ioctl.c 82.2 KB
Newer Older
C
Christoph Hellwig 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
/*
 * Copyright (C) 2007 Oracle.  All rights reserved.
 *
 * This program is free software; you can redistribute it and/or
 * modify it under the terms of the GNU General Public
 * License v2 as published by the Free Software Foundation.
 *
 * This program is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
 * General Public License for more details.
 *
 * You should have received a copy of the GNU General Public
 * License along with this program; if not, write to the
 * Free Software Foundation, Inc., 59 Temple Place - Suite 330,
 * Boston, MA 021110-1307, USA.
 */

#include <linux/kernel.h>
#include <linux/bio.h>
#include <linux/buffer_head.h>
#include <linux/file.h>
#include <linux/fs.h>
24
#include <linux/fsnotify.h>
C
Christoph Hellwig 已提交
25 26 27 28 29 30
#include <linux/pagemap.h>
#include <linux/highmem.h>
#include <linux/time.h>
#include <linux/init.h>
#include <linux/string.h>
#include <linux/backing-dev.h>
31
#include <linux/mount.h>
C
Christoph Hellwig 已提交
32
#include <linux/mpage.h>
33
#include <linux/namei.h>
C
Christoph Hellwig 已提交
34 35 36 37 38
#include <linux/swap.h>
#include <linux/writeback.h>
#include <linux/statfs.h>
#include <linux/compat.h>
#include <linux/bit_spinlock.h>
39
#include <linux/security.h>
C
Christoph Hellwig 已提交
40
#include <linux/xattr.h>
41
#include <linux/vmalloc.h>
42
#include <linux/slab.h>
43
#include <linux/blkdev.h>
C
Chris Mason 已提交
44
#include "compat.h"
C
Christoph Hellwig 已提交
45 46 47 48 49 50 51
#include "ctree.h"
#include "disk-io.h"
#include "transaction.h"
#include "btrfs_inode.h"
#include "ioctl.h"
#include "print-tree.h"
#include "volumes.h"
52
#include "locking.h"
53
#include "inode-map.h"
54
#include "backref.h"
55
#include "rcu-string.h"
C
Christoph Hellwig 已提交
56

57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86
/* Mask out flags that are inappropriate for the given type of inode. */
static inline __u32 btrfs_mask_flags(umode_t mode, __u32 flags)
{
	if (S_ISDIR(mode))
		return flags;
	else if (S_ISREG(mode))
		return flags & ~FS_DIRSYNC_FL;
	else
		return flags & (FS_NODUMP_FL | FS_NOATIME_FL);
}

/*
 * Export inode flags to the format expected by the FS_IOC_GETFLAGS ioctl.
 */
static unsigned int btrfs_flags_to_ioctl(unsigned int flags)
{
	unsigned int iflags = 0;

	if (flags & BTRFS_INODE_SYNC)
		iflags |= FS_SYNC_FL;
	if (flags & BTRFS_INODE_IMMUTABLE)
		iflags |= FS_IMMUTABLE_FL;
	if (flags & BTRFS_INODE_APPEND)
		iflags |= FS_APPEND_FL;
	if (flags & BTRFS_INODE_NODUMP)
		iflags |= FS_NODUMP_FL;
	if (flags & BTRFS_INODE_NOATIME)
		iflags |= FS_NOATIME_FL;
	if (flags & BTRFS_INODE_DIRSYNC)
		iflags |= FS_DIRSYNC_FL;
L
Li Zefan 已提交
87 88 89 90 91 92 93
	if (flags & BTRFS_INODE_NODATACOW)
		iflags |= FS_NOCOW_FL;

	if ((flags & BTRFS_INODE_COMPRESS) && !(flags & BTRFS_INODE_NOCOMPRESS))
		iflags |= FS_COMPR_FL;
	else if (flags & BTRFS_INODE_NOCOMPRESS)
		iflags |= FS_NOCOMP_FL;
94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121

	return iflags;
}

/*
 * Update inode->i_flags based on the btrfs internal flags.
 */
void btrfs_update_iflags(struct inode *inode)
{
	struct btrfs_inode *ip = BTRFS_I(inode);

	inode->i_flags &= ~(S_SYNC|S_APPEND|S_IMMUTABLE|S_NOATIME|S_DIRSYNC);

	if (ip->flags & BTRFS_INODE_SYNC)
		inode->i_flags |= S_SYNC;
	if (ip->flags & BTRFS_INODE_IMMUTABLE)
		inode->i_flags |= S_IMMUTABLE;
	if (ip->flags & BTRFS_INODE_APPEND)
		inode->i_flags |= S_APPEND;
	if (ip->flags & BTRFS_INODE_NOATIME)
		inode->i_flags |= S_NOATIME;
	if (ip->flags & BTRFS_INODE_DIRSYNC)
		inode->i_flags |= S_DIRSYNC;
}

/*
 * Inherit flags from the parent inode.
 *
122
 * Currently only the compression flags and the cow flags are inherited.
123 124 125
 */
void btrfs_inherit_iflags(struct inode *inode, struct inode *dir)
{
126 127 128 129 130 131
	unsigned int flags;

	if (!dir)
		return;

	flags = BTRFS_I(dir)->flags;
132

133 134 135 136 137 138 139 140 141 142
	if (flags & BTRFS_INODE_NOCOMPRESS) {
		BTRFS_I(inode)->flags &= ~BTRFS_INODE_COMPRESS;
		BTRFS_I(inode)->flags |= BTRFS_INODE_NOCOMPRESS;
	} else if (flags & BTRFS_INODE_COMPRESS) {
		BTRFS_I(inode)->flags &= ~BTRFS_INODE_NOCOMPRESS;
		BTRFS_I(inode)->flags |= BTRFS_INODE_COMPRESS;
	}

	if (flags & BTRFS_INODE_NODATACOW)
		BTRFS_I(inode)->flags |= BTRFS_INODE_NODATACOW;
143 144 145 146 147 148 149 150 151 152 153 154 155 156

	btrfs_update_iflags(inode);
}

static int btrfs_ioctl_getflags(struct file *file, void __user *arg)
{
	struct btrfs_inode *ip = BTRFS_I(file->f_path.dentry->d_inode);
	unsigned int flags = btrfs_flags_to_ioctl(ip->flags);

	if (copy_to_user(arg, &flags, sizeof(flags)))
		return -EFAULT;
	return 0;
}

157 158 159 160 161
static int check_flags(unsigned int flags)
{
	if (flags & ~(FS_IMMUTABLE_FL | FS_APPEND_FL | \
		      FS_NOATIME_FL | FS_NODUMP_FL | \
		      FS_SYNC_FL | FS_DIRSYNC_FL | \
L
Li Zefan 已提交
162 163
		      FS_NOCOMP_FL | FS_COMPR_FL |
		      FS_NOCOW_FL))
164 165 166 167 168 169 170 171
		return -EOPNOTSUPP;

	if ((flags & FS_NOCOMP_FL) && (flags & FS_COMPR_FL))
		return -EINVAL;

	return 0;
}

172 173 174 175 176 177 178 179
static int btrfs_ioctl_setflags(struct file *file, void __user *arg)
{
	struct inode *inode = file->f_path.dentry->d_inode;
	struct btrfs_inode *ip = BTRFS_I(inode);
	struct btrfs_root *root = ip->root;
	struct btrfs_trans_handle *trans;
	unsigned int flags, oldflags;
	int ret;
180 181
	u64 ip_oldflags;
	unsigned int i_oldflags;
182

L
Li Zefan 已提交
183 184 185
	if (btrfs_root_readonly(root))
		return -EROFS;

186 187 188
	if (copy_from_user(&flags, arg, sizeof(flags)))
		return -EFAULT;

189 190 191
	ret = check_flags(flags);
	if (ret)
		return ret;
C
Christoph Hellwig 已提交
192

193
	if (!inode_owner_or_capable(inode))
194 195 196 197
		return -EACCES;

	mutex_lock(&inode->i_mutex);

198 199 200
	ip_oldflags = ip->flags;
	i_oldflags = inode->i_flags;

201 202 203 204 205 206 207 208 209
	flags = btrfs_mask_flags(inode->i_mode, flags);
	oldflags = btrfs_flags_to_ioctl(ip->flags);
	if ((flags ^ oldflags) & (FS_APPEND_FL | FS_IMMUTABLE_FL)) {
		if (!capable(CAP_LINUX_IMMUTABLE)) {
			ret = -EPERM;
			goto out_unlock;
		}
	}

210
	ret = mnt_want_write_file(file);
211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237
	if (ret)
		goto out_unlock;

	if (flags & FS_SYNC_FL)
		ip->flags |= BTRFS_INODE_SYNC;
	else
		ip->flags &= ~BTRFS_INODE_SYNC;
	if (flags & FS_IMMUTABLE_FL)
		ip->flags |= BTRFS_INODE_IMMUTABLE;
	else
		ip->flags &= ~BTRFS_INODE_IMMUTABLE;
	if (flags & FS_APPEND_FL)
		ip->flags |= BTRFS_INODE_APPEND;
	else
		ip->flags &= ~BTRFS_INODE_APPEND;
	if (flags & FS_NODUMP_FL)
		ip->flags |= BTRFS_INODE_NODUMP;
	else
		ip->flags &= ~BTRFS_INODE_NODUMP;
	if (flags & FS_NOATIME_FL)
		ip->flags |= BTRFS_INODE_NOATIME;
	else
		ip->flags &= ~BTRFS_INODE_NOATIME;
	if (flags & FS_DIRSYNC_FL)
		ip->flags |= BTRFS_INODE_DIRSYNC;
	else
		ip->flags &= ~BTRFS_INODE_DIRSYNC;
L
Li Zefan 已提交
238 239 240 241
	if (flags & FS_NOCOW_FL)
		ip->flags |= BTRFS_INODE_NODATACOW;
	else
		ip->flags &= ~BTRFS_INODE_NODATACOW;
242

243 244 245 246 247 248 249 250 251 252 253
	/*
	 * The COMPRESS flag can only be changed by users, while the NOCOMPRESS
	 * flag may be changed automatically if compression code won't make
	 * things smaller.
	 */
	if (flags & FS_NOCOMP_FL) {
		ip->flags &= ~BTRFS_INODE_COMPRESS;
		ip->flags |= BTRFS_INODE_NOCOMPRESS;
	} else if (flags & FS_COMPR_FL) {
		ip->flags |= BTRFS_INODE_COMPRESS;
		ip->flags &= ~BTRFS_INODE_NOCOMPRESS;
L
Li Zefan 已提交
254 255
	} else {
		ip->flags &= ~(BTRFS_INODE_COMPRESS | BTRFS_INODE_NOCOMPRESS);
256
	}
257

258
	trans = btrfs_start_transaction(root, 1);
259 260 261 262
	if (IS_ERR(trans)) {
		ret = PTR_ERR(trans);
		goto out_drop;
	}
263

264
	btrfs_update_iflags(inode);
265
	inode_inc_iversion(inode);
266
	inode->i_ctime = CURRENT_TIME;
267 268 269
	ret = btrfs_update_inode(trans, root, inode);

	btrfs_end_transaction(trans, root);
270 271 272 273 274
 out_drop:
	if (ret) {
		ip->flags = ip_oldflags;
		inode->i_flags = i_oldflags;
	}
275

A
Al Viro 已提交
276
	mnt_drop_write_file(file);
277 278
 out_unlock:
	mutex_unlock(&inode->i_mutex);
279
	return ret;
280 281 282 283 284 285 286 287
}

static int btrfs_ioctl_getversion(struct file *file, int __user *arg)
{
	struct inode *inode = file->f_path.dentry->d_inode;

	return put_user(inode->i_generation, arg);
}
C
Christoph Hellwig 已提交
288

289 290
static noinline int btrfs_ioctl_fitrim(struct file *file, void __user *arg)
{
291
	struct btrfs_fs_info *fs_info = btrfs_sb(fdentry(file)->d_sb);
292 293 294 295 296
	struct btrfs_device *device;
	struct request_queue *q;
	struct fstrim_range range;
	u64 minlen = ULLONG_MAX;
	u64 num_devices = 0;
297
	u64 total_bytes = btrfs_super_total_bytes(fs_info->super_copy);
298 299 300 301 302
	int ret;

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

303 304 305
	rcu_read_lock();
	list_for_each_entry_rcu(device, &fs_info->fs_devices->devices,
				dev_list) {
306 307 308 309 310 311 312 313 314
		if (!device->bdev)
			continue;
		q = bdev_get_queue(device->bdev);
		if (blk_queue_discard(q)) {
			num_devices++;
			minlen = min((u64)q->limits.discard_granularity,
				     minlen);
		}
	}
315
	rcu_read_unlock();
316

317 318 319 320
	if (!num_devices)
		return -EOPNOTSUPP;
	if (copy_from_user(&range, arg, sizeof(range)))
		return -EFAULT;
321 322
	if (range.start > total_bytes)
		return -EINVAL;
323

324
	range.len = min(range.len, total_bytes - range.start);
325
	range.minlen = max(range.minlen, minlen);
326
	ret = btrfs_trim_fs(fs_info->tree_root, &range);
327 328 329 330 331 332 333 334 335
	if (ret < 0)
		return ret;

	if (copy_to_user(arg, &range, sizeof(range)))
		return -EFAULT;

	return 0;
}

336 337
static noinline int create_subvol(struct btrfs_root *root,
				  struct dentry *dentry,
S
Sage Weil 已提交
338 339
				  char *name, int namelen,
				  u64 *async_transid)
C
Christoph Hellwig 已提交
340 341 342 343 344 345
{
	struct btrfs_trans_handle *trans;
	struct btrfs_key key;
	struct btrfs_root_item root_item;
	struct btrfs_inode_item *inode_item;
	struct extent_buffer *leaf;
346
	struct btrfs_root *new_root;
347
	struct dentry *parent = dentry->d_parent;
348
	struct inode *dir;
C
Christoph Hellwig 已提交
349 350 351 352
	int ret;
	int err;
	u64 objectid;
	u64 new_dirid = BTRFS_FIRST_FREE_OBJECTID;
353
	u64 index = 0;
C
Christoph Hellwig 已提交
354

355
	ret = btrfs_find_free_objectid(root->fs_info->tree_root, &objectid);
356
	if (ret)
357
		return ret;
358 359 360

	dir = parent->d_inode;

J
Josef Bacik 已提交
361 362 363 364 365 366
	/*
	 * 1 - inode item
	 * 2 - refs
	 * 1 - root item
	 * 2 - dir items
	 */
367
	trans = btrfs_start_transaction(root, 6);
368
	if (IS_ERR(trans))
369
		return PTR_ERR(trans);
C
Christoph Hellwig 已提交
370

371
	leaf = btrfs_alloc_free_block(trans, root, root->leafsize,
372
				      0, objectid, NULL, 0, 0, 0);
373 374 375 376
	if (IS_ERR(leaf)) {
		ret = PTR_ERR(leaf);
		goto fail;
	}
C
Christoph Hellwig 已提交
377

378
	memset_extent_buffer(leaf, 0, 0, sizeof(struct btrfs_header));
C
Christoph Hellwig 已提交
379 380
	btrfs_set_header_bytenr(leaf, leaf->start);
	btrfs_set_header_generation(leaf, trans->transid);
381
	btrfs_set_header_backref_rev(leaf, BTRFS_MIXED_BACKREF_REV);
C
Christoph Hellwig 已提交
382 383 384 385 386
	btrfs_set_header_owner(leaf, objectid);

	write_extent_buffer(leaf, root->fs_info->fsid,
			    (unsigned long)btrfs_header_fsid(leaf),
			    BTRFS_FSID_SIZE);
387 388 389
	write_extent_buffer(leaf, root->fs_info->chunk_tree_uuid,
			    (unsigned long)btrfs_header_chunk_tree_uuid(leaf),
			    BTRFS_UUID_SIZE);
C
Christoph Hellwig 已提交
390 391 392 393 394 395 396
	btrfs_mark_buffer_dirty(leaf);

	inode_item = &root_item.inode;
	memset(inode_item, 0, sizeof(*inode_item));
	inode_item->generation = cpu_to_le64(1);
	inode_item->size = cpu_to_le64(3);
	inode_item->nlink = cpu_to_le32(1);
397
	inode_item->nbytes = cpu_to_le64(root->leafsize);
C
Christoph Hellwig 已提交
398 399
	inode_item->mode = cpu_to_le32(S_IFDIR | 0755);

400 401 402 403
	root_item.flags = 0;
	root_item.byte_limit = 0;
	inode_item->flags = cpu_to_le64(BTRFS_INODE_ROOT_ITEM_INIT);

C
Christoph Hellwig 已提交
404
	btrfs_set_root_bytenr(&root_item, leaf->start);
405
	btrfs_set_root_generation(&root_item, trans->transid);
C
Christoph Hellwig 已提交
406 407
	btrfs_set_root_level(&root_item, 0);
	btrfs_set_root_refs(&root_item, 1);
408
	btrfs_set_root_used(&root_item, leaf->len);
Y
Yan Zheng 已提交
409
	btrfs_set_root_last_snapshot(&root_item, 0);
C
Christoph Hellwig 已提交
410 411 412 413

	memset(&root_item.drop_progress, 0, sizeof(root_item.drop_progress));
	root_item.drop_level = 0;

