKTRACESTART:
	case BLKTRACESTOP:
	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 -ENOIOCTLCMD;
	}
}

/*
 * Always keep this in sync with compat_blkdev_ioctl()
 * to handle all incompatible commands in both functions.
 *
 * New commands must be compatible and go into blkdev_common_ioctl
 */
int blkdev_ioctl(struct block_device *bdev, fmode_t mode, unsigned cmd,
			unsigned long arg)
{
	int ret;
	loff_t size;
	void __user *argp = (void __user *)arg;

	switch (cmd) {
	/* These need separate implementations for the data structure */
	case HDIO_GETGEO:
		return blkdev_getgeo(bdev, argp);
	case BLKPG:
		return blkpg_ioctl(bdev, argp);

	/* Compat mode returns 32-bit data instead of 'long' */
	case BLKRAGET:
	case BLKFRAGET:
		if (!argp)
			return -EINVAL;
		return put_long(argp, (bdev->bd_bdi->ra_pages*PAGE_SIZE) / 512);
	case BLKGETSIZE:
		size = i_size_read(bdev->bd_inode);
		if ((size >> 9) > ~0UL)
			return -EFBIG;
		return put_ulong(argp, size >> 9);

	/* The data is compatible, but the command number is different */
	case BLKBSZGET: /* get block device soft block size (cf. BLKSSZGET) */
		return put_int(argp, block_size(bdev));
	case BLKBSZSET:
		return blkdev_bszset(bdev, mode, argp);
	case BLKGETSIZE64:
		return put_u64(argp, i_size_read(bdev->bd_inode));

	/* Incompatible alignment on i386 */
	case BLKTRACESETUP:
		return blk_trace_ioctl(bdev, cmd, argp);
	default:
		break;
	}

	ret = blkdev_common_ioctl(bdev, mode, cmd, arg, argp);
	if (ret != -ENOIOCTLCMD)
		return ret;

	if (!bdev->bd_disk->fops->ioctl)
		return -ENOTTY;
	return bdev->bd_disk->fops->ioctl(bdev, mode, cmd, arg);
}
EXPORT_SYMBOL_GPL(blkdev_ioctl); /* for /dev/raw */

#ifdef CONFIG_COMPAT

#define BLKBSZGET_32		_IOR(0x12, 112, int)
#define BLKBSZSET_32		_IOW(0x12, 113, int)
#define BLKGETSIZE64_32		_IOR(0x12, 114, int)

/* Most of the generic ioctls are handled in the normal fallback path.
   This assumes the blkdev's low level compat_ioctl always returns
   ENOIOCTLCMD for unknown ioctls. */
long compat_blkdev_ioctl(struct file *file, unsigned cmd, unsigned long arg)
{
	int ret;
	void __user *argp = compat_ptr(arg);
	struct block_device *bdev = I_BDEV(file->f_mapping->host);
	struct gendisk *disk = bdev->bd_disk;
	fmode_t mode = file->f_mode;
	loff_t size;

	/*
	 * O_NDELAY can be altered using fcntl(.., F_SETFL, ..), so we have
	 * to updated it before every ioctl.
	 */
	if (file->f_flags & O_NDELAY)
		mode |= FMODE_NDELAY;
	else
		mode &= ~FMODE_NDELAY;

	switch (cmd) {
	/* These need separate implementations for the data structure */
	case HDIO_GETGEO:
		return compat_hdio_getgeo(bdev, argp);
	case BLKPG:
		return compat_blkpg_ioctl(bdev, argp);

	/* Compat mode returns 32-bit data instead of 'long' */
	case BLKRAGET:
	case BLKFRAGET:
		if (!argp)
			return -EINVAL;
		return compat_put_long(argp,
			       (bdev->bd_bdi->ra_pages * PAGE_SIZE) / 512);
	case BLKGETSIZE:
		size = i_size_read(bdev->bd_inode);
		if ((size >> 9) > ~0UL)
			return -EFBIG;
		return compat_put_ulong(argp, size >> 9);

	/* The data is compatible, but the command number is different */
	case BLKBSZGET_32: /* get the logical block size (cf. BLKSSZGET) */
		return put_int(argp, bdev_logical_block_size(bdev));
	case BLKBSZSET_32:
		return blkdev_bszset(bdev, mode, argp);
	case BLKGETSIZE64_32:
		return put_u64(argp, i_size_read(bdev->bd_inode));

