ioctl.c 88.8 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>
44
#include <linux/uuid.h>
C
Chris Mason 已提交
45
#include "compat.h"
C
Christoph Hellwig 已提交
46 47 48 49 50 51 52
#include "ctree.h"
#include "disk-io.h"
#include "transaction.h"
#include "btrfs_inode.h"
#include "ioctl.h"
#include "print-tree.h"
#include "volumes.h"
53
#include "locking.h"
54
#include "inode-map.h"
55
#include "backref.h"
56
#include "rcu-string.h"
57
#include "send.h"
C
Christoph Hellwig 已提交
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 87 88
/* 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
89 90 91 92 93 94 95
	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;
96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123

	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.
 *
124
 * Currently only the compression flags and the cow flags are inherited.
125 126 127
 */
void btrfs_inherit_iflags(struct inode *inode, struct inode *dir)
{
128 129 130 131 132 133
	unsigned int flags;

	if (!dir)
		return;

	flags = BTRFS_I(dir)->flags;
134

135 136 137 138 139 140 141 142 143 144
	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;
145 146 147 148 149 150 151 152 153 154 155 156 157 158

	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;
}

159 160 161 162 163
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
164 165
		      FS_NOCOMP_FL | FS_COMPR_FL |
		      FS_NOCOW_FL))
166 167 168 169 170 171 172 173
		return -EOPNOTSUPP;

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

	return 0;
}

174 175 176 177 178 179 180 181
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;
182 183
	u64 ip_oldflags;
	unsigned int i_oldflags;
184

L
Li Zefan 已提交
185 186 187
	if (btrfs_root_readonly(root))
		return -EROFS;

188 189 190
	if (copy_from_user(&flags, arg, sizeof(flags)))
		return -EFAULT;

191 192 193
	ret = check_flags(flags);
	if (ret)
		return ret;
C
Christoph Hellwig 已提交
194

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

198 199 200 201
	ret = mnt_want_write_file(file);
	if (ret)
		return ret;

202 203
	mutex_lock(&inode->i_mutex);

204 205 206
	ip_oldflags = ip->flags;
	i_oldflags = inode->i_flags;

207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239
	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;
		}
	}

	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
240 241 242 243
	if (flags & FS_NOCOW_FL)
		ip->flags |= BTRFS_INODE_NODATACOW;
	else
		ip->flags &= ~BTRFS_INODE_NODATACOW;
244

245 246 247 248 249 250 251 252 253 254 255
	/*
	 * 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
256 257
	} else {
		ip->flags &= ~(BTRFS_INODE_COMPRESS | BTRFS_INODE_NOCOMPRESS);
258
	}
259

260
	trans = btrfs_start_transaction(root, 1);
261 262 263 264
	if (IS_ERR(trans)) {
		ret = PTR_ERR(trans);
		goto out_drop;
	}
265

266
	btrfs_update_iflags(inode);
267
	inode_inc_iversion(inode);
268
	inode->i_ctime = CURRENT_TIME;
269 270 271
	ret = btrfs_update_inode(trans, root, inode);

	btrfs_end_transaction(trans, root);
272 273 274 275 276
 out_drop:
	if (ret) {
		ip->flags = ip_oldflags;
		inode->i_flags = i_oldflags;
	}
277 278 279

 out_unlock:
	mutex_unlock(&inode->i_mutex);
280
	mnt_drop_write_file(file);
281
	return ret;
282 283 284 285 286 287 288 289
}

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
290

291 292
static noinline int btrfs_ioctl_fitrim(struct file *file, void __user *arg)
{
293
	struct btrfs_fs_info *fs_info = btrfs_sb(fdentry(file)->d_sb);
294 295 296 297 298
	struct btrfs_device *device;
	struct request_queue *q;
	struct fstrim_range range;
	u64 minlen = ULLONG_MAX;
	u64 num_devices = 0;
299
	u64 total_bytes = btrfs_super_total_bytes(fs_info->super_copy);
300 301 302 303 304
	int ret;

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

305 306 307
	rcu_read_lock();
	list_for_each_entry_rcu(device, &fs_info->fs_devices->devices,
				dev_list) {
308 309 310 311 312 313 314 315 316
		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);
		}
	}
317
	rcu_read_unlock();
318

319 320 321 322
	if (!num_devices)
		return -EOPNOTSUPP;
	if (copy_from_user(&range, arg, sizeof(range)))
		return -EFAULT;
323 324
	if (range.start > total_bytes)
		return -EINVAL;
325