414
	btrfs_tree_unlock(leaf);
C
Christoph Hellwig 已提交
415 416 417 418 419 420
	free_extent_buffer(leaf);
	leaf = NULL;

	btrfs_set_root_dirid(&root_item, new_dirid);

	key.objectid = objectid;
421
	key.offset = 0;
C
Christoph Hellwig 已提交
422 423 424 425 426 427
	btrfs_set_key_type(&key, BTRFS_ROOT_ITEM_KEY);
	ret = btrfs_insert_root(trans, root->fs_info->tree_root, &key,
				&root_item);
	if (ret)
		goto fail;

428 429
	key.offset = (u64)-1;
	new_root = btrfs_read_fs_root_no_name(root->fs_info, &key);
430 431 432 433 434
	if (IS_ERR(new_root)) {
		btrfs_abort_transaction(trans, root, PTR_ERR(new_root));
		ret = PTR_ERR(new_root);
		goto fail;
	}
435 436 437

	btrfs_record_root_in_trans(trans, new_root);

438
	ret = btrfs_create_subvol_root(trans, new_root, new_dirid);
439 440
	if (ret) {
		/* We potentially lose an unused inode item here */
441
		btrfs_abort_transaction(trans, root, ret);
442 443 444
		goto fail;
	}

C
Christoph Hellwig 已提交
445 446 447
	/*
	 * insert the directory item
	 */
448
	ret = btrfs_set_inode_index(dir, &index);
449 450 451 452
	if (ret) {
		btrfs_abort_transaction(trans, root, ret);
		goto fail;
	}
453 454

	ret = btrfs_insert_dir_item(trans, root,
455
				    name, namelen, dir, &key,
456
				    BTRFS_FT_DIR, index);
457 458
	if (ret) {
		btrfs_abort_transaction(trans, root, ret);
C
Christoph Hellwig 已提交
459
		goto fail;
460
	}
461

462 463 464 465
	btrfs_i_size_write(dir, dir->i_size + namelen * 2);
	ret = btrfs_update_inode(trans, root, dir);
	BUG_ON(ret);

466
	ret = btrfs_add_root_ref(trans, root->fs_info->tree_root,
467
				 objectid, root->root_key.objectid,
L
Li Zefan 已提交
468
				 btrfs_ino(dir), index, name, namelen);
469

470
	BUG_ON(ret);
C
Christoph Hellwig 已提交
471

472
	d_instantiate(dentry, btrfs_lookup_dentry(dir, dentry));
C
Christoph Hellwig 已提交
473
fail:
S
Sage Weil 已提交
474 475 476 477 478 479
	if (async_transid) {
		*async_transid = trans->transid;
		err = btrfs_commit_transaction_async(trans, root, 1);
	} else {
		err = btrfs_commit_transaction(trans, root);
	}
C
Christoph Hellwig 已提交
480 481 482 483 484
	if (err && !ret)
		ret = err;
	return ret;
}

S
Sage Weil 已提交
485
static int create_snapshot(struct btrfs_root *root, struct dentry *dentry,
L
Li Zefan 已提交
486 487
			   char *name, int namelen, u64 *async_transid,
			   bool readonly)
C
Christoph Hellwig 已提交
488
{
489
	struct inode *inode;
C
Christoph Hellwig 已提交
490 491
	struct btrfs_pending_snapshot *pending_snapshot;
	struct btrfs_trans_handle *trans;
492
	int ret;
C
Christoph Hellwig 已提交
493 494 495 496

	if (!root->ref_cows)
		return -EINVAL;

497
	pending_snapshot = kzalloc(sizeof(*pending_snapshot), GFP_NOFS);
498 499 500 501
	if (!pending_snapshot)
		return -ENOMEM;

	btrfs_init_block_rsv(&pending_snapshot->block_rsv);
502
	pending_snapshot->dentry = dentry;
C
Christoph Hellwig 已提交
503
	pending_snapshot->root = root;
L
Li Zefan 已提交
504
	pending_snapshot->readonly = readonly;
505 506 507 508 509 510 511 512 513 514

	trans = btrfs_start_transaction(root->fs_info->extent_root, 5);
	if (IS_ERR(trans)) {
		ret = PTR_ERR(trans);
		goto fail;
	}

	ret = btrfs_snap_reserve_metadata(trans, pending_snapshot);
	BUG_ON(ret);

515
	spin_lock(&root->fs_info->trans_lock);
C
Christoph Hellwig 已提交
516 517
	list_add(&pending_snapshot->list,
		 &trans->transaction->pending_snapshots);
518
	spin_unlock(&root->fs_info->trans_lock);
S
Sage Weil 已提交
519 520 521 522 523 524 525 526
	if (async_transid) {
		*async_transid = trans->transid;
		ret = btrfs_commit_transaction_async(trans,
				     root->fs_info->extent_root, 1);
	} else {
		ret = btrfs_commit_transaction(trans,
					       root->fs_info->extent_root);
	}
527
	BUG_ON(ret);
528 529 530 531 532

	ret = pending_snapshot->error;
	if (ret)
		goto fail;

533 534 535
	ret = btrfs_orphan_cleanup(pending_snapshot->snap);
	if (ret)
		goto fail;
C
Christoph Hellwig 已提交
536

537
	inode = btrfs_lookup_dentry(dentry->d_parent->d_inode, dentry);
538 539 540 541 542 543 544 545
	if (IS_ERR(inode)) {
		ret = PTR_ERR(inode);
		goto fail;
	}
	BUG_ON(!inode);
	d_instantiate(dentry, inode);
	ret = 0;
fail:
546
	kfree(pending_snapshot);
C
Christoph Hellwig 已提交
547 548 549
	return ret;
}

550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619
/*  copy of check_sticky in fs/namei.c()
* It's inline, so penalty for filesystems that don't use sticky bit is
* minimal.
*/
static inline int btrfs_check_sticky(struct inode *dir, struct inode *inode)
{
	uid_t fsuid = current_fsuid();

	if (!(dir->i_mode & S_ISVTX))
		return 0;
	if (inode->i_uid == fsuid)
		return 0;
	if (dir->i_uid == fsuid)
		return 0;
	return !capable(CAP_FOWNER);
}

/*  copy of may_delete in fs/namei.c()
 *	Check whether we can remove a link victim from directory dir, check
 *  whether the type of victim is right.
 *  1. We can't do it if dir is read-only (done in permission())
 *  2. We should have write and exec permissions on dir
 *  3. We can't remove anything from append-only dir
 *  4. We can't do anything with immutable dir (done in permission())
 *  5. If the sticky bit on dir is set we should either
 *	a. be owner of dir, or
 *	b. be owner of victim, or
 *	c. have CAP_FOWNER capability
 *  6. If the victim is append-only or immutable we can't do antyhing with
 *     links pointing to it.
 *  7. If we were asked to remove a directory and victim isn't one - ENOTDIR.
 *  8. If we were asked to remove a non-directory and victim isn't one - EISDIR.
 *  9. We can't remove a root or mountpoint.
 * 10. We don't allow removal of NFS sillyrenamed files; it's handled by
 *     nfs_async_unlink().
 */

static int btrfs_may_delete(struct inode *dir,struct dentry *victim,int isdir)
{
	int error;

	if (!victim->d_inode)
		return -ENOENT;

	BUG_ON(victim->d_parent->d_inode != dir);
	audit_inode_child(victim, dir);

	error = inode_permission(dir, MAY_WRITE | MAY_EXEC);
	if (error)
		return error;
	if (IS_APPEND(dir))
		return -EPERM;
	if (btrfs_check_sticky(dir, victim->d_inode)||
		IS_APPEND(victim->d_inode)||
	    IS_IMMUTABLE(victim->d_inode) || IS_SWAPFILE(victim->d_inode))
		return -EPERM;
	if (isdir) {
		if (!S_ISDIR(victim->d_inode->i_mode))
			return -ENOTDIR;
		if (IS_ROOT(victim))
			return -EBUSY;
	} else if (S_ISDIR(victim->d_inode->i_mode))
		return -EISDIR;
	if (IS_DEADDIR(dir))
		return -ENOENT;
	if (victim->d_flags & DCACHE_NFSFS_RENAMED)
		return -EBUSY;
	return 0;
}

620 621 622 623 624 625 626 627 628 629 630 631 632 633 634
/* copy of may_create in fs/namei.c() */
static inline int btrfs_may_create(struct inode *dir, struct dentry *child)
{
	if (child->d_inode)
		return -EEXIST;
	if (IS_DEADDIR(dir))
		return -ENOENT;
	return inode_permission(dir, MAY_WRITE | MAY_EXEC);
}

/*
 * Create a new subvolume below @parent.  This is largely modeled after
 * sys_mkdirat and vfs_mkdir, but we only do a single component lookup
 * inside this filesystem so it's quite a bit simpler.
 */
635 636
static noinline int btrfs_mksubvol(struct path *parent,
				   char *name, int namelen,
S
Sage Weil 已提交
637
				   struct btrfs_root *snap_src,
L
Li Zefan 已提交
638
				   u64 *async_transid, bool readonly)
639
{
640
	struct inode *dir  = parent->dentry->d_inode;
641 642 643
	struct dentry *dentry;
	int error;

644
	mutex_lock_nested(&dir->i_mutex, I_MUTEX_PARENT);
645 646 647 648 649 650 651 652 653 654 655 656 657 658

	dentry = lookup_one_len(name, parent->dentry, namelen);
	error = PTR_ERR(dentry);
	if (IS_ERR(dentry))
		goto out_unlock;

	error = -EEXIST;
	if (dentry->d_inode)
		goto out_dput;

	error = mnt_want_write(parent->mnt);
	if (error)
		goto out_dput;

659
	error = btrfs_may_create(dir, dentry);
660 661 662
	if (error)
		goto out_drop_write;

663 664 665 666 667
	down_read(&BTRFS_I(dir)->root->fs_info->subvol_sem);

	if (btrfs_root_refs(&BTRFS_I(dir)->root->root_item) == 0)
		goto out_up_read;

668
	if (snap_src) {
S
Sage Weil 已提交
669
		error = create_snapshot(snap_src, dentry,
L
Li Zefan 已提交
670
					name, namelen, async_transid, readonly);
671
	} else {
672
		error = create_subvol(BTRFS_I(dir)->root, dentry,
S
Sage Weil 已提交
673
				      name, namelen, async_transid);
674
	}
675 676 677 678
	if (!error)
		fsnotify_mkdir(dir, dentry);
out_up_read:
	up_read(&BTRFS_I(dir)->root->fs_info->subvol_sem);
679 680 681 682 683
out_drop_write:
	mnt_drop_write(parent->mnt);
out_dput:
	dput(dentry);
out_unlock:
684
	mutex_unlock(&dir->i_mutex);
685 686 687
	return error;
}

C
Chris Mason 已提交
688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738
/*
 * When we're defragging a range, we don't want to kick it off again
 * if it is really just waiting for delalloc to send it down.
 * If we find a nice big extent or delalloc range for the bytes in the
 * file you want to defrag, we return 0 to let you know to skip this
 * part of the file
 */
static int check_defrag_in_cache(struct inode *inode, u64 offset, int thresh)
{
	struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree;
	struct extent_map *em = NULL;
	struct extent_map_tree *em_tree = &BTRFS_I(inode)->extent_tree;
	u64 end;

	read_lock(&em_tree->lock);
	em = lookup_extent_mapping(em_tree, offset, PAGE_CACHE_SIZE);
	read_unlock(&em_tree->lock);

	if (em) {
		end = extent_map_end(em);
		free_extent_map(em);
		if (end - offset > thresh)
			return 0;
	}
	/* if we already have a nice delalloc here, just stop */
	thresh /= 2;
	end = count_range_bits(io_tree, &offset, offset + thresh,
			       thresh, EXTENT_DELALLOC, 1);
	if (end >= thresh)
		return 0;
	return 1;
}

/*
 * helper function to walk through a file and find extents
 * newer than a specific transid, and smaller than thresh.
 *
 * This is used by the defragging code to find new and small
 * extents
 */
static int find_new_extents(struct btrfs_root *root,
			    struct inode *inode, u64 newer_than,
			    u64 *off, int thresh)
{
	struct btrfs_path *path;
	struct btrfs_key min_key;
	struct btrfs_key max_key;
	struct extent_buffer *leaf;
	struct btrfs_file_extent_item *extent;
	int type;
	int ret;
739
	u64 ino = btrfs_ino(inode);
C
Chris Mason 已提交
740 741 742 743 744

	path = btrfs_alloc_path();
	if (!path)
		return -ENOMEM;

745
	min_key.objectid = ino;
C
Chris Mason 已提交
746 747 748
	min_key.type = BTRFS_EXTENT_DATA_KEY;
	min_key.offset = *off;

749
	max_key.objectid = ino;
C
Chris Mason 已提交
750 751 752 753 754 755 756 757 758 759
	max_key.type = (u8)-1;
	max_key.offset = (u64)-1;

	path->keep_locks = 1;

	while(1) {
		ret = btrfs_search_forward(root, &min_key, &max_key,
					   path, 0, newer_than);
		if (ret != 0)
			goto none;
760
		if (min_key.objectid != ino)
C
Chris Mason 已提交
761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788
			goto none;
		if (min_key.type != BTRFS_EXTENT_DATA_KEY)
			goto none;

		leaf = path->nodes[0];
		extent = btrfs_item_ptr(leaf, path->slots[0],
					struct btrfs_file_extent_item);

		type = btrfs_file_extent_type(leaf, extent);
		if (type == BTRFS_FILE_EXTENT_REG &&
		    btrfs_file_extent_num_bytes(leaf, extent) < thresh &&
		    check_defrag_in_cache(inode, min_key.offset, thresh)) {
			*off = min_key.offset;
			btrfs_free_path(path);
			return 0;
		}

		if (min_key.offset == (u64)-1)
			goto none;

		min_key.offset++;
		btrfs_release_path(path);
	}
none:
	btrfs_free_path(path);
	return -ENOENT;
}

L
Li Zefan 已提交
789
static struct extent_map *defrag_lookup_extent(struct inode *inode, u64 start)
790 791
{
	struct extent_map_tree *em_tree = &BTRFS_I(inode)->extent_tree;
L
Li Zefan 已提交
792 793 794
	struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree;
	struct extent_map *em;
	u64 len = PAGE_CACHE_SIZE;
795

L
Li Zefan 已提交
796 797 798 799
	/*
	 * hopefully we have this extent in the tree already, try without
	 * the full extent lock
	 */
800
	read_lock(&em_tree->lock);
L
Li Zefan 已提交
801
	em = lookup_extent_mapping(em_tree, start, len);
802 803
	read_unlock(&em_tree->lock);

L
Li Zefan 已提交
804 805 806 807 808 809 810 811 812 813 814 815
	if (!em) {
		/* get the big lock and read metadata off disk */
		lock_extent(io_tree, start, start + len - 1);
		em = btrfs_get_extent(inode, NULL, 0, start, len, 0);
		unlock_extent(io_tree, start, start + len - 1);

		if (IS_ERR(em))
			return NULL;
	}

	return em;
}
816

L
Li Zefan 已提交
817 818 819 820 821 822 823 824 825 826 827 828 829 830
static bool defrag_check_next_extent(struct inode *inode, struct extent_map *em)
{
	struct extent_map *next;
	bool ret = true;

	/* this is the last extent */
	if (em->start + em->len >= i_size_read(inode))
		return false;

	next = defrag_lookup_extent(inode, em->start + em->len);
	if (!next || next->block_start >= EXTENT_MAP_LAST_BYTE)
		ret = false;

	free_extent_map(next);
831 832 833
	return ret;
}

L
Li Zefan 已提交
834 835
static int should_defrag_range(struct inode *inode, u64 start, int thresh,
			       u64 *last_len, u64 *skip, u64 *defrag_end)
836
{
L
Li Zefan 已提交
837
	struct extent_map *em;
838
	int ret = 1;
L
Li Zefan 已提交
839
	bool next_mergeable = true;
840 841

	/*
842
	 * make sure that once we start defragging an extent, we keep on
843 844 845 846 847 848 849
	 * defragging it
	 */
	if (start < *defrag_end)
		return 1;

	*skip = 0;

L
Li Zefan 已提交
850 851 852
	em = defrag_lookup_extent(inode, start);
	if (!em)
		return 0;
853 854

	/* this will cover holes, and inline extents */
855
	if (em->block_start >= EXTENT_MAP_LAST_BYTE) {
856
		ret = 0;
857 858 859
		goto out;
	}

L
Li Zefan 已提交
860
	next_mergeable = defrag_check_next_extent(inode, em);
861 862