	/* Incompatible alignment on i386 */
	case BLKTRACESETUP32:
		return blk_trace_ioctl(bdev, cmd, argp);
	default:
		break;
	}

	ret = blkdev_common_ioctl(bdev, mode, cmd, arg, argp);
	if (ret == -ENOIOCTLCMD && disk->fops->compat_ioctl)
		ret = disk->fops->compat_ioctl(bdev, mode, cmd, arg);

	return ret;
}
#endif
Commit	Line	Data
3dcf60bc	1	// SPDX-License-Identifier: GPL-2.0
c59ede7b	2	#include <linux/capability.h>
ee6a129d	3	#include <linux/compat.h>
1da177e4	4	#include <linux/blkdev.h>
d5decd3b	5	#include <linux/export.h>
5a0e3ad6	6	#include <linux/gfp.h>
1da177e4	7	#include <linux/blkpg.h>
a885c8c4	8	#include <linux/hdreg.h>
1da177e4	9	#include <linux/backing-dev.h>
ff01bb48	10	#include <linux/fs.h>
2056a782	11	#include <linux/blktrace_api.h>
bbd3e064	12	#include <linux/pr.h>
7c0f6ba6	13	#include <linux/uaccess.h>
581e2600	14	#include "blk.h"
1da177e4	15
5fb889f5 AB	16	static int blkpg_do_ioctl(struct block_device *bdev,
5fb889f5 AB	17	struct blkpg_partition __user *upart, int op)
1da177e4	18	{
1da177e4 LT	19	struct blkpg_partition p;
1da177e4 LT	20	long long start, length;
1da177e4 LT	21
	22	if (!capable(CAP_SYS_ADMIN))
	23	return -EACCES;
5fb889f5	24	if (copy_from_user(&p, upart, sizeof(struct blkpg_partition)))
1da177e4	25	return -EFAULT;
fa01b1e9	26	if (bdev_is_partition(bdev))
1da177e4	27	return -EINVAL;
88e34126	28
fa9156ae CH	29	if (p.pno <= 0)
fa9156ae CH	30	return -EINVAL;
e71bf0d0	31
fa9156ae CH	32	if (op == BLKPG_DEL_PARTITION)
fa9156ae CH	33	return bdev_del_partition(bdev, p.pno);
e71bf0d0	34
fa9156ae CH	35	start = p.start >> SECTOR_SHIFT;
fa9156ae CH	36	length = p.length >> SECTOR_SHIFT;
e71bf0d0	37
fa9156ae CH	38	switch (op) {
	39	case BLKPG_ADD_PARTITION:
	40	/* check if partition is aligned to blocksize */
	41	if (p.start & (bdev_logical_block_size(bdev) - 1))
1da177e4	42	return -EINVAL;
fa9156ae CH	43	return bdev_add_partition(bdev, p.pno, start, length);
	44	case BLKPG_RESIZE_PARTITION:
	45	return bdev_resize_partition(bdev, p.pno, start, length);
	46	default:
	47	return -EINVAL;
1da177e4 LT	48	}
	49	}
	50
5fb889f5 AB	51	static int blkpg_ioctl(struct block_device *bdev,
	52	struct blkpg_ioctl_arg __user *arg)
	53	{
	54	struct blkpg_partition __user *udata;
	55	int op;
	56
	57	if (get_user(op, &arg->op) \|\| get_user(udata, &arg->data))
	58	return -EFAULT;
	59
	60	return blkpg_do_ioctl(bdev, udata, op);
	61	}
	62
	63	#ifdef CONFIG_COMPAT
	64	struct compat_blkpg_ioctl_arg {
	65	compat_int_t op;
	66	compat_int_t flags;
	67	compat_int_t datalen;
	68	compat_caddr_t data;
	69	};
	70
	71	static int compat_blkpg_ioctl(struct block_device *bdev,
	72	struct compat_blkpg_ioctl_arg __user *arg)
	73	{
	74	compat_caddr_t udata;
	75	int op;
	76
	77	if (get_user(op, &arg->op) \|\| get_user(udata, &arg->data))
	78	return -EFAULT;
	79
	80	return blkpg_do_ioctl(bdev, compat_ptr(udata), op);
	81	}
	82	#endif
	83
f0b870df	84	static int blkdev_reread_part(struct block_device *bdev)
1da177e4	85	{
f0b870df CH	86	int ret;
f0b870df CH	87
