ioctl.c 15.3 KB
Newer Older
1
#include <linux/capability.h>
Linus Torvalds's avatar
Linus Torvalds committed
2
#include <linux/blkdev.h>
3
#include <linux/export.h>
4
#include <linux/gfp.h>
Linus Torvalds's avatar
Linus Torvalds committed
5
#include <linux/blkpg.h>
6
#include <linux/hdreg.h>
Linus Torvalds's avatar
Linus Torvalds committed
7
#include <linux/backing-dev.h>
Al Viro's avatar
Al Viro committed
8
#include <linux/fs.h>
9
#include <linux/blktrace_api.h>
10
#include <linux/pr.h>
11
#include <linux/uaccess.h>
Linus Torvalds's avatar
Linus Torvalds committed
12 13 14 15 16

static int blkpg_ioctl(struct block_device *bdev, struct blkpg_ioctl_arg __user *arg)
{
	struct block_device *bdevp;
	struct gendisk *disk;
17
	struct hd_struct *part, *lpart;
Linus Torvalds's avatar
Linus Torvalds committed
18 19
	struct blkpg_ioctl_arg a;
	struct blkpg_partition p;
20
	struct disk_part_iter piter;
Linus Torvalds's avatar
Linus Torvalds committed
21
	long long start, length;
22
	int partno;
Linus Torvalds's avatar
Linus Torvalds committed
23 24 25 26 27 28 29 30 31 32

	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;
33
	partno = p.pno;
34
	if (partno <= 0)
Linus Torvalds's avatar
Linus Torvalds committed
35 36 37 38 39
		return -EINVAL;
	switch (a.op) {
		case BLKPG_ADD_PARTITION:
			start = p.start >> 9;
			length = p.length >> 9;
40 41
			/* check for fit in a hd_struct */
			if (sizeof(sector_t) == sizeof(long) &&
Linus Torvalds's avatar
Linus Torvalds committed
42 43 44
			    sizeof(long long) > sizeof(long)) {
				long pstart = start, plength = length;
				if (pstart != start || plength != length
45
				    || pstart < 0 || plength < 0 || partno > 65535)
Linus Torvalds's avatar
Linus Torvalds committed
46 47
					return -EINVAL;
			}
48 49 50
			/* check if partition is aligned to blocksize */
			if (p.start & (bdev_logical_block_size(bdev) - 1))
				return -EINVAL;
51

52
			mutex_lock(&bdev->bd_mutex);
53

Linus Torvalds's avatar
Linus Torvalds committed
54
			/* overlap? */
55 56 57 58 59 60
			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);
61
					mutex_unlock(&bdev->bd_mutex);
Linus Torvalds's avatar
Linus Torvalds committed
62 63 64
					return -EBUSY;
				}
			}
65 66
			disk_part_iter_exit(&piter);

Linus Torvalds's avatar
Linus Torvalds committed
67
			/* all seems OK */
68
			part = add_partition(disk, partno, start, length,
69
					     ADDPART_FLAG_NONE, NULL);
70
			mutex_unlock(&bdev->bd_mutex);
71
			return PTR_ERR_OR_ZERO(part);
Linus Torvalds's avatar
Linus Torvalds committed
72
		case BLKPG_DEL_PARTITION:
73 74
			part = disk_get_part(disk, partno);
			if (!part)
Linus Torvalds's avatar
Linus Torvalds committed
75
				return -ENXIO;
76 77 78

			bdevp = bdget(part_devt(part));
			disk_put_part(part);
Linus Torvalds's avatar
Linus Torvalds committed
79 80
			if (!bdevp)
				return -ENOMEM;
81

82
			mutex_lock(&bdevp->bd_mutex);
Linus Torvalds's avatar
Linus Torvalds committed
83
			if (bdevp->bd_openers) {
84
				mutex_unlock(&bdevp->bd_mutex);
Linus Torvalds's avatar
Linus Torvalds committed
85 86 87 88 89
				bdput(bdevp);
				return -EBUSY;
			}
			/* all seems OK */
			fsync_bdev(bdevp);
90
			invalidate_bdev(bdevp);
Linus Torvalds's avatar
Linus Torvalds committed
91

92
			mutex_lock_nested(&bdev->bd_mutex, 1);
93
			delete_partition(disk, partno);
94 95
			mutex_unlock(&bdev->bd_mutex);
			mutex_unlock(&bdevp->bd_mutex);
Linus Torvalds's avatar
Linus Torvalds committed
96 97
			bdput(bdevp);