	/*
L
Li Zefan 已提交
863 864
	 * we hit a real extent, if it is big or the next extent is not a
	 * real extent, don't bother defragging it
865
	 */
L
Li Zefan 已提交
866 867
	if ((*last_len == 0 || *last_len >= thresh) &&
	    (em->len >= thresh || !next_mergeable))
868
		ret = 0;
869
out:
870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889
	/*
	 * last_len ends up being a counter of how many bytes we've defragged.
	 * every time we choose not to defrag an extent, we reset *last_len
	 * so that the next tiny extent will force a defrag.
	 *
	 * The end result of this is that tiny extents before a single big
	 * extent will force at least part of that big extent to be defragged.
	 */
	if (ret) {
		*defrag_end = extent_map_end(em);
	} else {
		*last_len = 0;
		*skip = extent_map_end(em);
		*defrag_end = 0;
	}

	free_extent_map(em);
	return ret;
}

C
Chris Mason 已提交
890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905
/*
 * it doesn't do much good to defrag one or two pages
 * at a time.  This pulls in a nice chunk of pages
 * to COW and defrag.
 *
 * It also makes sure the delalloc code has enough
 * dirty data to avoid making new small extents as part
 * of the defrag
 *
 * It's a good idea to start RA on this range
 * before calling this.
 */
static int cluster_pages_for_defrag(struct inode *inode,
				    struct page **pages,
				    unsigned long start_index,
				    int num_pages)
C
Christoph Hellwig 已提交
906
{
C
Chris Mason 已提交
907 908 909 910
	unsigned long file_end;
	u64 isize = i_size_read(inode);
	u64 page_start;
	u64 page_end;
911
	u64 page_cnt;
C
Chris Mason 已提交
912 913 914
	int ret;
	int i;
	int i_done;
915
	struct btrfs_ordered_extent *ordered;
C
Chris Mason 已提交
916
	struct extent_state *cached_state = NULL;
917
	struct extent_io_tree *tree;
918
	gfp_t mask = btrfs_alloc_write_mask(inode->i_mapping);
C
Chris Mason 已提交
919 920

	file_end = (isize - 1) >> PAGE_CACHE_SHIFT;
921 922 923 924
	if (!isize || start_index > file_end)
		return 0;

	page_cnt = min_t(u64, (u64)num_pages, (u64)file_end - start_index + 1);
C
Chris Mason 已提交
925 926

	ret = btrfs_delalloc_reserve_space(inode,
927
					   page_cnt << PAGE_CACHE_SHIFT);
C
Chris Mason 已提交
928 929 930
	if (ret)
		return ret;
	i_done = 0;
931
	tree = &BTRFS_I(inode)->io_tree;
C
Chris Mason 已提交
932 933

	/* step one, lock all the pages */
934
	for (i = 0; i < page_cnt; i++) {
C
Chris Mason 已提交
935
		struct page *page;
936
again:
937
		page = find_or_create_page(inode->i_mapping,
938
					   start_index + i, mask);
C
Chris Mason 已提交
939 940 941
		if (!page)
			break;

942 943 944
		page_start = page_offset(page);
		page_end = page_start + PAGE_CACHE_SIZE - 1;
		while (1) {
945
			lock_extent(tree, page_start, page_end);
946 947
			ordered = btrfs_lookup_ordered_extent(inode,
							      page_start);
948
			unlock_extent(tree, page_start, page_end);
949 950 951 952 953 954 955
			if (!ordered)
				break;

			unlock_page(page);
			btrfs_start_ordered_extent(inode, ordered, 1);
			btrfs_put_ordered_extent(ordered);
			lock_page(page);
956 957 958 959 960 961 962 963 964
			/*
			 * we unlocked the page above, so we need check if
			 * it was released or not.
			 */
			if (page->mapping != inode->i_mapping) {
				unlock_page(page);
				page_cache_release(page);
				goto again;
			}
965 966
		}

C
Chris Mason 已提交
967 968 969 970 971 972 973 974 975 976
		if (!PageUptodate(page)) {
			btrfs_readpage(NULL, page);
			lock_page(page);
			if (!PageUptodate(page)) {
				unlock_page(page);
				page_cache_release(page);
				ret = -EIO;
				break;
			}
		}
977 978 979 980 981 982 983

		if (page->mapping != inode->i_mapping) {
			unlock_page(page);
			page_cache_release(page);
			goto again;
		}

C
Chris Mason 已提交
984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003
		pages[i] = page;
		i_done++;
	}
	if (!i_done || ret)
		goto out;

	if (!(inode->i_sb->s_flags & MS_ACTIVE))
		goto out;

	/*
	 * so now we have a nice long stream of locked
	 * and up to date pages, lets wait on them
	 */
	for (i = 0; i < i_done; i++)
		wait_on_page_writeback(pages[i]);

	page_start = page_offset(pages[0]);
	page_end = page_offset(pages[i_done - 1]) + PAGE_CACHE_SIZE;

	lock_extent_bits(&BTRFS_I(inode)->io_tree,
1004
			 page_start, page_end - 1, 0, &cached_state);
C
Chris Mason 已提交
1005 1006 1007 1008 1009
	clear_extent_bit(&BTRFS_I(inode)->io_tree, page_start,
			  page_end - 1, EXTENT_DIRTY | EXTENT_DELALLOC |
			  EXTENT_DO_ACCOUNTING, 0, 0, &cached_state,
			  GFP_NOFS);

1010
	if (i_done != page_cnt) {
1011 1012 1013
		spin_lock(&BTRFS_I(inode)->lock);
		BTRFS_I(inode)->outstanding_extents++;
		spin_unlock(&BTRFS_I(inode)->lock);
C
Chris Mason 已提交
1014
		btrfs_delalloc_release_space(inode,
1015
				     (page_cnt - i_done) << PAGE_CACHE_SHIFT);
C
Chris Mason 已提交
1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039
	}


	btrfs_set_extent_delalloc(inode, page_start, page_end - 1,
				  &cached_state);

	unlock_extent_cached(&BTRFS_I(inode)->io_tree,
			     page_start, page_end - 1, &cached_state,
			     GFP_NOFS);

	for (i = 0; i < i_done; i++) {
		clear_page_dirty_for_io(pages[i]);
		ClearPageChecked(pages[i]);
		set_page_extent_mapped(pages[i]);
		set_page_dirty(pages[i]);
		unlock_page(pages[i]);
		page_cache_release(pages[i]);
	}
	return i_done;
out:
	for (i = 0; i < i_done; i++) {
		unlock_page(pages[i]);
		page_cache_release(pages[i]);
	}
1040
	btrfs_delalloc_release_space(inode, page_cnt << PAGE_CACHE_SHIFT);
C
Chris Mason 已提交
1041 1042 1043 1044 1045 1046 1047 1048 1049
	return ret;

}

int btrfs_defrag_file(struct inode *inode, struct file *file,
		      struct btrfs_ioctl_defrag_range_args *range,
		      u64 newer_than, unsigned long max_to_defrag)
{
	struct btrfs_root *root = BTRFS_I(inode)->root;
1050
	struct btrfs_super_block *disk_super;
C
Chris Mason 已提交
1051
	struct file_ra_state *ra = NULL;
C
Christoph Hellwig 已提交
1052
	unsigned long last_index;
1053
	u64 isize = i_size_read(inode);
1054
	u64 features;
1055 1056 1057
	u64 last_len = 0;
	u64 skip = 0;
	u64 defrag_end = 0;
C
Chris Mason 已提交
1058
	u64 newer_off = range->start;
C
Christoph Hellwig 已提交
1059
	unsigned long i;
1060
	unsigned long ra_index = 0;
C
Christoph Hellwig 已提交
1061
	int ret;
C
Chris Mason 已提交
1062
	int defrag_count = 0;
1063
	int compress_type = BTRFS_COMPRESS_ZLIB;
C
Chris Mason 已提交
1064
	int extent_thresh = range->extent_thresh;
1065 1066
	int max_cluster = (256 * 1024) >> PAGE_CACHE_SHIFT;
	int cluster = max_cluster;
C
Chris Mason 已提交
1067 1068 1069 1070 1071
	u64 new_align = ~((u64)128 * 1024 - 1);
	struct page **pages = NULL;

	if (extent_thresh == 0)
		extent_thresh = 256 * 1024;
1072 1073 1074 1075 1076 1077 1078

	if (range->flags & BTRFS_DEFRAG_RANGE_COMPRESS) {
		if (range->compress_type > BTRFS_COMPRESS_TYPES)
			return -EINVAL;
		if (range->compress_type)
			compress_type = range->compress_type;
	}
C
Christoph Hellwig 已提交
1079

1080
	if (isize == 0)
1081 1082
		return 0;

C
Chris Mason 已提交
1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095
	/*
	 * if we were not given a file, allocate a readahead
	 * context
	 */
	if (!file) {
		ra = kzalloc(sizeof(*ra), GFP_NOFS);
		if (!ra)
			return -ENOMEM;
		file_ra_state_init(ra, inode->i_mapping);
	} else {
		ra = &file->f_ra;
	}

1096
	pages = kmalloc(sizeof(struct page *) * max_cluster,
C
Chris Mason 已提交
1097 1098 1099 1100 1101 1102 1103
			GFP_NOFS);
	if (!pages) {
		ret = -ENOMEM;
		goto out_ra;
	}

	/* find the last page to defrag */
C
Chris Mason 已提交
1104
	if (range->start + range->len > range->start) {
1105
		last_index = min_t(u64, isize - 1,
C
Chris Mason 已提交
1106 1107
			 range->start + range->len - 1) >> PAGE_CACHE_SHIFT;
	} else {
1108
		last_index = (isize - 1) >> PAGE_CACHE_SHIFT;
C
Chris Mason 已提交
1109 1110
	}

C
Chris Mason 已提交
1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126
	if (newer_than) {
		ret = find_new_extents(root, inode, newer_than,
				       &newer_off, 64 * 1024);
		if (!ret) {
			range->start = newer_off;
			/*
			 * we always align our defrag to help keep
			 * the extents in the file evenly spaced
			 */
			i = (newer_off & new_align) >> PAGE_CACHE_SHIFT;
		} else
			goto out_ra;
	} else {
		i = range->start >> PAGE_CACHE_SHIFT;
	}
	if (!max_to_defrag)
L
Liu Bo 已提交
1127
		max_to_defrag = last_index + 1;
C
Chris Mason 已提交
1128

L
Li Zefan 已提交
1129 1130 1131 1132 1133 1134 1135
	/*
	 * make writeback starts from i, so the defrag range can be
	 * written sequentially.
	 */
	if (i < inode->i_mapping->writeback_index)
		inode->i_mapping->writeback_index = i;

1136 1137 1138
	while (i <= last_index && defrag_count < max_to_defrag &&
	       (i < (i_size_read(inode) + PAGE_CACHE_SIZE - 1) >>
		PAGE_CACHE_SHIFT)) {
C
Chris Mason 已提交
1139 1140 1141 1142 1143 1144 1145
		/*
		 * make sure we stop running if someone unmounts
		 * the FS
		 */
		if (!(inode->i_sb->s_flags & MS_ACTIVE))
			break;

1146
		if (!should_defrag_range(inode, (u64)i << PAGE_CACHE_SHIFT,
L
Li Zefan 已提交
1147 1148
					 extent_thresh, &last_len, &skip,
					 &defrag_end)) {
1149 1150 1151 1152 1153 1154 1155 1156 1157
			unsigned long next;
			/*
			 * the should_defrag function tells us how much to skip
			 * bump our counter by the suggested amount
			 */
			next = (skip + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT;
			i = max(i + 1, next);
			continue;
		}
1158 1159 1160 1161 1162 1163 1164 1165 1166

		if (!newer_than) {
			cluster = (PAGE_CACHE_ALIGN(defrag_end) >>
				   PAGE_CACHE_SHIFT) - i;
			cluster = min(cluster, max_cluster);
		} else {
			cluster = max_cluster;
		}

C
Chris Mason 已提交
1167
		if (range->flags & BTRFS_DEFRAG_RANGE_COMPRESS)
1168
			BTRFS_I(inode)->force_compress = compress_type;
1169

1170 1171 1172 1173 1174 1175
		if (i + cluster > ra_index) {
			ra_index = max(i, ra_index);
			btrfs_force_ra(inode->i_mapping, ra, file, ra_index,
				       cluster);
			ra_index += max_cluster;
		}
1176

1177
		mutex_lock(&inode->i_mutex);
1178
		ret = cluster_pages_for_defrag(inode, pages, i, cluster);
1179 1180
		if (ret < 0) {
			mutex_unlock(&inode->i_mutex);
C
Chris Mason 已提交
1181
			goto out_ra;
1182
		}
C
Chris Mason 已提交
1183 1184 1185

		defrag_count += ret;
		balance_dirty_pages_ratelimited_nr(inode->i_mapping, ret);
1186
		mutex_unlock(&inode->i_mutex);
C
Chris Mason 已提交
1187 1188 1189 1190 1191

		if (newer_than) {
			if (newer_off == (u64)-1)
				break;

1192 1193 1194
			if (ret > 0)
				i += ret;

C
Chris Mason 已提交
1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205
			newer_off = max(newer_off + 1,
					(u64)i << PAGE_CACHE_SHIFT);

			ret = find_new_extents(root, inode,
					       newer_than, &newer_off,
					       64 * 1024);
			if (!ret) {
				range->start = newer_off;
				i = (newer_off & new_align) >> PAGE_CACHE_SHIFT;
			} else {
				break;
C
Christoph Hellwig 已提交
1206
			}
C
Chris Mason 已提交
1207
		} else {
1208
			if (ret > 0) {
L
Li Zefan 已提交
1209
				i += ret;
1210 1211
				last_len += ret << PAGE_CACHE_SHIFT;
			} else {
L
Li Zefan 已提交
1212
				i++;
1213 1214
				last_len = 0;
			}
C
Christoph Hellwig 已提交
1215 1216 1217
		}
	}

C
Chris Mason 已提交
1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235
	if ((range->flags & BTRFS_DEFRAG_RANGE_START_IO))
		filemap_flush(inode->i_mapping);

	if ((range->flags & BTRFS_DEFRAG_RANGE_COMPRESS)) {
		/* the filemap_flush will queue IO into the worker threads, but
		 * we have to make sure the IO is actually started and that
		 * ordered extents get created before we return
		 */
		atomic_inc(&root->fs_info->async_submit_draining);
		while (atomic_read(&root->fs_info->nr_async_submits) ||
		      atomic_read(&root->fs_info->async_delalloc_pages)) {
			wait_event(root->fs_info->async_submit_wait,
			   (atomic_read(&root->fs_info->nr_async_submits) == 0 &&
			    atomic_read(&root->fs_info->async_delalloc_pages) == 0));
		}
		atomic_dec(&root->fs_info->async_submit_draining);

		mutex_lock(&inode->i_mutex);
1236
		BTRFS_I(inode)->force_compress = BTRFS_COMPRESS_NONE;
C
Chris Mason 已提交
1237 1238 1239
		mutex_unlock(&inode->i_mutex);
	}

1240
	disk_super = root->fs_info->super_copy;
1241 1242 1243 1244 1245 1246
	features = btrfs_super_incompat_flags(disk_super);
	if (range->compress_type == BTRFS_COMPRESS_LZO) {
		features |= BTRFS_FEATURE_INCOMPAT_COMPRESS_LZO;
		btrfs_set_super_incompat_flags(disk_super, features);
	}

1247
	ret = defrag_count;
1248

C
Chris Mason 已提交
1249 1250 1251 1252
out_ra:
	if (!file)
		kfree(ra);
	kfree(pages);
1253
	return ret;
C
Christoph Hellwig 已提交
1254 1255
}

1256 1257
static noinline int btrfs_ioctl_resize(struct btrfs_root *root,
					void __user *arg)
C
Christoph Hellwig 已提交
1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269
{
	u64 new_size;
	u64 old_size;
	u64 devid = 1;
	struct btrfs_ioctl_vol_args *vol_args;
	struct btrfs_trans_handle *trans;
	struct btrfs_device *device = NULL;
	char *sizestr;
	char *devstr = NULL;
	int ret = 0;
	int mod = 0;

Y
Yan Zheng 已提交
1270 1271 1272
	if (root->fs_info->sb->s_flags & MS_RDONLY)
		return -EROFS;

1273 1274 1275
	if (!capable(CAP_SYS_ADMIN))
		return -EPERM;

1276 1277 1278 1279 1280 1281 1282
	mutex_lock(&root->fs_info->volume_mutex);
	if (root->fs_info->balance_ctl) {
		printk(KERN_INFO "btrfs: balance in progress\n");
		ret = -EINVAL;
		goto out;
	}

L
Li Zefan 已提交
1283
	vol_args = memdup_user(arg, sizeof(*vol_args));
1284 1285 1286 1287
	if (IS_ERR(vol_args)) {
		ret = PTR_ERR(vol_args);
		goto out;
	}
1288 1289

	vol_args->name[BTRFS_PATH_NAME_MAX] = '\0';
C
Christoph Hellwig 已提交
1290 1291 1292 1293 1294 1295 1296 1297 1298

	sizestr = vol_args->name;
	devstr = strchr(sizestr, ':');
	if (devstr) {
		char *end;
		sizestr = devstr + 1;
		*devstr = '\0';
		devstr = vol_args->name;
		devid = simple_strtoull(devstr, &end, 10);
1299
		printk(KERN_INFO "btrfs: resizing devid %llu\n",
1300
		       (unsigned long long)devid);
C
Christoph Hellwig 已提交
1301
	}
Y
Yan Zheng 已提交
1302
	device = btrfs_find_device(root, devid, NULL, NULL);
C
Christoph Hellwig 已提交
1303
	if (!device) {
1304
		printk(KERN_INFO "btrfs: resizer unable to find device %llu\n",
1305
		       (unsigned long long)devid);
C
Christoph Hellwig 已提交
1306
		ret = -EINVAL;
1307
		goto out_free;
C
Christoph Hellwig 已提交
1308
	}
L
Liu Bo 已提交
1309 1310
	if (device->fs_devices && device->fs_devices->seeding) {
		printk(KERN_INFO "btrfs: resizer unable to apply on "
1311 1312
		       "seeding device %llu\n",
		       (unsigned long long)devid);
L
Liu Bo 已提交
1313 1314 1315 1316
		ret = -EINVAL;
		goto out_free;
	}