fa01b1e9	88	if (!disk_part_scan_enabled(bdev->bd_disk) \|\| bdev_is_partition(bdev))
1da177e4 LT	89	return -EINVAL;
	90	if (!capable(CAP_SYS_ADMIN))
	91	return -EACCES;
be324177	92
b04a5636	93	mutex_lock(&bdev->bd_mutex);
f0b870df	94	ret = bdev_disk_changed(bdev, false);
c039e313	95	mutex_unlock(&bdev->bd_mutex);
be324177	96
f0b870df	97	return ret;
1da177e4 LT	98	}
1da177e4 LT	99
d8e4bb81 CH	100	static int blk_ioctl_discard(struct block_device *bdev, fmode_t mode,
d8e4bb81 CH	101	unsigned long arg, unsigned long flags)
d30a2605	102	{
d8e4bb81 CH	103	uint64_t range[2];
d8e4bb81 CH	104	uint64_t start, len;
351499a1	105	struct request_queue *q = bdev_get_queue(bdev);
384d87ef	106	int err;
d8e4bb81 CH	107
	108	if (!(mode & FMODE_WRITE))
	109	return -EBADF;
	110
351499a1 DM	111	if (!blk_queue_discard(q))
	112	return -EOPNOTSUPP;
	113
d8e4bb81 CH	114	if (copy_from_user(range, (void __user *)arg, sizeof(range)))
	115	return -EFAULT;
	116
	117	start = range[0];
	118	len = range[1];
8d57a98c	119
d30a2605 DW	120	if (start & 511)
	121	return -EINVAL;
	122	if (len & 511)
	123	return -EINVAL;
d30a2605	124
351499a1	125	if (start + len > i_size_read(bdev->bd_inode))
d30a2605	126	return -EINVAL;
384d87ef JK	127
	128	err = truncate_bdev_range(bdev, mode, start, start + len - 1);
	129	if (err)
	130	return err;
	131
351499a1 DM	132	return blkdev_issue_discard(bdev, start >> 9, len >> 9,
351499a1 DM	133	GFP_KERNEL, flags);
d30a2605 DW	134	}
d30a2605 DW	135
d8e4bb81 CH	136	static int blk_ioctl_zeroout(struct block_device *bdev, fmode_t mode,
d8e4bb81 CH	137	unsigned long arg)
66ba32dc	138	{
d8e4bb81	139	uint64_t range[2];
22dd6d35	140	uint64_t start, end, len;
384d87ef	141	int err;
d8e4bb81 CH	142
	143	if (!(mode & FMODE_WRITE))
	144	return -EBADF;
	145
	146	if (copy_from_user(range, (void __user *)arg, sizeof(range)))
	147	return -EFAULT;
	148
	149	start = range[0];
	150	len = range[1];
22dd6d35	151	end = start + len - 1;
d8e4bb81	152
66ba32dc MP	153	if (start & 511)
	154	return -EINVAL;
	155	if (len & 511)
	156	return -EINVAL;
22dd6d35 DW	157	if (end >= (uint64_t)i_size_read(bdev->bd_inode))
	158	return -EINVAL;
	159	if (end < start)
66ba32dc MP	160	return -EINVAL;
66ba32dc MP	161
22dd6d35	162	/* Invalidate the page cache, including dirty pages */
384d87ef JK	163	err = truncate_bdev_range(bdev, mode, start, end);
	164	if (err)
	165	return err;
22dd6d35 DW	166
22dd6d35 DW	167	return blkdev_issue_zeroout(bdev, start >> 9, len >> 9, GFP_KERNEL,
ee472d83	168	BLKDEV_ZERO_NOUNMAP);
66ba32dc MP	169	}
66ba32dc MP	170
9b81648c	171	static int put_ushort(unsigned short __user *argp, unsigned short val)
1da177e4	172	{
9b81648c	173	return put_user(val, argp);
1da177e4 LT	174	}
1da177e4 LT	175
9b81648c	176	static int put_int(int __user *argp, int val)
1da177e4	177	{
9b81648c	178	return put_user(val, argp);
1da177e4 LT	179	}
1da177e4 LT	180
9b81648c	181	static int put_uint(unsigned int __user *argp, unsigned int val)
ac481c20	182	{
9b81648c	183	return put_user(val, argp);
ac481c20 MP	184	}
ac481c20 MP	185