98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150
			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);
Linus Torvalds's avatar
Linus Torvalds committed
151 152 153 154 155 156
			return 0;
		default:
			return -EINVAL;
	}
}

157 158 159 160 161 162
/*
 * 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)
Linus Torvalds's avatar
Linus Torvalds committed
163 164 165
{
	struct gendisk *disk = bdev->bd_disk;

166
	if (!disk_part_scan_enabled(disk) || bdev != bdev->bd_contains)
Linus Torvalds's avatar
Linus Torvalds committed
167 168 169
		return -EINVAL;
	if (!capable(CAP_SYS_ADMIN))
		return -EACCES;
170 171 172 173 174 175 176 177 178 179 180

	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().
181 182 183 184 185 186
 *
 * 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
187 188 189 190 191
 */
int blkdev_reread_part(struct block_device *bdev)
{
	int res;

192
	mutex_lock(&bdev->bd_mutex);
193
	res = __blkdev_reread_part(bdev);
194
	mutex_unlock(&bdev->bd_mutex);
195

Linus Torvalds's avatar
Linus Torvalds committed
196 197
	return res;
}
198
EXPORT_SYMBOL(blkdev_reread_part);
Linus Torvalds's avatar
Linus Torvalds committed
199

200 201
static int blk_ioctl_discard(struct block_device *bdev, fmode_t mode,
		unsigned long arg, unsigned long flags)
202
{
203 204
	uint64_t range[2];
	uint64_t start, len;
205 206 207
	struct request_queue *q = bdev_get_queue(bdev);
	struct address_space *mapping = bdev->bd_inode->i_mapping;

208 209 210 211

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

212 213 214
	if (!blk_queue_discard(q))
		return -EOPNOTSUPP;

215 216 217 218 219
	if (copy_from_user(range, (void __user *)arg, sizeof(range)))
		return -EFAULT;

	start = range[0];
	len = range[1];
220

221 222 223 224 225
	if (start & 511)
		return -EINVAL;
	if (len & 511)
		return -EINVAL;

226
	if (start + len > i_size_read(bdev->bd_inode))
227
		return -EINVAL;
228
	truncate_inode_pages_range(mapping, start, start + len - 1);
229 230
	return blkdev_issue_discard(bdev, start >> 9, len >> 9,
				    GFP_KERNEL, flags);
231 232
}

233 234
static int blk_ioctl_zeroout(struct block_device *bdev, fmode_t mode,
		unsigned long arg)
235
{
236
	uint64_t range[2];
237 238
	struct address_space *mapping;
	uint64_t start, end, len;
239 240 241 242 243 244 245 246 247

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

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

	start = range[0];
	len = range[1];
248
	end = start + len - 1;
249

250 251 252 253
	if (start & 511)
		return -EINVAL;
	if (len & 511)
		return -EINVAL;
254 255 256
	if (end >= (uint64_t)i_size_read(bdev->bd_inode))
		return -EINVAL;
	if (end < start)
257 258
		return -EINVAL;

259 260 261 262 263
	/* 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,
264
			BLKDEV_ZERO_NOUNMAP);
265 266
}

Linus Torvalds's avatar
Linus Torvalds committed
267 268 269 270 271 272 273 274 275 276
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);
}

277 278 279 280 281
static int put_uint(unsigned long arg, unsigned int val)
{
	return put_user(val, (unsigned int __user *)arg);
}

Linus Torvalds's avatar
Linus Torvalds committed
282 283 284 285 286 287 288 289 290 291 292 293 294 295 296
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);
}

297 298 299 300
int __blkdev_driver_ioctl(struct block_device *bdev, fmode_t mode,
			unsigned cmd, unsigned long arg)
{
	struct gendisk *disk = bdev->bd_disk;
301 302 303

	if (disk->fops->ioctl)
		return disk->fops->ioctl(bdev, mode, cmd, arg);
304 305 306 307 308 309 310 311 312 313

	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);

314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403
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);
}

404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423
/*
 * 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;
}

424 425
static int blkdev_flushbuf(struct block_device *bdev, fmode_t mode,
		unsigned cmd, unsigned long arg)
426
{
427
	int ret;
428

429 430
	if (!capable(CAP_SYS_ADMIN))
		return -EACCES;
431

432 433 434
	ret = __blkdev_driver_ioctl(bdev, mode, cmd, arg);
	if (!is_unrecognized_ioctl(ret))
		return ret;
435

