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

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

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

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

			mutex_lock(&bdev->bd_mutex);

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

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

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

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

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

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

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

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

	lockdep_assert_held(&bdev->bd_mutex);

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

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

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

	return res;
}
EXPORT_SYMBOL(blkdev_reread_part);

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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