9b81648c	186	static int put_long(long __user *argp, long val)
1da177e4	187	{
9b81648c	188	return put_user(val, argp);
1da177e4 LT	189	}
1da177e4 LT	190
9b81648c	191	static int put_ulong(unsigned long __user *argp, unsigned long val)
1da177e4	192	{
9b81648c	193	return put_user(val, argp);
1da177e4 LT	194	}
1da177e4 LT	195
9b81648c	196	static int put_u64(u64 __user *argp, u64 val)
1da177e4	197	{
9b81648c	198	return put_user(val, argp);
1da177e4 LT	199	}
1da177e4 LT	200
bdc1ddad	201	#ifdef CONFIG_COMPAT
c8210a57	202	static int compat_put_long(compat_long_t __user *argp, long val)
bdc1ddad	203	{
9b81648c	204	return put_user(val, argp);
bdc1ddad AB	205	}
bdc1ddad AB	206
c8210a57	207	static int compat_put_ulong(compat_ulong_t __user *argp, compat_ulong_t val)
bdc1ddad	208	{
9b81648c	209	return put_user(val, argp);
bdc1ddad AB	210	}
	211	#endif
	212
ee6a129d AB	213	#ifdef CONFIG_COMPAT
	214	/*
	215	* This is the equivalent of compat_ptr_ioctl(), to be used by block
	216	* drivers that implement only commands that are completely compatible
	217	* between 32-bit and 64-bit user space
	218	*/
	219	int blkdev_compat_ptr_ioctl(struct block_device *bdev, fmode_t mode,
	220	unsigned cmd, unsigned long arg)
	221	{
	222	struct gendisk *disk = bdev->bd_disk;
	223
	224	if (disk->fops->ioctl)
	225	return disk->fops->ioctl(bdev, mode, cmd,
	226	(unsigned long)compat_ptr(arg));
	227
	228	return -ENOIOCTLCMD;
	229	}
	230	EXPORT_SYMBOL(blkdev_compat_ptr_ioctl);
	231	#endif
	232
bbd3e064 CH	233	static int blkdev_pr_register(struct block_device *bdev,
	234	struct pr_registration __user *arg)
	235	{
	236	const struct pr_ops *ops = bdev->bd_disk->fops->pr_ops;
	237	struct pr_registration reg;
	238
	239	if (!capable(CAP_SYS_ADMIN))
	240	return -EPERM;
	241	if (!ops \|\| !ops->pr_register)
	242	return -EOPNOTSUPP;
	243	if (copy_from_user(&reg, arg, sizeof(reg)))
	244	return -EFAULT;
	245
	246	if (reg.flags & ~PR_FL_IGNORE_KEY)
	247	return -EOPNOTSUPP;
	248	return ops->pr_register(bdev, reg.old_key, reg.new_key, reg.flags);
	249	}
	250
	251	static int blkdev_pr_reserve(struct block_device *bdev,
	252	struct pr_reservation __user *arg)
	253	{
	254	const struct pr_ops *ops = bdev->bd_disk->fops->pr_ops;
	255	struct pr_reservation rsv;
	256
	257	if (!capable(CAP_SYS_ADMIN))
	258	return -EPERM;
	259	if (!ops \|\| !ops->pr_reserve)
	260	return -EOPNOTSUPP;
	261	if (copy_from_user(&rsv, arg, sizeof(rsv)))
	262	return -EFAULT;
	263
	264	if (rsv.flags & ~PR_FL_IGNORE_KEY)
	265	return -EOPNOTSUPP;
	266	return ops->pr_reserve(bdev, rsv.key, rsv.type, rsv.flags);
	267	}
	268
	269	static int blkdev_pr_release(struct block_device *bdev,
	270	struct pr_reservation __user *arg)
	271	{
	272	const struct pr_ops *ops = bdev->bd_disk->fops->pr_ops;
	273	struct pr_reservation rsv;
	274
	275	if (!capable(CAP_SYS_ADMIN))
	276	return -EPERM;
	277	if (!ops \|\| !ops->pr_release)
	278	return -EOPNOTSUPP;
	279	if (copy_from_user(&rsv, arg, sizeof(rsv)))
	280	return -EFAULT;
	281
	282	if (rsv.flags)
