[linux-2.6-block.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().
 */
int blkdev_reread_part(struct block_device *bdev)
{
	int res;

	if (!mutex_trylock(&bdev->bd_mutex))
		return -EBUSY;
	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, uint64_t start,
			     uint64_t len, int secure)
{
	unsigned long flags = 0;

	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;
	if (secure)
		flags |= BLKDEV_DISCARD_SECURE;
	return blkdev_issue_discard(bdev, start, len, GFP_KERNEL, flags);
}

static int blk_ioctl_zeroout(struct block_device *bdev, uint64_t start,
			     uint64_t len)
{
	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;
}

/*
 * 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 gendisk *disk = bdev->bd_disk;
	struct backing_dev_info *bdi;
	loff_t size;
	int ret, n;
	unsigned int max_sectors;

	switch(cmd) {
	case BLKFLSBUF:
		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;

	case BLKROSET:
		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;

	case BLKDISCARD:
	case BLKSECDISCARD: {
		uint64_t range[2];

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

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

		return blk_ioctl_discard(bdev, range[0], range[1],
					 cmd == BLKSECDISCARD);
	}
	case BLKZEROOUT: {
		uint64_t range[2];

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

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

		return blk_ioctl_zeroout(bdev, range[0], range[1]);
	}

	case HDIO_GETGEO: {
		struct hd_geometry geo;