C
Christoph Hellwig 已提交
1317 1318 1319 1320 1321 1322 1323 1324 1325 1326
	if (!strcmp(sizestr, "max"))
		new_size = device->bdev->bd_inode->i_size;
	else {
		if (sizestr[0] == '-') {
			mod = -1;
			sizestr++;
		} else if (sizestr[0] == '+') {
			mod = 1;
			sizestr++;
		}
A
Akinobu Mita 已提交
1327
		new_size = memparse(sizestr, NULL);
C
Christoph Hellwig 已提交
1328 1329
		if (new_size == 0) {
			ret = -EINVAL;
1330
			goto out_free;
C
Christoph Hellwig 已提交
1331 1332 1333 1334 1335 1336 1337 1338
		}
	}

	old_size = device->total_bytes;

	if (mod < 0) {
		if (new_size > old_size) {
			ret = -EINVAL;
1339
			goto out_free;
C
Christoph Hellwig 已提交
1340 1341 1342 1343 1344 1345 1346 1347
		}
		new_size = old_size - new_size;
	} else if (mod > 0) {
		new_size = old_size + new_size;
	}

	if (new_size < 256 * 1024 * 1024) {
		ret = -EINVAL;
1348
		goto out_free;
C
Christoph Hellwig 已提交
1349 1350 1351
	}
	if (new_size > device->bdev->bd_inode->i_size) {
		ret = -EFBIG;
1352
		goto out_free;
C
Christoph Hellwig 已提交
1353 1354 1355 1356 1357
	}

	do_div(new_size, root->sectorsize);
	new_size *= root->sectorsize;

1358 1359 1360
	printk_in_rcu(KERN_INFO "btrfs: new size for %s is %llu\n",
		      rcu_str_deref(device->name),
		      (unsigned long long)new_size);
C
Christoph Hellwig 已提交
1361 1362

	if (new_size > old_size) {
1363
		trans = btrfs_start_transaction(root, 0);
1364 1365
		if (IS_ERR(trans)) {
			ret = PTR_ERR(trans);
1366
			goto out_free;
1367
		}
C
Christoph Hellwig 已提交
1368 1369
		ret = btrfs_grow_device(trans, device, new_size);
		btrfs_commit_transaction(trans, root);
1370
	} else if (new_size < old_size) {
C
Christoph Hellwig 已提交
1371 1372 1373
		ret = btrfs_shrink_device(device, new_size);
	}

1374
out_free:
C
Christoph Hellwig 已提交
1375
	kfree(vol_args);
1376 1377
out:
	mutex_unlock(&root->fs_info->volume_mutex);
C
Christoph Hellwig 已提交
1378 1379 1380
	return ret;
}

S
Sage Weil 已提交
1381 1382 1383 1384
static noinline int btrfs_ioctl_snap_create_transid(struct file *file,
						    char *name,
						    unsigned long fd,
						    int subvol,
L
Li Zefan 已提交
1385 1386
						    u64 *transid,
						    bool readonly)
C
Christoph Hellwig 已提交
1387
{
1388
	struct btrfs_root *root = BTRFS_I(fdentry(file)->d_inode)->root;
1389
	struct file *src_file;
C
Christoph Hellwig 已提交
1390
	int namelen;
1391
	int ret = 0;
C
Christoph Hellwig 已提交
1392

Y
Yan Zheng 已提交
1393 1394 1395
	if (root->fs_info->sb->s_flags & MS_RDONLY)
		return -EROFS;

S
Sage Weil 已提交
1396 1397
	namelen = strlen(name);
	if (strchr(name, '/')) {
C
Christoph Hellwig 已提交
1398 1399 1400 1401
		ret = -EINVAL;
		goto out;
	}

1402 1403 1404 1405 1406 1407
	if (name[0] == '.' &&
	   (namelen == 1 || (name[1] == '.' && namelen == 2))) {
		ret = -EEXIST;
		goto out;
	}

1408
	if (subvol) {
S
Sage Weil 已提交
1409
		ret = btrfs_mksubvol(&file->f_path, name, namelen,
L
Li Zefan 已提交
1410
				     NULL, transid, readonly);
1411
	} else {
1412
		struct inode *src_inode;
S
Sage Weil 已提交
1413
		src_file = fget(fd);
1414 1415 1416 1417 1418 1419 1420
		if (!src_file) {
			ret = -EINVAL;
			goto out;
		}

		src_inode = src_file->f_path.dentry->d_inode;
		if (src_inode->i_sb != file->f_path.dentry->d_inode->i_sb) {
C
Chris Mason 已提交
1421 1422
			printk(KERN_INFO "btrfs: Snapshot src from "
			       "another FS\n");
1423 1424 1425 1426
			ret = -EINVAL;
			fput(src_file);
			goto out;
		}
S
Sage Weil 已提交
1427 1428
		ret = btrfs_mksubvol(&file->f_path, name, namelen,
				     BTRFS_I(src_inode)->root,
L
Li Zefan 已提交
1429
				     transid, readonly);
1430
		fput(src_file);
1431
	}
C
Christoph Hellwig 已提交
1432
out:
S
Sage Weil 已提交
1433 1434 1435 1436
	return ret;
}

static noinline int btrfs_ioctl_snap_create(struct file *file,
1437
					    void __user *arg, int subvol)
S
Sage Weil 已提交
1438
{
1439
	struct btrfs_ioctl_vol_args *vol_args;
S
Sage Weil 已提交
1440 1441
	int ret;

1442 1443 1444 1445
	vol_args = memdup_user(arg, sizeof(*vol_args));
	if (IS_ERR(vol_args))
		return PTR_ERR(vol_args);
	vol_args->name[BTRFS_PATH_NAME_MAX] = '\0';
S
Sage Weil 已提交
1446

1447
	ret = btrfs_ioctl_snap_create_transid(file, vol_args->name,
L
Li Zefan 已提交
1448 1449
					      vol_args->fd, subvol,
					      NULL, false);
1450

1451 1452 1453
	kfree(vol_args);
	return ret;
}
1454

1455 1456 1457 1458 1459 1460 1461
static noinline int btrfs_ioctl_snap_create_v2(struct file *file,
					       void __user *arg, int subvol)
{
	struct btrfs_ioctl_vol_args_v2 *vol_args;
	int ret;
	u64 transid = 0;
	u64 *ptr = NULL;
L
Li Zefan 已提交
1462
	bool readonly = false;
1463

1464 1465 1466 1467
	vol_args = memdup_user(arg, sizeof(*vol_args));
	if (IS_ERR(vol_args))
		return PTR_ERR(vol_args);
	vol_args->name[BTRFS_SUBVOL_NAME_MAX] = '\0';
1468

L
Li Zefan 已提交
1469 1470 1471
	if (vol_args->flags &
	    ~(BTRFS_SUBVOL_CREATE_ASYNC | BTRFS_SUBVOL_RDONLY)) {
		ret = -EOPNOTSUPP;
1472
		goto out;
S
Sage Weil 已提交
1473
	}
1474 1475 1476

	if (vol_args->flags & BTRFS_SUBVOL_CREATE_ASYNC)
		ptr = &transid;
L
Li Zefan 已提交
1477 1478
	if (vol_args->flags & BTRFS_SUBVOL_RDONLY)
		readonly = true;
1479 1480

	ret = btrfs_ioctl_snap_create_transid(file, vol_args->name,
L
Li Zefan 已提交
1481 1482
					      vol_args->fd, subvol,
					      ptr, readonly);
1483 1484 1485 1486 1487 1488

	if (ret == 0 && ptr &&
	    copy_to_user(arg +
			 offsetof(struct btrfs_ioctl_vol_args_v2,
				  transid), ptr, sizeof(*ptr)))
		ret = -EFAULT;
1489
out:
C
Christoph Hellwig 已提交
1490 1491 1492 1493
	kfree(vol_args);
	return ret;
}

1494 1495 1496 1497 1498 1499 1500 1501
static noinline int btrfs_ioctl_subvol_getflags(struct file *file,
						void __user *arg)
{
	struct inode *inode = fdentry(file)->d_inode;
	struct btrfs_root *root = BTRFS_I(inode)->root;
	int ret = 0;
	u64 flags = 0;

L
Li Zefan 已提交
1502
	if (btrfs_ino(inode) != BTRFS_FIRST_FREE_OBJECTID)
1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528
		return -EINVAL;

	down_read(&root->fs_info->subvol_sem);
	if (btrfs_root_readonly(root))
		flags |= BTRFS_SUBVOL_RDONLY;
	up_read(&root->fs_info->subvol_sem);

	if (copy_to_user(arg, &flags, sizeof(flags)))
		ret = -EFAULT;

	return ret;
}

static noinline int btrfs_ioctl_subvol_setflags(struct file *file,
					      void __user *arg)
{
	struct inode *inode = fdentry(file)->d_inode;
	struct btrfs_root *root = BTRFS_I(inode)->root;
	struct btrfs_trans_handle *trans;
	u64 root_flags;
	u64 flags;
	int ret = 0;

	if (root->fs_info->sb->s_flags & MS_RDONLY)
		return -EROFS;

L
Li Zefan 已提交
1529
	if (btrfs_ino(inode) != BTRFS_FIRST_FREE_OBJECTID)
1530 1531 1532 1533 1534
		return -EINVAL;

	if (copy_from_user(&flags, arg, sizeof(flags)))
		return -EFAULT;

1535
	if (flags & BTRFS_SUBVOL_CREATE_ASYNC)
1536 1537 1538 1539 1540
		return -EINVAL;

	if (flags & ~BTRFS_SUBVOL_RDONLY)
		return -EOPNOTSUPP;

1541
	if (!inode_owner_or_capable(inode))
1542 1543
		return -EACCES;

1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563
	down_write(&root->fs_info->subvol_sem);

	/* nothing to do */
	if (!!(flags & BTRFS_SUBVOL_RDONLY) == btrfs_root_readonly(root))
		goto out;

	root_flags = btrfs_root_flags(&root->root_item);
	if (flags & BTRFS_SUBVOL_RDONLY)
		btrfs_set_root_flags(&root->root_item,
				     root_flags | BTRFS_ROOT_SUBVOL_RDONLY);
	else
		btrfs_set_root_flags(&root->root_item,
				     root_flags & ~BTRFS_ROOT_SUBVOL_RDONLY);

	trans = btrfs_start_transaction(root, 1);
	if (IS_ERR(trans)) {
		ret = PTR_ERR(trans);
		goto out_reset;
	}

1564
	ret = btrfs_update_root(trans, root->fs_info->tree_root,
1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575
				&root->root_key, &root->root_item);

	btrfs_commit_transaction(trans, root);
out_reset:
	if (ret)
		btrfs_set_root_flags(&root->root_item, root_flags);
out:
	up_write(&root->fs_info->subvol_sem);
	return ret;
}

1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611
/*
 * helper to check if the subvolume references other subvolumes
 */
static noinline int may_destroy_subvol(struct btrfs_root *root)
{
	struct btrfs_path *path;
	struct btrfs_key key;
	int ret;

	path = btrfs_alloc_path();
	if (!path)
		return -ENOMEM;

	key.objectid = root->root_key.objectid;
	key.type = BTRFS_ROOT_REF_KEY;
	key.offset = (u64)-1;

	ret = btrfs_search_slot(NULL, root->fs_info->tree_root,
				&key, path, 0, 0);
	if (ret < 0)
		goto out;
	BUG_ON(ret == 0);

	ret = 0;
	if (path->slots[0] > 0) {
		path->slots[0]--;
		btrfs_item_key_to_cpu(path->nodes[0], &key, path->slots[0]);
		if (key.objectid == root->root_key.objectid &&
		    key.type == BTRFS_ROOT_REF_KEY)
			ret = -ENOTEMPTY;
	}
out:
	btrfs_free_path(path);
	return ret;
}

1612 1613 1614
static noinline int key_in_sk(struct btrfs_key *key,
			      struct btrfs_ioctl_search_key *sk)
{
1615 1616 1617 1618 1619 1620 1621 1622 1623
	struct btrfs_key test;
	int ret;

	test.objectid = sk->min_objectid;
	test.type = sk->min_type;
	test.offset = sk->min_offset;

	ret = btrfs_comp_cpu_keys(key, &test);
	if (ret < 0)
1624
		return 0;
1625 1626 1627 1628 1629 1630 1631

	test.objectid = sk->max_objectid;
	test.type = sk->max_type;
	test.offset = sk->max_offset;

	ret = btrfs_comp_cpu_keys(key, &test);
	if (ret > 0)
1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697
		return 0;
	return 1;
}

static noinline int copy_to_sk(struct btrfs_root *root,
			       struct btrfs_path *path,
			       struct btrfs_key *key,
			       struct btrfs_ioctl_search_key *sk,
			       char *buf,
			       unsigned long *sk_offset,
			       int *num_found)
{
	u64 found_transid;
	struct extent_buffer *leaf;
	struct btrfs_ioctl_search_header sh;
	unsigned long item_off;
	unsigned long item_len;
	int nritems;
	int i;
	int slot;
	int ret = 0;

	leaf = path->nodes[0];
	slot = path->slots[0];
	nritems = btrfs_header_nritems(leaf);

	if (btrfs_header_generation(leaf) > sk->max_transid) {
		i = nritems;
		goto advance_key;
	}
	found_transid = btrfs_header_generation(leaf);

	for (i = slot; i < nritems; i++) {
		item_off = btrfs_item_ptr_offset(leaf, i);
		item_len = btrfs_item_size_nr(leaf, i);

		if (item_len > BTRFS_SEARCH_ARGS_BUFSIZE)
			item_len = 0;

		if (sizeof(sh) + item_len + *sk_offset >
		    BTRFS_SEARCH_ARGS_BUFSIZE) {
			ret = 1;
			goto overflow;
		}

		btrfs_item_key_to_cpu(leaf, key, i);
		if (!key_in_sk(key, sk))
			continue;

		sh.objectid = key->objectid;
		sh.offset = key->offset;
		sh.type = key->type;
		sh.len = item_len;
		sh.transid = found_transid;

		/* copy search result header */
		memcpy(buf + *sk_offset, &sh, sizeof(sh));
		*sk_offset += sizeof(sh);

		if (item_len) {
			char *p = buf + *sk_offset;
			/* copy the item */
			read_extent_buffer(leaf, p,
					   item_off, item_len);
			*sk_offset += item_len;
		}
1698
		(*num_found)++;
1699 1700 1701 1702 1703

		if (*num_found >= sk->nr_items)
			break;
	}
advance_key:
1704 1705
	ret = 0;
	if (key->offset < (u64)-1 && key->offset < sk->max_offset)
1706
		key->offset++;
1707 1708
	else if (key->type < (u8)-1 && key->type < sk->max_type) {
		key->offset = 0;
1709
		key->type++;
1710 1711 1712
	} else if (key->objectid < (u64)-1 && key->objectid < sk->max_objectid) {
		key->offset = 0;
		key->type = 0;
1713
		key->objectid++;
1714 1715
	} else
		ret = 1;
1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772
overflow:
	return ret;
}

static noinline int search_ioctl(struct inode *inode,
				 struct btrfs_ioctl_search_args *args)
{
	struct btrfs_root *root;
	struct btrfs_key key;
	struct btrfs_key max_key;
	struct btrfs_path *path;
	struct btrfs_ioctl_search_key *sk = &args->key;
	struct btrfs_fs_info *info = BTRFS_I(inode)->root->fs_info;
	int ret;
	int num_found = 0;
	unsigned long sk_offset = 0;

	path = btrfs_alloc_path();
	if (!path)
		return -ENOMEM;

	if (sk->tree_id == 0) {
		/* search the root of the inode that was passed */
		root = BTRFS_I(inode)->root;
	} else {
		key.objectid = sk->tree_id;
		key.type = BTRFS_ROOT_ITEM_KEY;
		key.offset = (u64)-1;
		root = btrfs_read_fs_root_no_name(info, &key);
		if (IS_ERR(root)) {
			printk(KERN_ERR "could not find root %llu\n",
			       sk->tree_id);
			btrfs_free_path(path);
			return -ENOENT;
		}
	}

	key.objectid = sk->min_objectid;
	key.type = sk->min_type;
	key.offset = sk->min_offset;

	max_key.objectid = sk->max_objectid;
	max_key.type = sk->max_type;
	max_key.offset = sk->max_offset;

	path->keep_locks = 1;

	while(1) {
		ret = btrfs_search_forward(root, &key, &max_key, path, 0,
					   sk->min_transid);
		if (ret != 0) {
			if (ret > 0)
				ret = 0;
			goto err;
		}
		ret = copy_to_sk(root, path, &key, sk, args->buf,
				 &sk_offset, &num_found);
1773
		btrfs_release_path(path);
1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794
		if (ret || num_found >= sk->nr_items)
			break;