	283	return -EOPNOTSUPP;
	284	return ops->pr_release(bdev, rsv.key, rsv.type);
	285	}
	286
	287	static int blkdev_pr_preempt(struct block_device *bdev,
	288	struct pr_preempt __user *arg, bool abort)
	289	{
	290	const struct pr_ops *ops = bdev->bd_disk->fops->pr_ops;
	291	struct pr_preempt p;
	292
	293	if (!capable(CAP_SYS_ADMIN))
	294	return -EPERM;
	295	if (!ops \|\| !ops->pr_preempt)
	296	return -EOPNOTSUPP;
297	if (copy_from_user(&p, arg, sizeof(p)))
298	return -EFAULT;
299
300	if (p.flags)
301	return -EOPNOTSUPP;
302	return ops->pr_preempt(bdev, p.old_key, p.new_key, p.type, abort);
303	}
304
305	static int blkdev_pr_clear(struct block_device *bdev,
306	struct pr_clear __user *arg)
307	{
308	const struct pr_ops *ops = bdev->bd_disk->fops->pr_ops;
309	struct pr_clear c;
310
311	if (!capable(CAP_SYS_ADMIN))
312	return -EPERM;
313	if (!ops \|\| !ops->pr_clear)
314	return -EOPNOTSUPP;
315	if (copy_from_user(&c, arg, sizeof(c)))
316	return -EFAULT;
317
318	if (c.flags)
319	return -EOPNOTSUPP;
320	return ops->pr_clear(bdev, c.key);
321	}
322
d8e4bb81 CH	323	static int blkdev_flushbuf(struct block_device *bdev, fmode_t mode,
d8e4bb81 CH	324	unsigned cmd, unsigned long arg)
bb93e3a5	325	{
d8e4bb81 CH	326	if (!capable(CAP_SYS_ADMIN))
d8e4bb81 CH	327	return -EACCES;
d8e4bb81 CH	328	fsync_bdev(bdev);
	329	invalidate_bdev(bdev);
	330	return 0;
	331	}
d30a2605	332
d8e4bb81 CH	333	static int blkdev_roset(struct block_device *bdev, fmode_t mode,
	334	unsigned cmd, unsigned long arg)
	335	{
	336	int ret, n;
d30a2605	337
bb749b31 ID	338	if (!capable(CAP_SYS_ADMIN))
	339	return -EACCES;
	340
d8e4bb81 CH	341	if (get_user(n, (int __user *)arg))
d8e4bb81 CH	342	return -EFAULT;
e00adcad CH	343	if (bdev->bd_disk->fops->set_read_only) {
	344	ret = bdev->bd_disk->fops->set_read_only(bdev, n);
	345	if (ret)
	346	return ret;
	347	}
83950d35	348	bdev->bd_read_only = n;
d8e4bb81 CH	349	return 0;
d8e4bb81 CH	350	}
d30a2605	351
d8e4bb81 CH	352	static int blkdev_getgeo(struct block_device *bdev,
	353	struct hd_geometry __user *argp)
	354	{
	355	struct gendisk *disk = bdev->bd_disk;
	356	struct hd_geometry geo;
	357	int ret;
d30a2605	358
d8e4bb81 CH	359	if (!argp)
	360	return -EINVAL;
	361	if (!disk->fops->getgeo)
	362	return -ENOTTY;
	363
	364	/*
	365	* We need to set the startsect first, the driver may
	366	* want to override it.
	367	*/
	368	memset(&geo, 0, sizeof(geo));
	369	geo.start = get_start_sect(bdev);
	370	ret = disk->fops->getgeo(bdev, &geo);
	371	if (ret)
	372	return ret;
	373	if (copy_to_user(argp, &geo, sizeof(geo)))
	374	return -EFAULT;
	375	return 0;
	376	}
66ba32dc	377
bdc1ddad AB	378	#ifdef CONFIG_COMPAT
	379	struct compat_hd_geometry {
	380	unsigned char heads;
	381	unsigned char sectors;
	382	unsigned short cylinders;
	383	u32 start;
	384	};
	385
9b81648c AB	386	static int compat_hdio_getgeo(struct block_device *bdev,
9b81648c AB	387	struct compat_hd_geometry __user *ugeo)
bdc1ddad	388	{
9b81648c	389	struct gendisk *disk = bdev->bd_disk;
bdc1ddad AB	390	struct hd_geometry geo;