		if (!arg)
			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((struct hd_geometry __user *)arg, &geo,
					sizeof(geo)))
			return -EFAULT;
		return 0;
	}
	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:
		/* set the logical block size */
		if (!capable(CAP_SYS_ADMIN))
			return -EACCES;
		if (!arg)
			return -EINVAL;
		if (get_user(n, (int __user *) arg))
			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;
	case BLKPG:
		ret = blkpg_ioctl(bdev, (struct blkpg_ioctl_arg __user *) arg);
		break;
	case BLKRRPART:
		ret = blkdev_reread_part(bdev);
		break;
	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:
		ret = blk_trace_ioctl(bdev, cmd, (char __user *) arg);
		break;
	default:
		ret = __blkdev_driver_ioctl(bdev, mode, cmd, arg);
	}
	return ret;
}
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().
	177	*/
	178	int blkdev_reread_part(struct block_device *bdev)
	179	{
	180	int res;
	181
c039e313	182	if (!mutex_trylock(&bdev->bd_mutex))
1da177e4	183	return -EBUSY;
be324177	184	res = __blkdev_reread_part(bdev);
c039e313	185	mutex_unlock(&bdev->bd_mutex);
be324177	186
1da177e4 LT	187	return res;
1da177e4 LT	188	}
be324177	189	EXPORT_SYMBOL(blkdev_reread_part);
1da177e4	190
d30a2605	191	static int blk_ioctl_discard(struct block_device *bdev, uint64_t start,
8d57a98c	192	uint64_t len, int secure)
d30a2605	193	{
dd3932ed	194	unsigned long flags = 0;
8d57a98c	195
d30a2605 DW	196	if (start & 511)
	197	return -EINVAL;
	198	if (len & 511)
	199	return -EINVAL;
	200	start >>= 9;
	201	len >>= 9;
	202
77304d2a	203	if (start + len > (i_size_read(bdev->bd_inode) >> 9))
d30a2605	204	return -EINVAL;
8d57a98c	205	if (secure)
dd3932ed	206	flags \|= BLKDEV_DISCARD_SECURE;
8d57a98c	207	return blkdev_issue_discard(bdev, start, len, GFP_KERNEL, flags);
d30a2605 DW	208	}
d30a2605 DW	209
66ba32dc MP	210	static int blk_ioctl_zeroout(struct block_device *bdev, uint64_t start,
	211	uint64_t len)
	212	{
	213	if (start & 511)
	214	return -EINVAL;
	215	if (len & 511)
	216	return -EINVAL;
	217	start >>= 9;
	218	len >>= 9;
	219
	220	if (start + len > (i_size_read(bdev->bd_inode) >> 9))
	221	return -EINVAL;
	222
d93ba7a5	223	return blkdev_issue_zeroout(bdev, start, len, GFP_KERNEL, false);
66ba32dc MP	224	}
66ba32dc MP	225
1da177e4 LT	226	static int put_ushort(unsigned long arg, unsigned short val)
	227	{
	228	return put_user(val, (unsigned short __user *)arg);
	229	}
	230
	231	static int put_int(unsigned long arg, int val)
	232	{
	233	return put_user(val, (int __user *)arg);
	234	}
	235
ac481c20 MP	236	static int put_uint(unsigned long arg, unsigned int val)
	237	{
	238	return put_user(val, (unsigned int __user *)arg);
	239	}
	240
1da177e4 LT	241	static int put_long(unsigned long arg, long val)
	242	{
	243	return put_user(val, (long __user *)arg);
	244	}
	245
	246	static int put_ulong(unsigned long arg, unsigned long val)
	247	{
	248	return put_user(val, (unsigned long __user *)arg);
	249	}
	250
	251	static int put_u64(unsigned long arg, u64 val)
	252	{
	253	return put_user(val, (u64 __user *)arg);
	254	}
	255
633a08b8 AV	256	int __blkdev_driver_ioctl(struct block_device *bdev, fmode_t mode,
	257	unsigned cmd, unsigned long arg)
	258	{
	259	struct gendisk *disk = bdev->bd_disk;
d4430d62 AV	260
	261	if (disk->fops->ioctl)
	262	return disk->fops->ioctl(bdev, mode, cmd, arg);
633a08b8	263
633a08b8 AV	264	return -ENOTTY;
	265	}
	266	/*
	267	* For the record: _GPL here is only because somebody decided to slap it
	268	* on the previous export. Sheer idiocy, since it wasn't copyrightable
	269	* at all and could be open-coded without any exports by anybody who cares.
	270	*/
	271	EXPORT_SYMBOL_GPL(__blkdev_driver_ioctl);
	272
07d106d0 LT	273	/*
	274	* Is it an unrecognized ioctl? The correct returns are either
	275	* ENOTTY (final) or ENOIOCTLCMD ("I don't know this one, try a
	276	* fallback"). ENOIOCTLCMD gets turned into ENOTTY by the ioctl
	277	* code before returning.
	278	*
	279	* Confused drivers sometimes return EINVAL, which is wrong. It
	280	* means "I understood the ioctl command, but the parameters to
	281	* it were wrong".
	282	*
	283	* We should aim to just fix the broken drivers, the EINVAL case
	284	* should go away.
	285	*/
	286	static inline int is_unrecognized_ioctl(int ret)
	287	{
	288	return ret == -EINVAL \|\|
	289	ret == -ENOTTY \|\|
	290	ret == -ENOIOCTLCMD;
	291	}
	292
f58c4c0a	293	/*
8a6cfeb6	294	* always keep this in sync with compat_blkdev_ioctl()
f58c4c0a	295	*/
56b26add	296	int blkdev_ioctl(struct block_device *bdev, fmode_t mode, unsigned cmd,
bb93e3a5 AB	297	unsigned long arg)
bb93e3a5 AB	298	{
bb93e3a5	299	struct gendisk *disk = bdev->bd_disk;
45048d09 AV	300	struct backing_dev_info *bdi;
45048d09 AV	301	loff_t size;
bb93e3a5	302	int ret, n;
63f26496	303	unsigned int max_sectors;
bb93e3a5 AB	304
bb93e3a5 AB	305	switch(cmd) {
1da177e4 LT	306	case BLKFLSBUF:
	307	if (!capable(CAP_SYS_ADMIN))
	308	return -EACCES;
bb93e3a5	309
e436fdae	310	ret = __blkdev_driver_ioctl(bdev, mode, cmd, arg);
07d106d0	311	if (!is_unrecognized_ioctl(ret))
bb93e3a5 AB	312	return ret;
bb93e3a5 AB	313
1da177e4	314	fsync_bdev(bdev);
f98393a6	315	invalidate_bdev(bdev);
1da177e4	316	return 0;
bb93e3a5	317
1da177e4	318	case BLKROSET:
e436fdae	319	ret = __blkdev_driver_ioctl(bdev, mode, cmd, arg);
07d106d0	320	if (!is_unrecognized_ioctl(ret))
bb93e3a5	321	return ret;
1da177e4 LT	322	if (!capable(CAP_SYS_ADMIN))
	323	return -EACCES;
	324	if (get_user(n, (int __user *)(arg)))
	325	return -EFAULT;
	326	set_device_ro(bdev, n);
	327	return 0;
d30a2605	328
8d57a98c AH	329	case BLKDISCARD:
8d57a98c AH	330	case BLKSECDISCARD: {
d30a2605 DW	331	uint64_t range[2];
d30a2605 DW	332
e436fdae	333	if (!(mode & FMODE_WRITE))
d30a2605 DW	334	return -EBADF;
	335
	336	if (copy_from_user(range, (void __user *)arg, sizeof(range)))
	337	return -EFAULT;
	338
8d57a98c AH	339	return blk_ioctl_discard(bdev, range[0], range[1],
8d57a98c AH	340	cmd == BLKSECDISCARD);
d30a2605	341	}
66ba32dc MP	342	case BLKZEROOUT: {
	343	uint64_t range[2];
	344
	345	if (!(mode & FMODE_WRITE))
	346	return -EBADF;
	347
	348	if (copy_from_user(range, (void __user *)arg, sizeof(range)))
	349	return -EFAULT;
	350
	351	return blk_ioctl_zeroout(bdev, range[0], range[1]);
	352	}
d30a2605	353
a885c8c4 CH	354	case HDIO_GETGEO: {
	355	struct hd_geometry geo;
	356
	357	if (!arg)
	358	return -EINVAL;
	359	if (!disk->fops->getgeo)
	360	return -ENOTTY;
	361
	362	/*
	363	* We need to set the startsect first, the driver may
	364	* want to override it.
	365	*/
a014741c	366	memset(&geo, 0, sizeof(geo));
a885c8c4 CH	367	geo.start = get_start_sect(bdev);
	368	ret = disk->fops->getgeo(bdev, &geo);
	369	if (ret)
	370	return ret;
	371	if (copy_to_user((struct hd_geometry __user *)arg, &geo,
	372	sizeof(geo)))
	373	return -EFAULT;
	374	return 0;
	375	}
45048d09 AV	376	case BLKRAGET:
	377	case BLKFRAGET:
	378	if (!arg)
	379	return -EINVAL;
	380	bdi = blk_get_backing_dev_info(bdev);
45048d09 AV	381	return put_long(arg, (bdi->ra_pages * PAGE_CACHE_SIZE) / 512);
	382	case BLKROGET:
	383	return put_int(arg, bdev_read_only(bdev) != 0);
ac481c20	384	case BLKBSZGET: /* get block device soft block size (cf. BLKSSZGET) */
45048d09	385	return put_int(arg, block_size(bdev));
ac481c20	386	case BLKSSZGET: /* get block device logical block size */
e1defc4f	387	return put_int(arg, bdev_logical_block_size(bdev));
ac481c20 MP	388	case BLKPBSZGET: /* get block device physical block size */
	389	return put_uint(arg, bdev_physical_block_size(bdev));
	390	case BLKIOMIN:
	391	return put_uint(arg, bdev_io_min(bdev));
	392	case BLKIOOPT:
	393	return put_uint(arg, bdev_io_opt(bdev));
	394	case BLKALIGNOFF:
	395	return put_int(arg, bdev_alignment_offset(bdev));
98262f27 MP	396	case BLKDISCARDZEROES:
98262f27 MP	397	return put_uint(arg, bdev_discard_zeroes_data(bdev));
45048d09	398	case BLKSECTGET:
63f26496 AM	399	max_sectors = min_t(unsigned int, USHRT_MAX,
	400	queue_max_sectors(bdev_get_queue(bdev)));
	401	return put_ushort(arg, max_sectors);
ef00f59c MP	402	case BLKROTATIONAL:
ef00f59c MP	403	return put_ushort(arg, !blk_queue_nonrot(bdev_get_queue(bdev)));
45048d09 AV	404	case BLKRASET:
	405	case BLKFRASET:
	406	if(!capable(CAP_SYS_ADMIN))
	407	return -EACCES;
	408	bdi = blk_get_backing_dev_info(bdev);
45048d09	409	bdi->ra_pages = (arg * 512) / PAGE_CACHE_SIZE;
45048d09 AV	410	return 0;
	411	case BLKBSZSET:
	412	/* set the logical block size */
	413	if (!capable(CAP_SYS_ADMIN))
	414	return -EACCES;
	415	if (!arg)
	416	return -EINVAL;
	417	if (get_user(n, (int __user *) arg))
	418	return -EFAULT;
3c522ced MS	419	if (!(mode & FMODE_EXCL)) {
	420	bdgrab(bdev);
	421	if (blkdev_get(bdev, mode \| FMODE_EXCL, &bdev) < 0)
	422	return -EBUSY;
	423	}
45048d09	424	ret = set_blocksize(bdev, n);
6af3a56e	425	if (!(mode & FMODE_EXCL))
e525fd89	426	blkdev_put(bdev, mode \| FMODE_EXCL);
bb93e3a5	427	return ret;
45048d09	428	case BLKPG:
45048d09	429	ret = blkpg_ioctl(bdev, (struct blkpg_ioctl_arg __user *) arg);
45048d09 AV	430	break;
45048d09 AV	431	case BLKRRPART:
45048d09	432	ret = blkdev_reread_part(bdev);
45048d09 AV	433	break;
45048d09 AV	434	case BLKGETSIZE:
77304d2a	435	size = i_size_read(bdev->bd_inode);
45048d09 AV	436	if ((size >> 9) > ~0UL)
	437	return -EFBIG;
	438	return put_ulong(arg, size >> 9);
	439	case BLKGETSIZE64:
77304d2a	440	return put_u64(arg, i_size_read(bdev->bd_inode));
45048d09 AV	441	case BLKTRACESTART:
	442	case BLKTRACESTOP:
	443	case BLKTRACESETUP:
	444	case BLKTRACETEARDOWN:
45048d09	445	ret = blk_trace_ioctl(bdev, cmd, (char __user *) arg);
45048d09 AV	446	break;
	447	default:
	448	ret = __blkdev_driver_ioctl(bdev, mode, cmd, arg);
	449	}
	450	return ret;
1da177e4	451	}
68f66feb	452	EXPORT_SYMBOL_GPL(blkdev_ioctl);