326
	range.len = min(range.len, total_bytes - range.start);
327
	range.minlen = max(range.minlen, minlen);
328
	ret = btrfs_trim_fs(fs_info->tree_root, &range);
329 330 331 332 333 334 335 336 337
	if (ret < 0)
		return ret;

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

	return 0;
}

338 339
static noinline int create_subvol(struct btrfs_root *root,
				  struct dentry *dentry,
S
Sage Weil 已提交
340
				  char *name, int namelen,
A
Arne Jansen 已提交
341 342
				  u64 *async_transid,
				  struct btrfs_qgroup_inherit **inherit)
C
Christoph Hellwig 已提交
343 344 345 346 347 348
{
	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;
349
	struct btrfs_root *new_root;
350
	struct dentry *parent = dentry->d_parent;
351
	struct inode *dir;
352
	struct timespec cur_time = CURRENT_TIME;
C
Christoph Hellwig 已提交
353 354 355 356
	int ret;
	int err;
	u64 objectid;
	u64 new_dirid = BTRFS_FIRST_FREE_OBJECTID;
357
	u64 index = 0;
358
	uuid_le new_uuid;
C
Christoph Hellwig 已提交
359

360
	ret = btrfs_find_free_objectid(root->fs_info->tree_root, &objectid);
361
	if (ret)
362
		return ret;
363 364 365

	dir = parent->d_inode;

J
Josef Bacik 已提交
366 367 368 369 370 371
	/*
	 * 1 - inode item
	 * 2 - refs
	 * 1 - root item
	 * 2 - dir items
	 */
372
	trans = btrfs_start_transaction(root, 6);
373
	if (IS_ERR(trans))
374
		return PTR_ERR(trans);
C
Christoph Hellwig 已提交
375

A
Arne Jansen 已提交
376 377 378 379 380
	ret = btrfs_qgroup_inherit(trans, root->fs_info, 0, objectid,
				   inherit ? *inherit : NULL);
	if (ret)
		goto fail;

381
	leaf = btrfs_alloc_free_block(trans, root, root->leafsize,
382
				      0, objectid, NULL, 0, 0, 0);
383 384 385 386
	if (IS_ERR(leaf)) {
		ret = PTR_ERR(leaf);
		goto fail;
	}
C
Christoph Hellwig 已提交
387

388
	memset_extent_buffer(leaf, 0, 0, sizeof(struct btrfs_header));
C
Christoph Hellwig 已提交
389 390
	btrfs_set_header_bytenr(leaf, leaf->start);
	btrfs_set_header_generation(leaf, trans->transid);
391
	btrfs_set_header_backref_rev(leaf, BTRFS_MIXED_BACKREF_REV);
C
Christoph Hellwig 已提交
392 393 394 395 396
	btrfs_set_header_owner(leaf, objectid);

	write_extent_buffer(leaf, root->fs_info->fsid,
			    (unsigned long)btrfs_header_fsid(leaf),
			    BTRFS_FSID_SIZE);
397 398 399
	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
400 401
	btrfs_mark_buffer_dirty(leaf);

402 403
	memset(&root_item, 0, sizeof(root_item));

C
Christoph Hellwig 已提交
404 405 406 407
	inode_item = &root_item.inode;
	inode_item->generation = cpu_to_le64(1);
	inode_item->size = cpu_to_le64(3);
	inode_item->nlink = cpu_to_le32(1);
408
	inode_item->nbytes = cpu_to_le64(root->leafsize);
C
Christoph Hellwig 已提交
409 410
	inode_item->mode = cpu_to_le32(S_IFDIR | 0755);

411 412 413 414
	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
415
	btrfs_set_root_bytenr(&root_item, leaf->start);
416
	btrfs_set_root_generation(&root_item, trans->transid);
C
Christoph Hellwig 已提交
417 418
	btrfs_set_root_level(&root_item, 0);
	btrfs_set_root_refs(&root_item, 1);
419
	btrfs_set_root_used(&root_item, leaf->len);
Y
Yan Zheng 已提交
420
	btrfs_set_root_last_snapshot(&root_item, 0);
C
Christoph Hellwig 已提交
421

422 423 424 425 426
	btrfs_set_root_generation_v2(&root_item,
			btrfs_root_generation(&root_item));
	uuid_le_gen(&new_uuid);
	memcpy(root_item.uuid, new_uuid.b, BTRFS_UUID_SIZE);
	root_item.otime.sec = cpu_to_le64(cur_time.tv_sec);
427
	root_item.otime.nsec = cpu_to_le32(cur_time.tv_nsec);
428 429 430
	root_item.ctime = root_item.otime;
	btrfs_set_root_ctransid(&root_item, trans->transid);
	btrfs_set_root_otransid(&root_item, trans->transid);
C
Christoph Hellwig 已提交
431