	}
	ret = 0;
err:
	sk->nr_items = num_found;
	btrfs_free_path(path);
	return ret;
}

static noinline int btrfs_ioctl_tree_search(struct file *file,
					   void __user *argp)
{
	 struct btrfs_ioctl_search_args *args;
	 struct inode *inode;
	 int ret;

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

J
Julia Lawall 已提交
1795 1796 1797
	args = memdup_user(argp, sizeof(*args));
	if (IS_ERR(args))
		return PTR_ERR(args);
1798 1799 1800 1801 1802 1803 1804 1805 1806

	inode = fdentry(file)->d_inode;
	ret = search_ioctl(inode, args);
	if (ret == 0 && copy_to_user(argp, args, sizeof(*args)))
		ret = -EFAULT;
	kfree(args);
	return ret;
}

1807
/*
1808 1809 1810
 * Search INODE_REFs to identify path name of 'dirid' directory
 * in a 'tree_id' tree. and sets path name to 'name'.
 */
1811 1812 1813 1814 1815
static noinline int btrfs_search_path_in_tree(struct btrfs_fs_info *info,
				u64 tree_id, u64 dirid, char *name)
{
	struct btrfs_root *root;
	struct btrfs_key key;
1816
	char *ptr;
1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833
	int ret = -1;
	int slot;
	int len;
	int total_len = 0;
	struct btrfs_inode_ref *iref;
	struct extent_buffer *l;
	struct btrfs_path *path;

	if (dirid == BTRFS_FIRST_FREE_OBJECTID) {
		name[0]='\0';
		return 0;
	}

	path = btrfs_alloc_path();
	if (!path)
		return -ENOMEM;

1834
	ptr = &name[BTRFS_INO_LOOKUP_PATH_MAX];
1835 1836 1837 1838 1839 1840 1841

	key.objectid = tree_id;
	key.type = BTRFS_ROOT_ITEM_KEY;
	key.offset = (u64)-1;
	root = btrfs_read_fs_root_no_name(info, &key);
	if (IS_ERR(root)) {
		printk(KERN_ERR "could not find root %llu\n", tree_id);
1842 1843
		ret = -ENOENT;
		goto out;
1844 1845 1846 1847
	}

	key.objectid = dirid;
	key.type = BTRFS_INODE_REF_KEY;
1848
	key.offset = (u64)-1;
1849 1850 1851 1852 1853 1854 1855 1856

	while(1) {
		ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
		if (ret < 0)
			goto out;

		l = path->nodes[0];
		slot = path->slots[0];
1857 1858
		if (ret > 0 && slot > 0)
			slot--;
1859 1860 1861
		btrfs_item_key_to_cpu(l, &key, slot);

		if (ret > 0 && (key.objectid != dirid ||
1862 1863
				key.type != BTRFS_INODE_REF_KEY)) {
			ret = -ENOENT;
1864
			goto out;
1865
		}
1866 1867 1868 1869 1870

		iref = btrfs_item_ptr(l, slot, struct btrfs_inode_ref);
		len = btrfs_inode_ref_name_len(l, iref);
		ptr -= len + 1;
		total_len += len + 1;
1871
		if (ptr < name)
1872 1873 1874 1875 1876 1877 1878 1879
			goto out;

		*(ptr + len) = '/';
		read_extent_buffer(l, ptr,(unsigned long)(iref + 1), len);

		if (key.offset == BTRFS_FIRST_FREE_OBJECTID)
			break;

1880
		btrfs_release_path(path);
1881
		key.objectid = key.offset;
1882
		key.offset = (u64)-1;
1883 1884
		dirid = key.objectid;
	}
1885
	if (ptr < name)
1886
		goto out;
1887
	memmove(name, ptr, total_len);
1888 1889 1890 1891
	name[total_len]='\0';
	ret = 0;
out:
	btrfs_free_path(path);
1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904
	return ret;
}

static noinline int btrfs_ioctl_ino_lookup(struct file *file,
					   void __user *argp)
{
	 struct btrfs_ioctl_ino_lookup_args *args;
	 struct inode *inode;
	 int ret;

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

J
Julia Lawall 已提交
1905 1906 1907
	args = memdup_user(argp, sizeof(*args));
	if (IS_ERR(args))
		return PTR_ERR(args);
1908

1909 1910
	inode = fdentry(file)->d_inode;

1911 1912 1913
	if (args->treeid == 0)
		args->treeid = BTRFS_I(inode)->root->root_key.objectid;

1914 1915 1916 1917 1918 1919 1920 1921
	ret = btrfs_search_path_in_tree(BTRFS_I(inode)->root->fs_info,
					args->treeid, args->objectid,
					args->name);

	if (ret == 0 && copy_to_user(argp, args, sizeof(*args)))
		ret = -EFAULT;

	kfree(args);
1922 1923 1924
	return ret;
}

1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951
static noinline int btrfs_ioctl_snap_destroy(struct file *file,
					     void __user *arg)
{
	struct dentry *parent = fdentry(file);
	struct dentry *dentry;
	struct inode *dir = parent->d_inode;
	struct inode *inode;
	struct btrfs_root *root = BTRFS_I(dir)->root;
	struct btrfs_root *dest = NULL;
	struct btrfs_ioctl_vol_args *vol_args;
	struct btrfs_trans_handle *trans;
	int namelen;
	int ret;
	int err = 0;

	vol_args = memdup_user(arg, sizeof(*vol_args));
	if (IS_ERR(vol_args))
		return PTR_ERR(vol_args);

	vol_args->name[BTRFS_PATH_NAME_MAX] = '\0';
	namelen = strlen(vol_args->name);
	if (strchr(vol_args->name, '/') ||
	    strncmp(vol_args->name, "..", namelen) == 0) {
		err = -EINVAL;
		goto out;
	}

1952
	err = mnt_want_write_file(file);
1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968
	if (err)
		goto out;

	mutex_lock_nested(&dir->i_mutex, I_MUTEX_PARENT);
	dentry = lookup_one_len(vol_args->name, parent, namelen);
	if (IS_ERR(dentry)) {
		err = PTR_ERR(dentry);
		goto out_unlock_dir;
	}

	if (!dentry->d_inode) {
		err = -ENOENT;
		goto out_dput;
	}

	inode = dentry->d_inode;
1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008
	dest = BTRFS_I(inode)->root;
	if (!capable(CAP_SYS_ADMIN)){
		/*
		 * Regular user.  Only allow this with a special mount
		 * option, when the user has write+exec access to the
		 * subvol root, and when rmdir(2) would have been
		 * allowed.
		 *
		 * Note that this is _not_ check that the subvol is
		 * empty or doesn't contain data that we wouldn't
		 * otherwise be able to delete.
		 *
		 * Users who want to delete empty subvols should try
		 * rmdir(2).
		 */
		err = -EPERM;
		if (!btrfs_test_opt(root, USER_SUBVOL_RM_ALLOWED))
			goto out_dput;

		/*
		 * Do not allow deletion if the parent dir is the same
		 * as the dir to be deleted.  That means the ioctl
		 * must be called on the dentry referencing the root
		 * of the subvol, not a random directory contained
		 * within it.
		 */
		err = -EINVAL;
		if (root == dest)
			goto out_dput;

		err = inode_permission(inode, MAY_WRITE | MAY_EXEC);
		if (err)
			goto out_dput;

		/* check if subvolume may be deleted by a non-root user */
		err = btrfs_may_delete(dir, dentry, 1);
		if (err)
			goto out_dput;
	}

L
Li Zefan 已提交
2009
	if (btrfs_ino(inode) != BTRFS_FIRST_FREE_OBJECTID) {
2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024
		err = -EINVAL;
		goto out_dput;
	}

	mutex_lock(&inode->i_mutex);
	err = d_invalidate(dentry);
	if (err)
		goto out_unlock;

	down_write(&root->fs_info->subvol_sem);

	err = may_destroy_subvol(dest);
	if (err)
		goto out_up_write;

2025 2026 2027
	trans = btrfs_start_transaction(root, 0);
	if (IS_ERR(trans)) {
		err = PTR_ERR(trans);
2028
		goto out_up_write;
2029 2030 2031
	}
	trans->block_rsv = &root->fs_info->global_block_rsv;

2032 2033 2034 2035
	ret = btrfs_unlink_subvol(trans, root, dir,
				dest->root_key.objectid,
				dentry->d_name.name,
				dentry->d_name.len);
2036 2037 2038 2039 2040
	if (ret) {
		err = ret;
		btrfs_abort_transaction(trans, root, ret);
		goto out_end_trans;
	}
2041 2042 2043 2044 2045 2046 2047 2048

	btrfs_record_root_in_trans(trans, dest);

	memset(&dest->root_item.drop_progress, 0,
		sizeof(dest->root_item.drop_progress));
	dest->root_item.drop_level = 0;
	btrfs_set_root_refs(&dest->root_item, 0);

2049 2050 2051 2052
	if (!xchg(&dest->orphan_item_inserted, 1)) {
		ret = btrfs_insert_orphan_item(trans,
					root->fs_info->tree_root,
					dest->root_key.objectid);
2053 2054 2055 2056 2057
		if (ret) {
			btrfs_abort_transaction(trans, root, ret);
			err = ret;
			goto out_end_trans;
		}
2058
	}
2059
out_end_trans:
S
Sage Weil 已提交
2060
	ret = btrfs_end_transaction(trans, root);
2061 2062
	if (ret && !err)
		err = ret;
2063 2064 2065 2066 2067 2068
	inode->i_flags |= S_DEAD;
out_up_write:
	up_write(&root->fs_info->subvol_sem);
out_unlock:
	mutex_unlock(&inode->i_mutex);
	if (!err) {
2069
		shrink_dcache_sb(root->fs_info->sb);
2070 2071 2072 2073 2074 2075 2076
		btrfs_invalidate_inodes(dest);
		d_delete(dentry);
	}
out_dput:
	dput(dentry);
out_unlock_dir:
	mutex_unlock(&dir->i_mutex);
A
Al Viro 已提交
2077
	mnt_drop_write_file(file);
2078 2079 2080 2081 2082
out:
	kfree(vol_args);
	return err;
}

C
Chris Mason 已提交
2083
static int btrfs_ioctl_defrag(struct file *file, void __user *argp)
C
Christoph Hellwig 已提交
2084 2085 2086
{
	struct inode *inode = fdentry(file)->d_inode;
	struct btrfs_root *root = BTRFS_I(inode)->root;
C
Chris Mason 已提交
2087
	struct btrfs_ioctl_defrag_range_args *range;
Y
Yan Zheng 已提交
2088 2089
	int ret;

L
Li Zefan 已提交
2090 2091 2092
	if (btrfs_root_readonly(root))
		return -EROFS;

2093
	ret = mnt_want_write_file(file);
Y
Yan Zheng 已提交
2094 2095
	if (ret)
		return ret;
C
Christoph Hellwig 已提交
2096 2097 2098

	switch (inode->i_mode & S_IFMT) {
	case S_IFDIR:
2099 2100 2101 2102
		if (!capable(CAP_SYS_ADMIN)) {
			ret = -EPERM;
			goto out;
		}
2103 2104 2105 2106
		ret = btrfs_defrag_root(root, 0);
		if (ret)
			goto out;
		ret = btrfs_defrag_root(root->fs_info->extent_root, 0);
C
Christoph Hellwig 已提交
2107 2108
		break;
	case S_IFREG:
2109 2110 2111 2112
		if (!(file->f_mode & FMODE_WRITE)) {
			ret = -EINVAL;
			goto out;
		}
C
Chris Mason 已提交
2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124

		range = kzalloc(sizeof(*range), GFP_KERNEL);
		if (!range) {
			ret = -ENOMEM;
			goto out;
		}

		if (argp) {
			if (copy_from_user(range, argp,
					   sizeof(*range))) {
				ret = -EFAULT;
				kfree(range);
2125
				goto out;
C
Chris Mason 已提交
2126 2127 2128 2129 2130 2131 2132 2133 2134 2135
			}
			/* compression requires us to start the IO */
			if ((range->flags & BTRFS_DEFRAG_RANGE_COMPRESS)) {
				range->flags |= BTRFS_DEFRAG_RANGE_START_IO;
				range->extent_thresh = (u32)-1;
			}
		} else {
			/* the rest are all set to zero by kzalloc */
			range->len = (u64)-1;
		}
C
Chris Mason 已提交
2136 2137 2138 2139
		ret = btrfs_defrag_file(fdentry(file)->d_inode, file,
					range, 0, 0);
		if (ret > 0)
			ret = 0;
C
Chris Mason 已提交
2140
		kfree(range);
C
Christoph Hellwig 已提交
2141
		break;
2142 2143
	default:
		ret = -EINVAL;
C
Christoph Hellwig 已提交
2144
	}
2145
out:
A
Al Viro 已提交
2146
	mnt_drop_write_file(file);
2147
	return ret;
C
Christoph Hellwig 已提交
2148 2149
}

2150
static long btrfs_ioctl_add_dev(struct btrfs_root *root, void __user *arg)
C
Christoph Hellwig 已提交
2151 2152 2153 2154
{
	struct btrfs_ioctl_vol_args *vol_args;
	int ret;

2155 2156 2157
	if (!capable(CAP_SYS_ADMIN))
		return -EPERM;

2158 2159 2160 2161 2162 2163 2164
	mutex_lock(&root->fs_info->volume_mutex);
	if (root->fs_info->balance_ctl) {
		printk(KERN_INFO "btrfs: balance in progress\n");
		ret = -EINVAL;
		goto out;
	}

L
Li Zefan 已提交
2165
	vol_args = memdup_user(arg, sizeof(*vol_args));
2166 2167 2168 2169
	if (IS_ERR(vol_args)) {
		ret = PTR_ERR(vol_args);
		goto out;
	}
C
Christoph Hellwig 已提交
2170

2171
	vol_args->name[BTRFS_PATH_NAME_MAX] = '\0';
C
Christoph Hellwig 已提交
2172 2173 2174
	ret = btrfs_init_new_device(root, vol_args->name);

	kfree(vol_args);
2175 2176
out:
	mutex_unlock(&root->fs_info->volume_mutex);
C
Christoph Hellwig 已提交
2177 2178 2179
	return ret;
}

2180
static long btrfs_ioctl_rm_dev(struct btrfs_root *root, void __user *arg)
C
Christoph Hellwig 已提交
2181 2182 2183 2184
{
	struct btrfs_ioctl_vol_args *vol_args;
	int ret;

2185 2186 2187
	if (!capable(CAP_SYS_ADMIN))
		return -EPERM;

Y
Yan Zheng 已提交
2188 2189 2190
	if (root->fs_info->sb->s_flags & MS_RDONLY)
		return -EROFS;

2191 2192 2193 2194 2195 2196 2197
	mutex_lock(&root->fs_info->volume_mutex);
	if (root->fs_info->balance_ctl) {
		printk(KERN_INFO "btrfs: balance in progress\n");
		ret = -EINVAL;
		goto out;
	}

L
Li Zefan 已提交
2198
	vol_args = memdup_user(arg, sizeof(*vol_args));
2199 2200 2201 2202
	if (IS_ERR(vol_args)) {
		ret = PTR_ERR(vol_args);
		goto out;
	}
C
Christoph Hellwig 已提交
2203

2204
	vol_args->name[BTRFS_PATH_NAME_MAX] = '\0';
C
Christoph Hellwig 已提交
2205 2206 2207
	ret = btrfs_rm_device(root, vol_args->name);

	kfree(vol_args);
2208 2209
out:
	mutex_unlock(&root->fs_info->volume_mutex);
C
Christoph Hellwig 已提交
2210 2211 2212
	return ret;
}

J
Jan Schmidt 已提交
2213 2214
static long btrfs_ioctl_fs_info(struct btrfs_root *root, void __user *arg)
{
2215
	struct btrfs_ioctl_fs_info_args *fi_args;
J
Jan Schmidt 已提交
2216 2217 2218
	struct btrfs_device *device;
	struct btrfs_device *next;
	struct btrfs_fs_devices *fs_devices = root->fs_info->fs_devices;
2219
	int ret = 0;
J
Jan Schmidt 已提交
2220 2221 2222 2223

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

2224 2225 2226 2227 2228 2229
	fi_args = kzalloc(sizeof(*fi_args), GFP_KERNEL);
	if (!fi_args)
		return -ENOMEM;

	fi_args->num_devices = fs_devices->num_devices;
	memcpy(&fi_args->fsid, root->fs_info->fsid, sizeof(fi_args->fsid));
J
Jan Schmidt 已提交
2230 2231 2232

	mutex_lock(&fs_devices->device_list_mutex);
	list_for_each_entry_safe(device, next, &fs_devices->devices, dev_list) {
2233 2234
		if (device->devid > fi_args->max_id)
			fi_args->max_id = device->devid;
J
Jan Schmidt 已提交
2235 2236 2237
	}
	mutex_unlock(&fs_devices->device_list_mutex);

2238 2239
	if (copy_to_user(arg, fi_args, sizeof(*fi_args)))
		ret = -EFAULT;
J
Jan Schmidt 已提交
2240