	391	int ret;
	392
	393	if (!ugeo)
	394	return -EINVAL;
	395	if (!disk->fops->getgeo)
	396	return -ENOTTY;
	397
	398	memset(&geo, 0, sizeof(geo));
	399	/*
	400	* We need to set the startsect first, the driver may
	401	* want to override it.
	402	*/
	403	geo.start = get_start_sect(bdev);
	404	ret = disk->fops->getgeo(bdev, &geo);
	405	if (ret)
	406	return ret;
	407
	408	ret = copy_to_user(ugeo, &geo, 4);
	409	ret \|= put_user(geo.start, &ugeo->start);
	410	if (ret)
	411	ret = -EFAULT;
	412
	413	return ret;
	414	}
	415	#endif
	416
d8e4bb81 CH	417	/* set the logical block size */
	418	static int blkdev_bszset(struct block_device *bdev, fmode_t mode,
	419	int __user *argp)
	420	{
	421	int ret, n;
66ba32dc	422
d8e4bb81 CH	423	if (!capable(CAP_SYS_ADMIN))
	424	return -EACCES;
	425	if (!argp)
	426	return -EINVAL;
	427	if (get_user(n, argp))
	428	return -EFAULT;
66ba32dc	429
47816282 CH	430	if (mode & FMODE_EXCL)
47816282 CH	431	return set_blocksize(bdev, n);
d30a2605	432
47816282 CH	433	if (IS_ERR(blkdev_get_by_dev(bdev->bd_dev, mode \| FMODE_EXCL, &bdev)))
47816282 CH	434	return -EBUSY;
d8e4bb81	435	ret = set_blocksize(bdev, n);
47816282 CH	436	blkdev_put(bdev, mode \| FMODE_EXCL);
47816282 CH	437
d8e4bb81 CH	438	return ret;
d8e4bb81 CH	439	}
a885c8c4	440
d8e4bb81	441	/*
9b81648c AB	442	* Common commands that are handled the same way on native and compat
	443	* user space. Note the separate arg/argp parameters that are needed
	444	* to deal with the compat_ptr() conversion.
d8e4bb81	445	*/
9b81648c AB	446	static int blkdev_common_ioctl(struct block_device *bdev, fmode_t mode,
9b81648c AB	447	unsigned cmd, unsigned long arg, void __user *argp)
d8e4bb81	448	{
d8e4bb81 CH	449	unsigned int max_sectors;
	450
	451	switch (cmd) {
	452	case BLKFLSBUF:
	453	return blkdev_flushbuf(bdev, mode, cmd, arg);
	454	case BLKROSET:
	455	return blkdev_roset(bdev, mode, cmd, arg);
	456	case BLKDISCARD:
	457	return blk_ioctl_discard(bdev, mode, arg, 0);
	458	case BLKSECDISCARD:
	459	return blk_ioctl_discard(bdev, mode, arg,
	460	BLKDEV_DISCARD_SECURE);
	461	case BLKZEROOUT:
	462	return blk_ioctl_zeroout(bdev, mode, arg);
3ed05a98 ST	463	case BLKREPORTZONE:
	464	return blkdev_report_zones_ioctl(bdev, mode, cmd, arg);
	465	case BLKRESETZONE:
e876df1f AJ	466	case BLKOPENZONE:
	467	case BLKCLOSEZONE:
	468	case BLKFINISHZONE:
	469	return blkdev_zone_mgmt_ioctl(bdev, mode, cmd, arg);
72cd8757	470	case BLKGETZONESZ:
9b81648c	471	return put_uint(argp, bdev_zone_sectors(bdev));
65e4e3ee	472	case BLKGETNRZONES:
9b81648c	473	return put_uint(argp, blkdev_nr_zones(bdev->bd_disk));
45048d09	474	case BLKROGET:
9b81648c	475	return put_int(argp, bdev_read_only(bdev) != 0);
ac481c20	476	case BLKSSZGET: /* get block device logical block size */
9b81648c	477	return put_int(argp, bdev_logical_block_size(bdev));
ac481c20	478	case BLKPBSZGET: /* get block device physical block size */
9b81648c	479	return put_uint(argp, bdev_physical_block_size(bdev));
ac481c20	480	case BLKIOMIN:
9b81648c	481	return put_uint(argp, bdev_io_min(bdev));
ac481c20	482	case BLKIOOPT:
9b81648c	483	return put_uint(argp, bdev_io_opt(bdev));
ac481c20	484	case BLKALIGNOFF:
9b81648c	485	return put_int(argp, bdev_alignment_offset(bdev));
98262f27	486	case BLKDISCARDZEROES:
9b81648c	487	return put_uint(argp, 0);
45048d09	488	case BLKSECTGET:
63f26496 AM	489	max_sectors = min_t(unsigned int, USHRT_MAX,
63f26496 AM	490	queue_max_sectors(bdev_get_queue(bdev)));
9b81648c	491	return put_ushort(argp, max_sectors);
ef00f59c	492	case BLKROTATIONAL:
9b81648c	493	return put_ushort(argp, !blk_queue_nonrot(bdev_get_queue(bdev)));
45048d09 AV	494	case BLKRASET:
	495	case BLKFRASET:
	496	if(!capable(CAP_SYS_ADMIN))
	497	return -EACCES;
efa7c9f9	498	bdev->bd_bdi->ra_pages = (arg * 512) / PAGE_SIZE;
45048d09	499	return 0;
45048d09	500	case BLKRRPART:
d8e4bb81	501	return blkdev_reread_part(bdev);
45048d09 AV	502	case BLKTRACESTART:
45048d09 AV	503	case BLKTRACESTOP:
45048d09	504	case BLKTRACETEARDOWN:
d8e4bb81	505	return blk_trace_ioctl(bdev, cmd, argp);
bbd3e064 CH	506	case IOC_PR_REGISTER:
	507	return blkdev_pr_register(bdev, argp);
	508	case IOC_PR_RESERVE:
	509	return blkdev_pr_reserve(bdev, argp);
	510	case IOC_PR_RELEASE:
	511	return blkdev_pr_release(bdev, argp);
	512	case IOC_PR_PREEMPT:
	513	return blkdev_pr_preempt(bdev, argp, false);
	514	case IOC_PR_PREEMPT_ABORT:
	515	return blkdev_pr_preempt(bdev, argp, true);
	516	case IOC_PR_CLEAR:
	517	return blkdev_pr_clear(bdev, argp);
45048d09	518	default:
9b81648c	519	return -ENOIOCTLCMD;
45048d09	520	}
1da177e4	521	}
9b81648c AB	522
	523	/*
	524	* Always keep this in sync with compat_blkdev_ioctl()
	525	* to handle all incompatible commands in both functions.
	526	*
	527	* New commands must be compatible and go into blkdev_common_ioctl
	528	*/
	529	int blkdev_ioctl(struct block_device *bdev, fmode_t mode, unsigned cmd,
	530	unsigned long arg)
	531	{
	532	int ret;
	533	loff_t size;
	534	void __user argp = (void __user )arg;
	535
	536	switch (cmd) {
	537	/* These need separate implementations for the data structure */
	538	case HDIO_GETGEO:
	539	return blkdev_getgeo(bdev, argp);
	540	case BLKPG:
	541	return blkpg_ioctl(bdev, argp);
	542
	543	/* Compat mode returns 32-bit data instead of 'long' */
	544	case BLKRAGET:
	545	case BLKFRAGET:
	546	if (!argp)
	547	return -EINVAL;
	548	return put_long(argp, (bdev->bd_bdi->ra_pages*PAGE_SIZE) / 512);
	549	case BLKGETSIZE:
	550	size = i_size_read(bdev->bd_inode);
	551	if ((size >> 9) > ~0UL)
	552	return -EFBIG;
	553	return put_ulong(argp, size >> 9);
	554
	555	/* The data is compatible, but the command number is different */
	556	case BLKBSZGET: /* get block device soft block size (cf. BLKSSZGET) */
	557	return put_int(argp, block_size(bdev));
	558	case BLKBSZSET:
	559	return blkdev_bszset(bdev, mode, argp);
	560	case BLKGETSIZE64:
	561	return put_u64(argp, i_size_read(bdev->bd_inode));
	562
	563	/* Incompatible alignment on i386 */
	564	case BLKTRACESETUP:
	565	return blk_trace_ioctl(bdev, cmd, argp);