432
	btrfs_tree_unlock(leaf);
C
Christoph Hellwig 已提交
433 434 435 436 437 438
	free_extent_buffer(leaf);
	leaf = NULL;

	btrfs_set_root_dirid(&root_item, new_dirid);

	key.objectid = objectid;
439
	key.offset = 0;
C
Christoph Hellwig 已提交
440 441 442 443 444 445
	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;

446 447
	key.offset = (u64)-1;
	new_root = btrfs_read_fs_root_no_name(root->fs_info, &key);
448 449 450 451 452
	if (IS_ERR(new_root)) {
		btrfs_abort_transaction(trans, root, PTR_ERR(new_root));
		ret = PTR_ERR(new_root);
		goto fail;
	}
453 454 455

	btrfs_record_root_in_trans(trans, new_root);

456
	ret = btrfs_create_subvol_root(trans, new_root, new_dirid);
457 458
	if (ret) {
		/* We potentially lose an unused inode item here */
459
		btrfs_abort_transaction(trans, root, ret);
460 461 462
		goto fail;
	}

C
Christoph Hellwig 已提交
463 464 465
	/*
	 * insert the directory item
	 */
466
	ret = btrfs_set_inode_index(dir, &index);
467 468 469 470
	if (ret) {
		btrfs_abort_transaction(trans, root, ret);
		goto fail;
	}
471 472

	ret = btrfs_insert_dir_item(trans, root,
473
				    name, namelen, dir, &key,
474
				    BTRFS_FT_DIR, index);
475 476
	if (ret) {
		btrfs_abort_transaction(trans, root, ret);
C
Christoph Hellwig 已提交
477
		goto fail;
478
	}
479

480 481 482 483
	btrfs_i_size_write(dir, dir->i_size + namelen * 2);
	ret = btrfs_update_inode(trans, root, dir);
	BUG_ON(ret);

484
	ret = btrfs_add_root_ref(trans, root->fs_info->tree_root,
485
				 objectid, root->root_key.objectid,
L
Li Zefan 已提交
486
				 btrfs_ino(dir), index, name, namelen);
487

488
	BUG_ON(ret);
C
Christoph Hellwig 已提交
489

490
	d_instantiate(dentry, btrfs_lookup_dentry(dir, dentry));
C
Christoph Hellwig 已提交
491
fail:
S
Sage Weil 已提交
492 493 494 495 496 497
	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
498 499 500 501 502
	if (err && !ret)
		ret = err;
	return ret;
}

S
Sage Weil 已提交
503
static int create_snapshot(struct btrfs_root *root, struct dentry *dentry,
L
Li Zefan 已提交
504
			   char *name, int namelen, u64 *async_transid,
A
Arne Jansen 已提交
505
			   bool readonly, struct btrfs_qgroup_inherit **inherit)
C
Christoph Hellwig 已提交
506
{
507
	struct inode *inode;
C
Christoph Hellwig 已提交
508 509
	struct btrfs_pending_snapshot *pending_snapshot;
	struct btrfs_trans_handle *trans;
510
	int ret;
C
Christoph Hellwig 已提交
511 512 513 514

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

515
	pending_snapshot = kzalloc(sizeof(*pending_snapshot), GFP_NOFS);
516 517 518
	if (!pending_snapshot)
		return -ENOMEM;

519 520
	btrfs_init_block_rsv(&pending_snapshot->block_rsv,
			     BTRFS_BLOCK_RSV_TEMP);
521
	pending_snapshot->dentry = dentry;
C
Christoph Hellwig 已提交
522
	pending_snapshot->root = root;
L
Li Zefan 已提交
523
	pending_snapshot->readonly = readonly;
A
Arne Jansen 已提交
524 525 526 527
	if (inherit) {
		pending_snapshot->inherit = *inherit;
		*inherit = NULL;	/* take responsibility to free it */
	}
528

529
	trans = btrfs_start_transaction(root->fs_info->extent_root, 6);
530 531 532 533 534 535 536 537
	if (IS_ERR(trans)) {
		ret = PTR_ERR(trans);
		goto fail;
	}

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

538
	spin_lock(&root->fs_info->trans_lock);
C
Christoph Hellwig 已提交
539 540
	list_add(&pending_snapshot->list,
		 &trans->transaction->pending_snapshots);
541
	spin_unlock(&root->fs_info->trans_lock);
S
Sage Weil 已提交
542 543 544 545 546 547 548 549
	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);
	}
550
	BUG_ON(ret);
551 552 553 554 555