2241 2242
	kfree(fi_args);
	return ret;
J
Jan Schmidt 已提交
2243 2244 2245 2246 2247 2248 2249 2250 2251 2252 2253 2254 2255 2256 2257 2258 2259 2260 2261 2262 2263 2264 2265 2266 2267 2268 2269 2270 2271 2272 2273 2274 2275 2276
}

static long btrfs_ioctl_dev_info(struct btrfs_root *root, void __user *arg)
{
	struct btrfs_ioctl_dev_info_args *di_args;
	struct btrfs_device *dev;
	struct btrfs_fs_devices *fs_devices = root->fs_info->fs_devices;
	int ret = 0;
	char *s_uuid = NULL;
	char empty_uuid[BTRFS_UUID_SIZE] = {0};

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

	di_args = memdup_user(arg, sizeof(*di_args));
	if (IS_ERR(di_args))
		return PTR_ERR(di_args);

	if (memcmp(empty_uuid, di_args->uuid, BTRFS_UUID_SIZE) != 0)
		s_uuid = di_args->uuid;

	mutex_lock(&fs_devices->device_list_mutex);
	dev = btrfs_find_device(root, di_args->devid, s_uuid, NULL);
	mutex_unlock(&fs_devices->device_list_mutex);

	if (!dev) {
		ret = -ENODEV;
		goto out;
	}

	di_args->devid = dev->devid;
	di_args->bytes_used = dev->bytes_used;
	di_args->total_bytes = dev->total_bytes;
	memcpy(di_args->uuid, dev->uuid, sizeof(di_args->uuid));
2277
	if (dev->name) {
2278 2279 2280 2281 2282 2283
		struct rcu_string *name;

		rcu_read_lock();
		name = rcu_dereference(dev->name);
		strncpy(di_args->path, name->str, sizeof(di_args->path));
		rcu_read_unlock();
2284 2285
		di_args->path[sizeof(di_args->path) - 1] = 0;
	} else {
2286
		di_args->path[0] = '\0';
2287
	}
J
Jan Schmidt 已提交
2288 2289 2290 2291 2292 2293 2294 2295 2296

out:
	if (ret == 0 && copy_to_user(arg, di_args, sizeof(*di_args)))
		ret = -EFAULT;

	kfree(di_args);
	return ret;
}

2297 2298
static noinline long btrfs_ioctl_clone(struct file *file, unsigned long srcfd,
				       u64 off, u64 olen, u64 destoff)
C
Christoph Hellwig 已提交
2299 2300 2301 2302 2303 2304 2305 2306
{
	struct inode *inode = fdentry(file)->d_inode;
	struct btrfs_root *root = BTRFS_I(inode)->root;
	struct file *src_file;
	struct inode *src;
	struct btrfs_trans_handle *trans;
	struct btrfs_path *path;
	struct extent_buffer *leaf;
Y
Yan Zheng 已提交
2307 2308
	char *buf;
	struct btrfs_key key;
C
Christoph Hellwig 已提交
2309 2310
	u32 nritems;
	int slot;
Y
Yan Zheng 已提交
2311
	int ret;
2312 2313 2314
	u64 len = olen;
	u64 bs = root->fs_info->sb->s_blocksize;
	u64 hint_byte;
2315

2316 2317 2318 2319 2320 2321 2322 2323 2324 2325
	/*
	 * TODO:
	 * - split compressed inline extents.  annoying: we need to
	 *   decompress into destination's address_space (the file offset
	 *   may change, so source mapping won't do), then recompress (or
	 *   otherwise reinsert) a subrange.
	 * - allow ranges within the same file to be cloned (provided
	 *   they don't overlap)?
	 */

2326
	/* the destination must be opened for writing */
2327
	if (!(file->f_mode & FMODE_WRITE) || (file->f_flags & O_APPEND))
2328 2329
		return -EINVAL;

L
Li Zefan 已提交
2330 2331 2332
	if (btrfs_root_readonly(root))
		return -EROFS;

2333
	ret = mnt_want_write_file(file);
Y
Yan Zheng 已提交
2334 2335 2336
	if (ret)
		return ret;

2337
	src_file = fget(srcfd);
2338 2339 2340 2341
	if (!src_file) {
		ret = -EBADF;
		goto out_drop_write;
	}
2342

C
Christoph Hellwig 已提交
2343 2344
	src = src_file->f_dentry->d_inode;

2345 2346 2347 2348
	ret = -EINVAL;
	if (src == inode)
		goto out_fput;

2349 2350 2351 2352
	/* the src must be open for reading */
	if (!(src_file->f_mode & FMODE_READ))
		goto out_fput;

2353 2354 2355 2356 2357
	/* don't make the dst file partly checksummed */
	if ((BTRFS_I(src)->flags & BTRFS_INODE_NODATASUM) !=
	    (BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM))
		goto out_fput;

Y
Yan Zheng 已提交
2358 2359 2360 2361
	ret = -EISDIR;
	if (S_ISDIR(src->i_mode) || S_ISDIR(inode->i_mode))
		goto out_fput;

C
Christoph Hellwig 已提交
2362
	ret = -EXDEV;
Y
Yan Zheng 已提交
2363 2364 2365 2366 2367 2368 2369 2370 2371 2372 2373
	if (src->i_sb != inode->i_sb || BTRFS_I(src)->root != root)
		goto out_fput;

	ret = -ENOMEM;
	buf = vmalloc(btrfs_level_size(root, 0));
	if (!buf)
		goto out_fput;

	path = btrfs_alloc_path();
	if (!path) {
		vfree(buf);
C
Christoph Hellwig 已提交
2374
		goto out_fput;
Y
Yan Zheng 已提交
2375 2376
	}
	path->reada = 2;
C
Christoph Hellwig 已提交
2377 2378

	if (inode < src) {
2379 2380
		mutex_lock_nested(&inode->i_mutex, I_MUTEX_PARENT);
		mutex_lock_nested(&src->i_mutex, I_MUTEX_CHILD);
C
Christoph Hellwig 已提交
2381
	} else {
2382 2383
		mutex_lock_nested(&src->i_mutex, I_MUTEX_PARENT);
		mutex_lock_nested(&inode->i_mutex, I_MUTEX_CHILD);
C
Christoph Hellwig 已提交
2384 2385
	}

2386 2387
	/* determine range to clone */
	ret = -EINVAL;
2388
	if (off + len > src->i_size || off + len < off)
C
Christoph Hellwig 已提交
2389
		goto out_unlock;
2390 2391 2392 2393
	if (len == 0)
		olen = len = src->i_size - off;
	/* if we extend to eof, continue to block boundary */
	if (off + len == src->i_size)
2394
		len = ALIGN(src->i_size, bs) - off;
2395 2396

	/* verify the end result is block aligned */
2397 2398
	if (!IS_ALIGNED(off, bs) || !IS_ALIGNED(off + len, bs) ||
	    !IS_ALIGNED(destoff, bs))
2399 2400
		goto out_unlock;

2401 2402 2403 2404 2405 2406
	if (destoff > inode->i_size) {
		ret = btrfs_cont_expand(inode, inode->i_size, destoff);
		if (ret)
			goto out_unlock;
	}

2407 2408 2409 2410
	/* truncate page cache pages from target inode range */
	truncate_inode_pages_range(&inode->i_data, destoff,
				   PAGE_CACHE_ALIGN(destoff + len) - 1);

C
Christoph Hellwig 已提交
2411 2412 2413
	/* do any pending delalloc/csum calc on src, one way or
	   another, and lock file content */
	while (1) {
Z
Zheng Yan 已提交
2414
		struct btrfs_ordered_extent *ordered;
2415
		lock_extent(&BTRFS_I(src)->io_tree, off, off+len);
2416 2417 2418 2419
		ordered = btrfs_lookup_first_ordered_extent(src, off+len);
		if (!ordered &&
		    !test_range_bit(&BTRFS_I(src)->io_tree, off, off+len,
				   EXTENT_DELALLOC, 0, NULL))
C
Christoph Hellwig 已提交
2420
			break;
2421
		unlock_extent(&BTRFS_I(src)->io_tree, off, off+len);
Y
Yan Zheng 已提交
2422 2423
		if (ordered)
			btrfs_put_ordered_extent(ordered);
2424
		btrfs_wait_ordered_range(src, off, len);
C
Christoph Hellwig 已提交
2425 2426
	}

2427
	/* clone data */
L
Li Zefan 已提交
2428
	key.objectid = btrfs_ino(src);
Y
Yan Zheng 已提交
2429 2430
	key.type = BTRFS_EXTENT_DATA_KEY;
	key.offset = 0;
C
Christoph Hellwig 已提交
2431 2432 2433 2434 2435 2436

	while (1) {
		/*
		 * note the key will change type as we walk through the
		 * tree.
		 */
2437
		ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
C
Christoph Hellwig 已提交
2438 2439 2440
		if (ret < 0)
			goto out;

Y
Yan Zheng 已提交
2441 2442
		nritems = btrfs_header_nritems(path->nodes[0]);
		if (path->slots[0] >= nritems) {
C
Christoph Hellwig 已提交
2443 2444 2445 2446 2447
			ret = btrfs_next_leaf(root, path);
			if (ret < 0)
				goto out;
			if (ret > 0)
				break;
Y
Yan Zheng 已提交
2448
			nritems = btrfs_header_nritems(path->nodes[0]);
C
Christoph Hellwig 已提交
2449 2450 2451 2452
		}
		leaf = path->nodes[0];
		slot = path->slots[0];

Y
Yan Zheng 已提交
2453
		btrfs_item_key_to_cpu(leaf, &key, slot);
2454
		if (btrfs_key_type(&key) > BTRFS_EXTENT_DATA_KEY ||
L
Li Zefan 已提交
2455
		    key.objectid != btrfs_ino(src))
C
Christoph Hellwig 已提交
2456 2457
			break;

2458 2459 2460
		if (btrfs_key_type(&key) == BTRFS_EXTENT_DATA_KEY) {
			struct btrfs_file_extent_item *extent;
			int type;
Z
Zheng Yan 已提交
2461 2462
			u32 size;
			struct btrfs_key new_key;
2463 2464 2465
			u64 disko = 0, diskl = 0;
			u64 datao = 0, datal = 0;
			u8 comp;
2466
			u64 endoff;
Z
Zheng Yan 已提交
2467 2468 2469 2470 2471

			size = btrfs_item_size_nr(leaf, slot);
			read_extent_buffer(leaf, buf,
					   btrfs_item_ptr_offset(leaf, slot),
					   size);
2472 2473 2474 2475 2476

			extent = btrfs_item_ptr(leaf, slot,
						struct btrfs_file_extent_item);
			comp = btrfs_file_extent_compression(leaf, extent);
			type = btrfs_file_extent_type(leaf, extent);
2477 2478
			if (type == BTRFS_FILE_EXTENT_REG ||
			    type == BTRFS_FILE_EXTENT_PREALLOC) {
C
Chris Mason 已提交
2479 2480 2481 2482
				disko = btrfs_file_extent_disk_bytenr(leaf,
								      extent);
				diskl = btrfs_file_extent_disk_num_bytes(leaf,
								 extent);
2483
				datao = btrfs_file_extent_offset(leaf, extent);
C
Chris Mason 已提交
2484 2485
				datal = btrfs_file_extent_num_bytes(leaf,
								    extent);
2486 2487 2488 2489 2490
			} else if (type == BTRFS_FILE_EXTENT_INLINE) {
				/* take upper bound, may be compressed */
				datal = btrfs_file_extent_ram_bytes(leaf,
								    extent);
			}
2491
			btrfs_release_path(path);
Z
Zheng Yan 已提交
2492

2493
			if (key.offset + datal <= off ||
2494 2495 2496
			    key.offset >= off+len)
				goto next;

Z
Zheng Yan 已提交
2497
			memcpy(&new_key, &key, sizeof(new_key));
L
Li Zefan 已提交
2498
			new_key.objectid = btrfs_ino(inode);
2499 2500 2501 2502
			if (off <= key.offset)
				new_key.offset = key.offset + destoff - off;
			else
				new_key.offset = destoff;
Z
Zheng Yan 已提交
2503

2504 2505 2506 2507 2508 2509
			/*
			 * 1 - adjusting old extent (we may have to split it)
			 * 1 - add new extent
			 * 1 - inode update
			 */
			trans = btrfs_start_transaction(root, 3);
2510 2511 2512 2513 2514
			if (IS_ERR(trans)) {
				ret = PTR_ERR(trans);
				goto out;
			}

2515 2516
			if (type == BTRFS_FILE_EXTENT_REG ||
			    type == BTRFS_FILE_EXTENT_PREALLOC) {
2517 2518 2519 2520 2521 2522 2523 2524 2525 2526
				/*
				 *    a  | --- range to clone ---|  b
				 * | ------------- extent ------------- |
				 */

				/* substract range b */
				if (key.offset + datal > off + len)
					datal = off + len - key.offset;

				/* substract range a */
2527 2528 2529 2530 2531 2532 2533 2534 2535
				if (off > key.offset) {
					datao += off - key.offset;
					datal -= off - key.offset;
				}

				ret = btrfs_drop_extents(trans, inode,
							 new_key.offset,
							 new_key.offset + datal,
							 &hint_byte, 1);
2536 2537 2538 2539 2540 2541
				if (ret) {
					btrfs_abort_transaction(trans, root,
								ret);
					btrfs_end_transaction(trans, root);
					goto out;
				}
2542

2543 2544
				ret = btrfs_insert_empty_item(trans, root, path,
							      &new_key, size);
2545 2546 2547 2548 2549 2550
				if (ret) {
					btrfs_abort_transaction(trans, root,
								ret);
					btrfs_end_transaction(trans, root);
					goto out;
				}
2551 2552 2553 2554

				leaf = path->nodes[0];
				slot = path->slots[0];
				write_extent_buffer(leaf, buf,
Z
Zheng Yan 已提交
2555 2556
					    btrfs_item_ptr_offset(leaf, slot),
					    size);
Y
Yan Zheng 已提交
2557

2558
				extent = btrfs_item_ptr(leaf, slot,
C
Christoph Hellwig 已提交
2559
						struct btrfs_file_extent_item);
2560 2561 2562 2563 2564 2565 2566 2567 2568 2569 2570

				/* disko == 0 means it's a hole */
				if (!disko)
					datao = 0;

				btrfs_set_file_extent_offset(leaf, extent,
							     datao);
				btrfs_set_file_extent_num_bytes(leaf, extent,
								datal);
				if (disko) {
					inode_add_bytes(inode, datal);
Y
Yan Zheng 已提交
2571
					ret = btrfs_inc_extent_ref(trans, root,
2572 2573
							disko, diskl, 0,
							root->root_key.objectid,
L
Li Zefan 已提交
2574
							btrfs_ino(inode),
A
Arne Jansen 已提交
2575 2576
							new_key.offset - datao,
							0);
2577 2578 2579 2580 2581 2582 2583 2584 2585
					if (ret) {
						btrfs_abort_transaction(trans,
									root,
									ret);
						btrfs_end_transaction(trans,
								      root);
						goto out;

					}
C
Christoph Hellwig 已提交
2586
				}
2587 2588 2589 2590 2591 2592 2593
			} else if (type == BTRFS_FILE_EXTENT_INLINE) {
				u64 skip = 0;
				u64 trim = 0;
				if (off > key.offset) {
					skip = off - key.offset;
					new_key.offset += skip;
				}
C
Chris Mason 已提交
2594

2595 2596
				if (key.offset + datal > off+len)
					trim = key.offset + datal - (off+len);
C
Chris Mason 已提交
2597

2598 2599
				if (comp && (skip || trim)) {
					ret = -EINVAL;
2600
					btrfs_end_transaction(trans, root);
2601 2602 2603 2604
					goto out;
				}
				size -= skip + trim;
				datal -= skip + trim;
2605 2606 2607 2608 2609

				ret = btrfs_drop_extents(trans, inode,
							 new_key.offset,
							 new_key.offset + datal,
							 &hint_byte, 1);
2610 2611 2612 2613 2614 2615
				if (ret) {
					btrfs_abort_transaction(trans, root,
								ret);
					btrfs_end_transaction(trans, root);
					goto out;
				}
2616

2617 2618
				ret = btrfs_insert_empty_item(trans, root, path,
							      &new_key, size);
2619 2620 2621 2622 2623 2624
				if (ret) {
					btrfs_abort_transaction(trans, root,
								ret);
					btrfs_end_transaction(trans, root);
					goto out;
				}
2625 2626

				if (skip) {
C
Chris Mason 已提交
2627 2628
					u32 start =
					  btrfs_file_extent_calc_inline_size(0);
2629 2630 2631 2632 2633 2634 2635 2636 2637 2638
					memmove(buf+start, buf+start+skip,
						datal);
				}

				leaf = path->nodes[0];
				slot = path->slots[0];
				write_extent_buffer(leaf, buf,
					    btrfs_item_ptr_offset(leaf, slot),
					    size);
				inode_add_bytes(inode, datal);
C
Christoph Hellwig 已提交
2639
			}
2640 2641

			btrfs_mark_buffer_dirty(leaf);
2642
			btrfs_release_path(path);
2643

2644
			inode_inc_iversion(inode);
2645
			inode->i_mtime = inode->i_ctime = CURRENT_TIME;
2646 2647 2648 2649 2650 2651 2652

			/*
			 * we round up to the block size at eof when
			 * determining which extents to clone above,
			 * but shouldn't round up the file size
			 */
			endoff = new_key.offset + datal;
2653 2654
			if (endoff > destoff+olen)
				endoff = destoff+olen;
2655 2656 2657
			if (endoff > inode->i_size)
				btrfs_i_size_write(inode, endoff);