436 437 438 439
	fsync_bdev(bdev);
	invalidate_bdev(bdev);
	return 0;
}
440

441 442 443 444
static int blkdev_roset(struct block_device *bdev, fmode_t mode,
		unsigned cmd, unsigned long arg)
{
	int ret, n;
445

446 447 448
	if (!capable(CAP_SYS_ADMIN))
		return -EACCES;

449 450 451 452 453 454 455 456
	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;
}
457

458 459 460 461 462 463
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;
464

465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482
	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;
}
483

484 485 486 487 488
/* set the logical block size */
static int blkdev_bszset(struct block_device *bdev, fmode_t mode,
		int __user *argp)
{
	int ret, n;
489

490 491 492 493 494 495
	if (!capable(CAP_SYS_ADMIN))
		return -EACCES;
	if (!argp)
		return -EINVAL;
	if (get_user(n, argp))
		return -EFAULT;
496

497 498 499 500
	if (!(mode & FMODE_EXCL)) {
		bdgrab(bdev);
		if (blkdev_get(bdev, mode | FMODE_EXCL, &bdev) < 0)
			return -EBUSY;
501
	}
502

503 504 505 506 507
	ret = set_blocksize(bdev, n);
	if (!(mode & FMODE_EXCL))
		blkdev_put(bdev, mode | FMODE_EXCL);
	return ret;
}
508

509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530
/*
 * 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);
531 532 533 534
	case BLKREPORTZONE:
		return blkdev_report_zones_ioctl(bdev, mode, cmd, arg);
	case BLKRESETZONE:
		return blkdev_reset_zones_ioctl(bdev, mode, cmd, arg);
535 536
	case BLKGETZONESZ:
		return put_uint(arg, bdev_zone_sectors(bdev));
537 538
	case BLKGETNRZONES:
		return put_uint(arg, blkdev_nr_zones(bdev));
539 540
	case HDIO_GETGEO:
		return blkdev_getgeo(bdev, argp);
541 542 543 544
	case BLKRAGET:
	case BLKFRAGET:
		if (!arg)
			return -EINVAL;
545
		return put_long(arg, (bdev->bd_bdi->ra_pages*PAGE_SIZE) / 512);
546 547
	case BLKROGET:
		return put_int(arg, bdev_read_only(bdev) != 0);
548
	case BLKBSZGET: /* get block device soft block size (cf. BLKSSZGET) */
549
		return put_int(arg, block_size(bdev));
550
	case BLKSSZGET: /* get block device logical block size */
551
		return put_int(arg, bdev_logical_block_size(bdev));
552 553 554 555 556 557 558 559
	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));
560
	case BLKDISCARDZEROES:
561
		return put_uint(arg, 0);
562
	case BLKSECTGET:
563 564 565
		max_sectors = min_t(unsigned int, USHRT_MAX,
				    queue_max_sectors(bdev_get_queue(bdev)));
		return put_ushort(arg, max_sectors);
566 567
	case BLKROTATIONAL:
		return put_ushort(arg, !blk_queue_nonrot(bdev_get_queue(bdev)));
568 569 570 571
	case BLKRASET:
	case BLKFRASET:
		if(!capable(CAP_SYS_ADMIN))
			return -EACCES;
572
		bdev->bd_bdi->ra_pages = (arg * 512) / PAGE_SIZE;
573 574
		return 0;
	case BLKBSZSET:
575
		return blkdev_bszset(bdev, mode, argp);
576
	case BLKPG:
577
		return blkpg_ioctl(bdev, argp);
578
	case BLKRRPART:
579
		return blkdev_reread_part(bdev);
580
	case BLKGETSIZE:
581
		size = i_size_read(bdev->bd_inode);
582 583 584 585
		if ((size >> 9) > ~0UL)
			return -EFBIG;
		return put_ulong(arg, size >> 9);
	case BLKGETSIZE64:
586
		return put_u64(arg, i_size_read(bdev->bd_inode));
587 588 589 590
	case BLKTRACESTART:
	case BLKTRACESTOP:
	case BLKTRACESETUP:
	case BLKTRACETEARDOWN:
591
		return blk_trace_ioctl(bdev, cmd, argp);
592 593 594 595 596 597 598 599 600 601 602 603
	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);
604
	default:
605
		return __blkdev_driver_ioctl(bdev, mode, cmd, arg);
606
	}
Linus Torvalds's avatar
Linus Torvalds committed
607
}
608
EXPORT_SYMBOL_GPL(blkdev_ioctl);