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

556 557 558
	ret = btrfs_orphan_cleanup(pending_snapshot->snap);
	if (ret)
		goto fail;
C
Christoph Hellwig 已提交
559

560
	inode = btrfs_lookup_dentry(dentry->d_parent->d_inode, dentry);
561 562 563 564 565 566 567 568
	if (IS_ERR(inode)) {
		ret = PTR_ERR(inode);
		goto fail;
	}
	BUG_ON(!inode);
	d_instantiate(dentry, inode);
	ret = 0;
fail:
569
	kfree(pending_snapshot);
C
Christoph Hellwig 已提交
570 571 572
	return ret;
}

573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642
/*  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;
}

643 644 645 646 647 648 649 650 651 652 653 654 655 656 657
/* 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.
 */
658 659
static noinline int btrfs_mksubvol(struct path *parent,
				   char *name, int namelen,
S
Sage Weil 已提交
660
				   struct btrfs_root *snap_src,
A
Arne Jansen 已提交
661 662
				   u64 *async_transid, bool readonly,
				   struct btrfs_qgroup_inherit **inherit)
663
{
664
	struct inode *dir  = parent->dentry->d_inode;
665 666 667
	struct dentry *dentry;
	int error;

668
	mutex_lock_nested(&dir->i_mutex, I_MUTEX_PARENT);
669 670 671 672 673 674 675 676 677 678

	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;

679
	error = btrfs_may_create(dir, dentry);
680
	if (error)
681
		goto out_dput;
682

683 684 685 686 687
	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;

688
	if (snap_src) {
A
Arne Jansen 已提交
689 690
		error = create_snapshot(snap_src, dentry, name, namelen,
					async_transid, readonly, inherit);
691
	} else {
692
		error = create_subvol(BTRFS_I(dir)->root, dentry,
A
Arne Jansen 已提交
693
				      name, namelen, async_transid, inherit);
694
	}
695 696 697 698
	if (!error)
		fsnotify_mkdir(dir, dentry);
out_up_read:
	up_read(&BTRFS_I(dir)->root->fs_info->subvol_sem);
699 700 701
out_dput:
	dput(dentry);
out_unlock:
702
	mutex_unlock(&dir->i_mutex);
703 704 705
	return error;
}

C
Chris Mason 已提交
706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756
/*
 * 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;
757
	u64 ino = btrfs_ino(inode);
C
Chris Mason 已提交
758 759 760 761 762

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

763
	min_key.objectid = ino;
C
Chris Mason 已提交
764 765 766
	min_key.type = BTRFS_EXTENT_DATA_KEY;
	min_key.offset = *off;

767
	max_key.objectid = ino;
C
Chris Mason 已提交
768 769 770 771 772 773 774 775 776 777
	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;
778
		if (min_key.objectid != ino)
C
Chris Mason 已提交
779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806
			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
807
static struct extent_map *defrag_lookup_extent(struct inode *inode, u64 start)
808 809
{
	struct extent_map_tree *em_tree = &BTRFS_I(inode)->extent_tree;
L
Li Zefan 已提交
810 811 812
	struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree;
	struct extent_map *em;
	u64 len = PAGE_CACHE_SIZE;
813

L
Li Zefan 已提交
814 815 816 817
	/*
	 * hopefully we have this extent in the tree already, try without
	 * the full extent lock
	 */
818
	read_lock(&em_tree->lock);
L
Li Zefan 已提交
819
	em = lookup_extent_mapping(em_tree, start, len);
820 821
	read_unlock(&em_tree->lock);

L
Li Zefan 已提交
822 823 824 825 826 827 828 829 830 831 832 833
	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;
}
834

L
Li Zefan 已提交
835 836 837 838 839 840 841 842 843 844 845 846 847 848
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);
849 850 851
	return ret;
}

L
Li Zefan 已提交
852
static int should_defrag_range(struct inode *inode, u64 start, int thresh,
853 854
			       u64 *last_len, u64 *skip, u64 *defrag_end,
			       int compress)
855
{
L
Li Zefan 已提交
856
	struct extent_map *em;
857
	int ret = 1;
L
Li Zefan 已提交
858
	bool next_mergeable = true;
859 860

	/*
861
	 * make sure that once we start defragging an extent, we keep on
862 863 864 865 866 867 868
	 * defragging it
	 */
	if (start < *defrag_end)
		return 1;

	*skip = 0;

L
Li Zefan 已提交
869 870 871
	em = defrag_lookup_extent(inode, start);
	if (!em)
		return 0;
872 873