2658
			ret = btrfs_update_inode(trans, root, inode);
2659 2660 2661 2662 2663 2664
			if (ret) {
				btrfs_abort_transaction(trans, root, ret);
				btrfs_end_transaction(trans, root);
				goto out;
			}
			ret = btrfs_end_transaction(trans, root);
2665
		}
C
Chris Mason 已提交
2666
next:
2667
		btrfs_release_path(path);
C
Christoph Hellwig 已提交
2668 2669 2670 2671
		key.offset++;
	}
	ret = 0;
out:
2672
	btrfs_release_path(path);
2673
	unlock_extent(&BTRFS_I(src)->io_tree, off, off+len);
C
Christoph Hellwig 已提交
2674 2675 2676
out_unlock:
	mutex_unlock(&src->i_mutex);
	mutex_unlock(&inode->i_mutex);
Y
Yan Zheng 已提交
2677 2678
	vfree(buf);
	btrfs_free_path(path);
C
Christoph Hellwig 已提交
2679 2680
out_fput:
	fput(src_file);
2681
out_drop_write:
A
Al Viro 已提交
2682
	mnt_drop_write_file(file);
C
Christoph Hellwig 已提交
2683 2684 2685
	return ret;
}

2686
static long btrfs_ioctl_clone_range(struct file *file, void __user *argp)
2687 2688 2689
{
	struct btrfs_ioctl_clone_range_args args;

2690
	if (copy_from_user(&args, argp, sizeof(args)))
2691 2692 2693 2694 2695
		return -EFAULT;
	return btrfs_ioctl_clone(file, args.src_fd, args.src_offset,
				 args.src_length, args.dest_offset);
}

C
Christoph Hellwig 已提交
2696 2697 2698 2699 2700 2701
/*
 * there are many ways the trans_start and trans_end ioctls can lead
 * to deadlocks.  They should only be used by applications that
 * basically own the machine, and have a very in depth understanding
 * of all the possible deadlocks and enospc problems.
 */
2702
static long btrfs_ioctl_trans_start(struct file *file)
C
Christoph Hellwig 已提交
2703 2704 2705 2706
{
	struct inode *inode = fdentry(file)->d_inode;
	struct btrfs_root *root = BTRFS_I(inode)->root;
	struct btrfs_trans_handle *trans;
2707
	int ret;
C
Christoph Hellwig 已提交
2708

2709
	ret = -EPERM;
2710
	if (!capable(CAP_SYS_ADMIN))
2711
		goto out;
2712

2713 2714
	ret = -EINPROGRESS;
	if (file->private_data)
C
Christoph Hellwig 已提交
2715
		goto out;
2716

L
Li Zefan 已提交
2717 2718 2719 2720
	ret = -EROFS;
	if (btrfs_root_readonly(root))
		goto out;

2721
	ret = mnt_want_write_file(file);
Y
Yan Zheng 已提交
2722 2723 2724
	if (ret)
		goto out;

J
Josef Bacik 已提交
2725
	atomic_inc(&root->fs_info->open_ioctl_trans);
2726

2727
	ret = -ENOMEM;
2728
	trans = btrfs_start_ioctl_transaction(root);
2729
	if (IS_ERR(trans))
2730 2731 2732 2733 2734 2735
		goto out_drop;

	file->private_data = trans;
	return 0;

out_drop:
J
Josef Bacik 已提交
2736
	atomic_dec(&root->fs_info->open_ioctl_trans);
A
Al Viro 已提交
2737
	mnt_drop_write_file(file);
C
Christoph Hellwig 已提交
2738 2739 2740 2741
out:
	return ret;
}

2742 2743 2744 2745 2746 2747 2748 2749 2750 2751 2752 2753 2754 2755 2756 2757 2758 2759 2760 2761 2762 2763 2764 2765 2766 2767 2768 2769 2770 2771 2772 2773 2774 2775 2776 2777 2778 2779 2780 2781 2782
static long btrfs_ioctl_default_subvol(struct file *file, void __user *argp)
{
	struct inode *inode = fdentry(file)->d_inode;
	struct btrfs_root *root = BTRFS_I(inode)->root;
	struct btrfs_root *new_root;
	struct btrfs_dir_item *di;
	struct btrfs_trans_handle *trans;
	struct btrfs_path *path;
	struct btrfs_key location;
	struct btrfs_disk_key disk_key;
	struct btrfs_super_block *disk_super;
	u64 features;
	u64 objectid = 0;
	u64 dir_id;

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

	if (copy_from_user(&objectid, argp, sizeof(objectid)))
		return -EFAULT;

	if (!objectid)
		objectid = root->root_key.objectid;

	location.objectid = objectid;
	location.type = BTRFS_ROOT_ITEM_KEY;
	location.offset = (u64)-1;

	new_root = btrfs_read_fs_root_no_name(root->fs_info, &location);
	if (IS_ERR(new_root))
		return PTR_ERR(new_root);

	if (btrfs_root_refs(&new_root->root_item) == 0)
		return -ENOENT;

	path = btrfs_alloc_path();
	if (!path)
		return -ENOMEM;
	path->leave_spinning = 1;

	trans = btrfs_start_transaction(root, 1);
2783
	if (IS_ERR(trans)) {
2784
		btrfs_free_path(path);
2785
		return PTR_ERR(trans);
2786 2787
	}

2788
	dir_id = btrfs_super_root_dir(root->fs_info->super_copy);
2789 2790
	di = btrfs_lookup_dir_item(trans, root->fs_info->tree_root, path,
				   dir_id, "default", 7, 1);
2791
	if (IS_ERR_OR_NULL(di)) {
2792 2793 2794 2795 2796 2797 2798 2799 2800 2801 2802 2803
		btrfs_free_path(path);
		btrfs_end_transaction(trans, root);
		printk(KERN_ERR "Umm, you don't have the default dir item, "
		       "this isn't going to work\n");
		return -ENOENT;
	}

	btrfs_cpu_key_to_disk(&disk_key, &new_root->root_key);
	btrfs_set_dir_item_key(path->nodes[0], di, &disk_key);
	btrfs_mark_buffer_dirty(path->nodes[0]);
	btrfs_free_path(path);

2804
	disk_super = root->fs_info->super_copy;
2805 2806 2807 2808 2809 2810 2811 2812 2813 2814
	features = btrfs_super_incompat_flags(disk_super);
	if (!(features & BTRFS_FEATURE_INCOMPAT_DEFAULT_SUBVOL)) {
		features |= BTRFS_FEATURE_INCOMPAT_DEFAULT_SUBVOL;
		btrfs_set_super_incompat_flags(disk_super, features);
	}
	btrfs_end_transaction(trans, root);

	return 0;
}

2815 2816 2817 2818 2819 2820 2821 2822 2823 2824 2825 2826 2827 2828 2829 2830
static void get_block_group_info(struct list_head *groups_list,
				 struct btrfs_ioctl_space_info *space)
{
	struct btrfs_block_group_cache *block_group;

	space->total_bytes = 0;
	space->used_bytes = 0;
	space->flags = 0;
	list_for_each_entry(block_group, groups_list, list) {
		space->flags = block_group->flags;
		space->total_bytes += block_group->key.offset;
		space->used_bytes +=
			btrfs_block_group_used(&block_group->item);
	}
}

J
Josef Bacik 已提交
2831 2832 2833 2834 2835
long btrfs_ioctl_space_info(struct btrfs_root *root, void __user *arg)
{
	struct btrfs_ioctl_space_args space_args;
	struct btrfs_ioctl_space_info space;
	struct btrfs_ioctl_space_info *dest;
2836
	struct btrfs_ioctl_space_info *dest_orig;
2837
	struct btrfs_ioctl_space_info __user *user_dest;
J
Josef Bacik 已提交
2838
	struct btrfs_space_info *info;
2839 2840 2841 2842 2843
	u64 types[] = {BTRFS_BLOCK_GROUP_DATA,
		       BTRFS_BLOCK_GROUP_SYSTEM,
		       BTRFS_BLOCK_GROUP_METADATA,
		       BTRFS_BLOCK_GROUP_DATA | BTRFS_BLOCK_GROUP_METADATA};
	int num_types = 4;
2844
	int alloc_size;
J
Josef Bacik 已提交
2845
	int ret = 0;
2846
	u64 slot_count = 0;
2847
	int i, c;
J
Josef Bacik 已提交
2848 2849 2850 2851 2852 2853

	if (copy_from_user(&space_args,
			   (struct btrfs_ioctl_space_args __user *)arg,
			   sizeof(space_args)))
		return -EFAULT;

2854 2855 2856 2857 2858 2859 2860 2861 2862 2863 2864 2865 2866 2867 2868 2869 2870 2871 2872 2873 2874 2875 2876 2877
	for (i = 0; i < num_types; i++) {
		struct btrfs_space_info *tmp;

		info = NULL;
		rcu_read_lock();
		list_for_each_entry_rcu(tmp, &root->fs_info->space_info,
					list) {
			if (tmp->flags == types[i]) {
				info = tmp;
				break;
			}
		}
		rcu_read_unlock();

		if (!info)
			continue;

		down_read(&info->groups_sem);
		for (c = 0; c < BTRFS_NR_RAID_TYPES; c++) {
			if (!list_empty(&info->block_groups[c]))
				slot_count++;
		}
		up_read(&info->groups_sem);
	}
2878 2879 2880 2881 2882 2883

	/* space_slots == 0 means they are asking for a count */
	if (space_args.space_slots == 0) {
		space_args.total_spaces = slot_count;
		goto out;
	}
2884

2885
	slot_count = min_t(u64, space_args.space_slots, slot_count);
2886

2887
	alloc_size = sizeof(*dest) * slot_count;
2888

2889 2890 2891 2892 2893 2894
	/* we generally have at most 6 or so space infos, one for each raid
	 * level.  So, a whole page should be more than enough for everyone
	 */
	if (alloc_size > PAGE_CACHE_SIZE)
		return -ENOMEM;

J
Josef Bacik 已提交
2895
	space_args.total_spaces = 0;
2896 2897 2898 2899
	dest = kmalloc(alloc_size, GFP_NOFS);
	if (!dest)
		return -ENOMEM;
	dest_orig = dest;
J
Josef Bacik 已提交
2900

2901
	/* now we have a buffer to copy into */
2902 2903 2904
	for (i = 0; i < num_types; i++) {
		struct btrfs_space_info *tmp;

2905 2906 2907
		if (!slot_count)
			break;

2908 2909 2910 2911 2912 2913 2914 2915 2916 2917
		info = NULL;
		rcu_read_lock();
		list_for_each_entry_rcu(tmp, &root->fs_info->space_info,
					list) {
			if (tmp->flags == types[i]) {
				info = tmp;
				break;
			}
		}
		rcu_read_unlock();
2918

2919 2920 2921 2922 2923 2924 2925 2926 2927 2928
		if (!info)
			continue;
		down_read(&info->groups_sem);
		for (c = 0; c < BTRFS_NR_RAID_TYPES; c++) {
			if (!list_empty(&info->block_groups[c])) {
				get_block_group_info(&info->block_groups[c],
						     &space);
				memcpy(dest, &space, sizeof(space));
				dest++;
				space_args.total_spaces++;
2929
				slot_count--;
2930
			}
2931 2932
			if (!slot_count)
				break;
2933 2934
		}
		up_read(&info->groups_sem);
J
Josef Bacik 已提交
2935 2936
	}

D
Daniel J Blueman 已提交
2937
	user_dest = (struct btrfs_ioctl_space_info __user *)
2938 2939 2940 2941 2942 2943 2944 2945
		(arg + sizeof(struct btrfs_ioctl_space_args));

	if (copy_to_user(user_dest, dest_orig, alloc_size))
		ret = -EFAULT;

	kfree(dest_orig);
out:
	if (ret == 0 && copy_to_user(arg, &space_args, sizeof(space_args)))
J
Josef Bacik 已提交
2946 2947 2948 2949 2950
		ret = -EFAULT;

	return ret;
}

C
Christoph Hellwig 已提交
2951 2952 2953 2954 2955 2956 2957 2958 2959 2960 2961 2962 2963
/*
 * there are many ways the trans_start and trans_end ioctls can lead
 * to deadlocks.  They should only be used by applications that
 * basically own the machine, and have a very in depth understanding
 * of all the possible deadlocks and enospc problems.
 */
long btrfs_ioctl_trans_end(struct file *file)
{
	struct inode *inode = fdentry(file)->d_inode;
	struct btrfs_root *root = BTRFS_I(inode)->root;
	struct btrfs_trans_handle *trans;

	trans = file->private_data;
2964 2965
	if (!trans)
		return -EINVAL;
C
Christoph Hellwig 已提交
2966
	file->private_data = NULL;
2967

2968 2969
	btrfs_end_transaction(trans, root);

J
Josef Bacik 已提交
2970
	atomic_dec(&root->fs_info->open_ioctl_trans);
2971

A
Al Viro 已提交
2972
	mnt_drop_write_file(file);
2973
	return 0;
C
Christoph Hellwig 已提交
2974 2975
}

2976 2977 2978 2979 2980
static noinline long btrfs_ioctl_start_sync(struct file *file, void __user *argp)
{
	struct btrfs_root *root = BTRFS_I(file->f_dentry->d_inode)->root;
	struct btrfs_trans_handle *trans;
	u64 transid;
T
Tsutomu Itoh 已提交
2981
	int ret;
2982 2983

	trans = btrfs_start_transaction(root, 0);
2984 2985
	if (IS_ERR(trans))
		return PTR_ERR(trans);
2986
	transid = trans->transid;
T
Tsutomu Itoh 已提交
2987
	ret = btrfs_commit_transaction_async(trans, root, 0);
2988 2989
	if (ret) {
		btrfs_end_transaction(trans, root);
T
Tsutomu Itoh 已提交
2990
		return ret;
2991
	}
2992 2993 2994 2995 2996 2997 2998 2999 3000 3001 3002 3003 3004 3005 3006 3007 3008 3009 3010 3011 3012

	if (argp)
		if (copy_to_user(argp, &transid, sizeof(transid)))
			return -EFAULT;
	return 0;
}

static noinline long btrfs_ioctl_wait_sync(struct file *file, void __user *argp)
{
	struct btrfs_root *root = BTRFS_I(file->f_dentry->d_inode)->root;
	u64 transid;

	if (argp) {
		if (copy_from_user(&transid, argp, sizeof(transid)))
			return -EFAULT;
	} else {
		transid = 0;  /* current trans */
	}
	return btrfs_wait_for_commit(root, transid);
}

J
Jan Schmidt 已提交
3013 3014 3015 3016 3017 3018 3019 3020 3021 3022 3023 3024 3025
static long btrfs_ioctl_scrub(struct btrfs_root *root, void __user *arg)
{
	int ret;
	struct btrfs_ioctl_scrub_args *sa;

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

	sa = memdup_user(arg, sizeof(*sa));
	if (IS_ERR(sa))
		return PTR_ERR(sa);

	ret = btrfs_scrub_dev(root, sa->devid, sa->start, sa->end,
A
Arne Jansen 已提交
3026
			      &sa->progress, sa->flags & BTRFS_SCRUB_READONLY);
J
Jan Schmidt 已提交
3027 3028 3029 3030 3031 3032 3033 3034 3035 3036 3037 3038 3039 3040 3041 3042 3043 3044 3045 3046 3047 3048 3049 3050 3051 3052 3053 3054 3055 3056 3057 3058 3059 3060 3061 3062 3063 3064

	if (copy_to_user(arg, sa, sizeof(*sa)))
		ret = -EFAULT;

	kfree(sa);
	return ret;
}

static long btrfs_ioctl_scrub_cancel(struct btrfs_root *root, void __user *arg)
{
	if (!capable(CAP_SYS_ADMIN))
		return -EPERM;

	return btrfs_scrub_cancel(root);
}

static long btrfs_ioctl_scrub_progress(struct btrfs_root *root,
				       void __user *arg)
{
	struct btrfs_ioctl_scrub_args *sa;
	int ret;

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

	sa = memdup_user(arg, sizeof(*sa));
	if (IS_ERR(sa))
		return PTR_ERR(sa);

	ret = btrfs_scrub_progress(root, sa->devid, &sa->progress);

	if (copy_to_user(arg, sa, sizeof(*sa)))
		ret = -EFAULT;

	kfree(sa);
	return ret;
}

3065 3066 3067 3068 3069 3070 3071 3072 3073 3074 3075 3076 3077 3078 3079 3080 3081 3082 3083 3084 3085 3086
static long btrfs_ioctl_get_dev_stats(struct btrfs_root *root,
				      void __user *arg, int reset_after_read)
{
	struct btrfs_ioctl_get_dev_stats *sa;
	int ret;

	if (reset_after_read && !capable(CAP_SYS_ADMIN))
		return -EPERM;

	sa = memdup_user(arg, sizeof(*sa));
	if (IS_ERR(sa))
		return PTR_ERR(sa);

	ret = btrfs_get_dev_stats(root, sa, reset_after_read);

	if (copy_to_user(arg, sa, sizeof(*sa)))
		ret = -EFAULT;

	kfree(sa);
	return ret;
}

3087 3088 3089 3090
static long btrfs_ioctl_ino_to_path(struct btrfs_root *root, void __user *arg)
{
	int ret = 0;
	int i;
3091
	u64 rel_ptr;
3092
	int size;
3093
	struct btrfs_ioctl_ino_path_args *ipa = NULL;
3094 3095 3096 3097 3098 3099 3100 3101 3102 3103 3104 3105 3106 3107 3108 3109 3110 3111 3112 3113 3114 3115 3116 3117 3118 3119 3120 3121 3122 3123 3124 3125
	struct inode_fs_paths *ipath = NULL;
	struct btrfs_path *path;