	566	default:
	567	break;
	568	}
	569
	570	ret = blkdev_common_ioctl(bdev, mode, cmd, arg, argp);
a7cb3d2f CH	571	if (ret != -ENOIOCTLCMD)
a7cb3d2f CH	572	return ret;
9b81648c	573
a7cb3d2f CH	574	if (!bdev->bd_disk->fops->ioctl)
	575	return -ENOTTY;
	576	return bdev->bd_disk->fops->ioctl(bdev, mode, cmd, arg);
9b81648c AB	577	}
9b81648c AB	578	EXPORT_SYMBOL_GPL(blkdev_ioctl); /* for /dev/raw */
bdc1ddad AB	579
bdc1ddad AB	580	#ifdef CONFIG_COMPAT
9b81648c	581
bdc1ddad AB	582	#define BLKBSZGET_32 _IOR(0x12, 112, int)
	583	#define BLKBSZSET_32 _IOW(0x12, 113, int)
	584	#define BLKGETSIZE64_32 _IOR(0x12, 114, int)
	585
	586	/* Most of the generic ioctls are handled in the normal fallback path.
	587	This assumes the blkdev's low level compat_ioctl always returns
	588	ENOIOCTLCMD for unknown ioctls. */
	589	long compat_blkdev_ioctl(struct file *file, unsigned cmd, unsigned long arg)
	590	{
9b81648c AB	591	int ret;
9b81648c AB	592	void __user *argp = compat_ptr(arg);
4e7b5671	593	struct block_device *bdev = I_BDEV(file->f_mapping->host);
bdc1ddad AB	594	struct gendisk *disk = bdev->bd_disk;
	595	fmode_t mode = file->f_mode;
	596	loff_t size;
bdc1ddad AB	597
	598	/*
	599	* O_NDELAY can be altered using fcntl(.., F_SETFL, ..), so we have
	600	* to updated it before every ioctl.
	601	*/
	602	if (file->f_flags & O_NDELAY)
	603	mode \|= FMODE_NDELAY;
	604	else
	605	mode &= ~FMODE_NDELAY;
	606
	607	switch (cmd) {
9b81648c	608	/* These need separate implementations for the data structure */
bdc1ddad	609	case HDIO_GETGEO:
9b81648c	610	return compat_hdio_getgeo(bdev, argp);
bdc1ddad	611	case BLKPG:
9b81648c AB	612	return compat_blkpg_ioctl(bdev, argp);
	613
	614	/* Compat mode returns 32-bit data instead of 'long' */
bdc1ddad AB	615	case BLKRAGET:
bdc1ddad AB	616	case BLKFRAGET:
9b81648c	617	if (!argp)
bdc1ddad	618	return -EINVAL;
9b81648c	619	return compat_put_long(argp,
bdc1ddad	620	(bdev->bd_bdi->ra_pages * PAGE_SIZE) / 512);
bdc1ddad AB	621	case BLKGETSIZE:
	622	size = i_size_read(bdev->bd_inode);
	623	if ((size >> 9) > ~0UL)
	624	return -EFBIG;
9b81648c	625	return compat_put_ulong(argp, size >> 9);
bdc1ddad	626
9b81648c AB	627	/* The data is compatible, but the command number is different */
	628	case BLKBSZGET_32: /* get the logical block size (cf. BLKSSZGET) */
	629	return put_int(argp, bdev_logical_block_size(bdev));
	630	case BLKBSZSET_32:
	631	return blkdev_bszset(bdev, mode, argp);
bdc1ddad	632	case BLKGETSIZE64_32:
9b81648c	633	return put_u64(argp, i_size_read(bdev->bd_inode));
bdc1ddad	634
9b81648c	635	/* Incompatible alignment on i386 */
bdc1ddad	636	case BLKTRACESETUP32:
9b81648c	637	return blk_trace_ioctl(bdev, cmd, argp);
bdc1ddad	638	default:
9b81648c	639	break;
bdc1ddad	640	}
9b81648c AB	641
	642	ret = blkdev_common_ioctl(bdev, mode, cmd, arg, argp);
	643	if (ret == -ENOIOCTLCMD && disk->fops->compat_ioctl)
	644	ret = disk->fops->compat_ioctl(bdev, mode, cmd, arg);
	645
	646	return ret;
bdc1ddad AB	647	}
bdc1ddad AB	648	#endif