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

	path = btrfs_alloc_path();
	if (!path) {
		ret = -ENOMEM;
		goto out;
	}

	ipa = memdup_user(arg, sizeof(*ipa));
	if (IS_ERR(ipa)) {
		ret = PTR_ERR(ipa);
		ipa = NULL;
		goto out;
	}

	size = min_t(u32, ipa->size, 4096);
	ipath = init_ipath(size, root, path);
	if (IS_ERR(ipath)) {
		ret = PTR_ERR(ipath);
		ipath = NULL;
		goto out;
	}

	ret = paths_from_inode(ipa->inum, ipath);
	if (ret < 0)
		goto out;

	for (i = 0; i < ipath->fspath->elem_cnt; ++i) {
3126 3127
		rel_ptr = ipath->fspath->val[i] -
			  (u64)(unsigned long)ipath->fspath->val;
3128
		ipath->fspath->val[i] = rel_ptr;
3129 3130
	}

3131 3132
	ret = copy_to_user((void *)(unsigned long)ipa->fspath,
			   (void *)(unsigned long)ipath->fspath, size);
3133 3134 3135 3136 3137 3138 3139 3140 3141 3142 3143 3144 3145 3146 3147 3148 3149 3150 3151 3152 3153 3154 3155 3156 3157 3158 3159 3160 3161 3162 3163 3164 3165 3166 3167 3168 3169 3170
	if (ret) {
		ret = -EFAULT;
		goto out;
	}

out:
	btrfs_free_path(path);
	free_ipath(ipath);
	kfree(ipa);

	return ret;
}

static int build_ino_list(u64 inum, u64 offset, u64 root, void *ctx)
{
	struct btrfs_data_container *inodes = ctx;
	const size_t c = 3 * sizeof(u64);

	if (inodes->bytes_left >= c) {
		inodes->bytes_left -= c;
		inodes->val[inodes->elem_cnt] = inum;
		inodes->val[inodes->elem_cnt + 1] = offset;
		inodes->val[inodes->elem_cnt + 2] = root;
		inodes->elem_cnt += 3;
	} else {
		inodes->bytes_missing += c - inodes->bytes_left;
		inodes->bytes_left = 0;
		inodes->elem_missed += 3;
	}

	return 0;
}

static long btrfs_ioctl_logical_to_ino(struct btrfs_root *root,
					void __user *arg)
{
	int ret = 0;
	int size;
J
Jan Schmidt 已提交
3171
	u64 extent_item_pos;
3172 3173 3174 3175 3176 3177 3178 3179 3180 3181 3182 3183 3184 3185 3186 3187 3188 3189 3190 3191 3192 3193 3194 3195 3196 3197 3198 3199 3200 3201
	struct btrfs_ioctl_logical_ino_args *loi;
	struct btrfs_data_container *inodes = NULL;
	struct btrfs_path *path = NULL;
	struct btrfs_key key;

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

	loi = memdup_user(arg, sizeof(*loi));
	if (IS_ERR(loi)) {
		ret = PTR_ERR(loi);
		loi = NULL;
		goto out;
	}

	path = btrfs_alloc_path();
	if (!path) {
		ret = -ENOMEM;
		goto out;
	}

	size = min_t(u32, loi->size, 4096);
	inodes = init_data_container(size);
	if (IS_ERR(inodes)) {
		ret = PTR_ERR(inodes);
		inodes = NULL;
		goto out;
	}

	ret = extent_from_logical(root->fs_info, loi->logical, path, &key);
J
Jan Schmidt 已提交
3202
	btrfs_release_path(path);
3203 3204 3205 3206 3207 3208

	if (ret & BTRFS_EXTENT_FLAG_TREE_BLOCK)
		ret = -ENOENT;
	if (ret < 0)
		goto out;

J
Jan Schmidt 已提交
3209
	extent_item_pos = loi->logical - key.objectid;
3210 3211
	ret = iterate_extent_inodes(root->fs_info, key.objectid,
					extent_item_pos, 0, build_ino_list,
J
Jan Schmidt 已提交
3212
					inodes);
3213 3214 3215 3216

	if (ret < 0)
		goto out;

3217 3218
	ret = copy_to_user((void *)(unsigned long)loi->inodes,
			   (void *)(unsigned long)inodes, size);
3219 3220 3221 3222 3223 3224 3225 3226 3227 3228 3229
	if (ret)
		ret = -EFAULT;

out:
	btrfs_free_path(path);
	kfree(inodes);
	kfree(loi);

	return ret;
}

3230
void update_ioctl_balance_args(struct btrfs_fs_info *fs_info, int lock,
3231 3232 3233 3234 3235 3236
			       struct btrfs_ioctl_balance_args *bargs)
{
	struct btrfs_balance_control *bctl = fs_info->balance_ctl;

	bargs->flags = bctl->flags;

3237 3238 3239 3240
	if (atomic_read(&fs_info->balance_running))
		bargs->state |= BTRFS_BALANCE_STATE_RUNNING;
	if (atomic_read(&fs_info->balance_pause_req))
		bargs->state |= BTRFS_BALANCE_STATE_PAUSE_REQ;
3241 3242
	if (atomic_read(&fs_info->balance_cancel_req))
		bargs->state |= BTRFS_BALANCE_STATE_CANCEL_REQ;
3243

3244 3245 3246
	memcpy(&bargs->data, &bctl->data, sizeof(bargs->data));
	memcpy(&bargs->meta, &bctl->meta, sizeof(bargs->meta));
	memcpy(&bargs->sys, &bctl->sys, sizeof(bargs->sys));
3247 3248 3249 3250 3251 3252 3253 3254

	if (lock) {
		spin_lock(&fs_info->balance_lock);
		memcpy(&bargs->stat, &bctl->stat, sizeof(bargs->stat));
		spin_unlock(&fs_info->balance_lock);
	} else {
		memcpy(&bargs->stat, &bctl->stat, sizeof(bargs->stat));
	}
3255 3256
}

3257
static long btrfs_ioctl_balance(struct file *file, void __user *arg)
3258
{
3259
	struct btrfs_root *root = BTRFS_I(fdentry(file)->d_inode)->root;
3260 3261 3262 3263 3264 3265 3266 3267 3268 3269 3270
	struct btrfs_fs_info *fs_info = root->fs_info;
	struct btrfs_ioctl_balance_args *bargs;
	struct btrfs_balance_control *bctl;
	int ret;

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

	if (fs_info->sb->s_flags & MS_RDONLY)
		return -EROFS;

3271
	ret = mnt_want_write_file(file);
3272 3273 3274
	if (ret)
		return ret;

3275 3276 3277 3278 3279 3280 3281 3282 3283
	mutex_lock(&fs_info->volume_mutex);
	mutex_lock(&fs_info->balance_mutex);

	if (arg) {
		bargs = memdup_user(arg, sizeof(*bargs));
		if (IS_ERR(bargs)) {
			ret = PTR_ERR(bargs);
			goto out;
		}
3284 3285 3286 3287 3288 3289 3290 3291 3292 3293 3294 3295 3296 3297

		if (bargs->flags & BTRFS_BALANCE_RESUME) {
			if (!fs_info->balance_ctl) {
				ret = -ENOTCONN;
				goto out_bargs;
			}

			bctl = fs_info->balance_ctl;
			spin_lock(&fs_info->balance_lock);
			bctl->flags |= BTRFS_BALANCE_RESUME;
			spin_unlock(&fs_info->balance_lock);

			goto do_balance;
		}
3298 3299 3300 3301
	} else {
		bargs = NULL;
	}

3302 3303 3304 3305 3306
	if (fs_info->balance_ctl) {
		ret = -EINPROGRESS;
		goto out_bargs;
	}

3307 3308 3309 3310 3311 3312 3313 3314 3315 3316 3317 3318 3319
	bctl = kzalloc(sizeof(*bctl), GFP_NOFS);
	if (!bctl) {
		ret = -ENOMEM;
		goto out_bargs;
	}

	bctl->fs_info = fs_info;
	if (arg) {
		memcpy(&bctl->data, &bargs->data, sizeof(bctl->data));
		memcpy(&bctl->meta, &bargs->meta, sizeof(bctl->meta));
		memcpy(&bctl->sys, &bargs->sys, sizeof(bctl->sys));

		bctl->flags = bargs->flags;
3320 3321 3322
	} else {
		/* balance everything - no filters */
		bctl->flags |= BTRFS_BALANCE_TYPE_MASK;
3323 3324
	}

3325
do_balance:
3326 3327
	ret = btrfs_balance(bctl, bargs);
	/*
3328 3329
	 * bctl is freed in __cancel_balance or in free_fs_info if
	 * restriper was paused all the way until unmount
3330 3331 3332 3333 3334 3335 3336 3337 3338 3339 3340
	 */
	if (arg) {
		if (copy_to_user(arg, bargs, sizeof(*bargs)))
			ret = -EFAULT;
	}

out_bargs:
	kfree(bargs);
out:
	mutex_unlock(&fs_info->balance_mutex);
	mutex_unlock(&fs_info->volume_mutex);
3341
	mnt_drop_write_file(file);
3342 3343 3344
	return ret;
}

3345 3346 3347 3348 3349 3350 3351 3352
static long btrfs_ioctl_balance_ctl(struct btrfs_root *root, int cmd)
{
	if (!capable(CAP_SYS_ADMIN))
		return -EPERM;

	switch (cmd) {
	case BTRFS_BALANCE_CTL_PAUSE:
		return btrfs_pause_balance(root->fs_info);
3353 3354
	case BTRFS_BALANCE_CTL_CANCEL:
		return btrfs_cancel_balance(root->fs_info);
3355 3356 3357 3358 3359
	}

	return -EINVAL;
}

3360 3361 3362 3363 3364 3365 3366 3367 3368 3369 3370 3371 3372 3373 3374 3375 3376 3377 3378 3379 3380 3381 3382 3383 3384 3385 3386 3387 3388 3389 3390 3391 3392
static long btrfs_ioctl_balance_progress(struct btrfs_root *root,
					 void __user *arg)
{
	struct btrfs_fs_info *fs_info = root->fs_info;
	struct btrfs_ioctl_balance_args *bargs;
	int ret = 0;

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

	mutex_lock(&fs_info->balance_mutex);
	if (!fs_info->balance_ctl) {
		ret = -ENOTCONN;
		goto out;
	}

	bargs = kzalloc(sizeof(*bargs), GFP_NOFS);
	if (!bargs) {
		ret = -ENOMEM;
		goto out;
	}

	update_ioctl_balance_args(fs_info, 1, bargs);

	if (copy_to_user(arg, bargs, sizeof(*bargs)))
		ret = -EFAULT;

	kfree(bargs);
out:
	mutex_unlock(&fs_info->balance_mutex);
	return ret;
}

C
Christoph Hellwig 已提交
3393 3394 3395 3396
long btrfs_ioctl(struct file *file, unsigned int
		cmd, unsigned long arg)
{
	struct btrfs_root *root = BTRFS_I(fdentry(file)->d_inode)->root;
3397
	void __user *argp = (void __user *)arg;
C
Christoph Hellwig 已提交
3398 3399

	switch (cmd) {
3400 3401 3402 3403 3404 3405
	case FS_IOC_GETFLAGS:
		return btrfs_ioctl_getflags(file, argp);
	case FS_IOC_SETFLAGS:
		return btrfs_ioctl_setflags(file, argp);
	case FS_IOC_GETVERSION:
		return btrfs_ioctl_getversion(file, argp);
3406 3407
	case FITRIM:
		return btrfs_ioctl_fitrim(file, argp);
C
Christoph Hellwig 已提交
3408
	case BTRFS_IOC_SNAP_CREATE:
3409
		return btrfs_ioctl_snap_create(file, argp, 0);
3410
	case BTRFS_IOC_SNAP_CREATE_V2:
3411
		return btrfs_ioctl_snap_create_v2(file, argp, 0);
3412
	case BTRFS_IOC_SUBVOL_CREATE:
3413
		return btrfs_ioctl_snap_create(file, argp, 1);
3414 3415
	case BTRFS_IOC_SNAP_DESTROY:
		return btrfs_ioctl_snap_destroy(file, argp);
3416 3417 3418 3419
	case BTRFS_IOC_SUBVOL_GETFLAGS:
		return btrfs_ioctl_subvol_getflags(file, argp);
	case BTRFS_IOC_SUBVOL_SETFLAGS:
		return btrfs_ioctl_subvol_setflags(file, argp);
3420 3421
	case BTRFS_IOC_DEFAULT_SUBVOL:
		return btrfs_ioctl_default_subvol(file, argp);
C
Christoph Hellwig 已提交
3422
	case BTRFS_IOC_DEFRAG:
C
Chris Mason 已提交
3423 3424 3425
		return btrfs_ioctl_defrag(file, NULL);
	case BTRFS_IOC_DEFRAG_RANGE:
		return btrfs_ioctl_defrag(file, argp);
C
Christoph Hellwig 已提交
3426
	case BTRFS_IOC_RESIZE:
3427
		return btrfs_ioctl_resize(root, argp);
C
Christoph Hellwig 已提交
3428
	case BTRFS_IOC_ADD_DEV:
3429
		return btrfs_ioctl_add_dev(root, argp);
C
Christoph Hellwig 已提交
3430
	case BTRFS_IOC_RM_DEV:
3431
		return btrfs_ioctl_rm_dev(root, argp);
J
Jan Schmidt 已提交
3432 3433 3434 3435
	case BTRFS_IOC_FS_INFO:
		return btrfs_ioctl_fs_info(root, argp);
	case BTRFS_IOC_DEV_INFO:
		return btrfs_ioctl_dev_info(root, argp);
C
Christoph Hellwig 已提交
3436
	case BTRFS_IOC_BALANCE:
3437
		return btrfs_ioctl_balance(file, NULL);
C
Christoph Hellwig 已提交
3438
	case BTRFS_IOC_CLONE:
3439 3440
		return btrfs_ioctl_clone(file, arg, 0, 0, 0);
	case BTRFS_IOC_CLONE_RANGE:
3441
		return btrfs_ioctl_clone_range(file, argp);
C
Christoph Hellwig 已提交
3442 3443 3444 3445
	case BTRFS_IOC_TRANS_START:
		return btrfs_ioctl_trans_start(file);
	case BTRFS_IOC_TRANS_END:
		return btrfs_ioctl_trans_end(file);
3446 3447 3448 3449
	case BTRFS_IOC_TREE_SEARCH:
		return btrfs_ioctl_tree_search(file, argp);
	case BTRFS_IOC_INO_LOOKUP:
		return btrfs_ioctl_ino_lookup(file, argp);
3450 3451 3452 3453
	case BTRFS_IOC_INO_PATHS:
		return btrfs_ioctl_ino_to_path(root, argp);
	case BTRFS_IOC_LOGICAL_INO:
		return btrfs_ioctl_logical_to_ino(root, argp);
J
Josef Bacik 已提交
3454 3455
	case BTRFS_IOC_SPACE_INFO:
		return btrfs_ioctl_space_info(root, argp);
C
Christoph Hellwig 已提交
3456 3457 3458
	case BTRFS_IOC_SYNC:
		btrfs_sync_fs(file->f_dentry->d_sb, 1);
		return 0;
3459 3460 3461 3462
	case BTRFS_IOC_START_SYNC:
		return btrfs_ioctl_start_sync(file, argp);
	case BTRFS_IOC_WAIT_SYNC:
		return btrfs_ioctl_wait_sync(file, argp);
J
Jan Schmidt 已提交
3463 3464 3465 3466 3467 3468
	case BTRFS_IOC_SCRUB:
		return btrfs_ioctl_scrub(root, argp);
	case BTRFS_IOC_SCRUB_CANCEL:
		return btrfs_ioctl_scrub_cancel(root, argp);
	case BTRFS_IOC_SCRUB_PROGRESS:
		return btrfs_ioctl_scrub_progress(root, argp);
3469
	case BTRFS_IOC_BALANCE_V2:
3470
		return btrfs_ioctl_balance(file, argp);
3471 3472
	case BTRFS_IOC_BALANCE_CTL:
		return btrfs_ioctl_balance_ctl(root, arg);
3473 3474
	case BTRFS_IOC_BALANCE_PROGRESS:
		return btrfs_ioctl_balance_progress(root, argp);
3475 3476 3477 3478
	case BTRFS_IOC_GET_DEV_STATS:
		return btrfs_ioctl_get_dev_stats(root, argp, 0);
	case BTRFS_IOC_GET_AND_RESET_DEV_STATS:
		return btrfs_ioctl_get_dev_stats(root, argp, 1);
C
Christoph Hellwig 已提交
3479 3480 3481 3482
	}

	return -ENOTTY;
}