	/* this will cover holes, and inline extents */
874
	if (em->block_start >= EXTENT_MAP_LAST_BYTE) {
875
		ret = 0;
876 877 878
		goto out;
	}

L
Li Zefan 已提交
879
	next_mergeable = defrag_check_next_extent(inode, em);
880 881

	/*
L
Li Zefan 已提交
882 883
	 * we hit a real extent, if it is big or the next extent is not a
	 * real extent, don't bother defragging it
884
	 */
885
	if (!compress && (*last_len == 0 || *last_len >= thresh) &&
L
Li Zefan 已提交
886
	    (em->len >= thresh || !next_mergeable))
887
		ret = 0;
888
out:
889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908
	/*
	 * 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
909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924
/*
 * 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
925
{
C
Chris Mason 已提交
926 927 928 929
	unsigned long file_end;
	u64 isize = i_size_read(inode);
	u64 page_start;
	u64 page_end;
930
	u64 page_cnt;
C
Chris Mason 已提交
931 932 933
	int ret;
	int i;
	int i_done;
934
	struct btrfs_ordered_extent *ordered;
C
Chris Mason 已提交
935
	struct extent_state *cached_state = NULL;
936
	struct extent_io_tree *tree;
937
	gfp_t mask = btrfs_alloc_write_mask(inode->i_mapping);
C
Chris Mason 已提交
938 939

	file_end = (isize - 1) >> PAGE_CACHE_SHIFT;
940 941 942 943
	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
944 945

	ret = btrfs_delalloc_reserve_space(inode,
946
					   page_cnt << PAGE_CACHE_SHIFT);
C
Chris Mason 已提交
947 948 949
	if (ret)
		return ret;
	i_done = 0;
950
	tree = &BTRFS_I(inode)->io_tree;
C
Chris Mason 已提交
951 952

	/* step one, lock all the pages */
953
	for (i = 0; i < page_cnt; i++) {
C
Chris Mason 已提交
954
		struct page *page;
955
again:
956
		page = find_or_create_page(inode->i_mapping,
957
					   start_index + i, mask);
C
Chris Mason 已提交
958 959 960
		if (!page)
			break;

961 962 963
		page_start = page_offset(page);
		page_end = page_start + PAGE_CACHE_SIZE - 1;
		while (1) {
964
			lock_extent(tree, page_start, page_end);
965 966
			ordered = btrfs_lookup_ordered_extent(inode,
							      page_start);
967
			unlock_extent(tree, page_start, page_end);
968 969 970 971 972 973 974
			if (!ordered)
				break;

			unlock_page(page);
			btrfs_start_ordered_extent(inode, ordered, 1);
			btrfs_put_ordered_extent(ordered);
			lock_page(page);
975 976 977 978 979 980 981 982 983
			/*
			 * 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;
			}
984 985
		}

C
Chris Mason 已提交
986 987 988 989 990 991 992 993 994 995
		if (!PageUptodate(page)) {
			btrfs_readpage(NULL, page);
			lock_page(page);
			if (!PageUptodate(page)) {
				unlock_page(page);
				page_cache_release(page);
				ret = -EIO;
				break;
			}
		}
996 997 998 999 1000 1001 1002

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

C
Chris Mason 已提交
1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022
		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,
1023
			 page_start, page_end - 1, 0, &cached_state);
C
Chris Mason 已提交
1024 1025
	clear_extent_bit(&BTRFS_I(inode)->io_tree, page_start,
			  page_end - 1, EXTENT_DIRTY | EXTENT_DELALLOC |
1026 1027
			  EXTENT_DO_ACCOUNTING | EXTENT_DEFRAG, 0, 0,
			  &cached_state, GFP_NOFS);
C
Chris Mason 已提交
1028

1029
	if (i_done != page_cnt) {
1030 1031 1032
		spin_lock(&BTRFS_I(inode)->lock);
		BTRFS_I(inode)->outstanding_extents++;
		spin_unlock(&BTRFS_I(inode)->lock);
C
Chris Mason 已提交
1033
		btrfs_delalloc_release_space(inode,
1034
				     (page_cnt - i_done) << PAGE_CACHE_SHIFT);
C
Chris Mason 已提交
1035 1036 1037
	}


1038 1039
	set_extent_defrag(&BTRFS_I(inode)->io_tree, page_start, page_end - 1,
			  &cached_state, GFP_NOFS);
C
Chris Mason 已提交
1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058

	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]);
	}
1059
	btrfs_delalloc_release_space(inode, page_cnt << PAGE_CACHE_SHIFT);
C
Chris Mason 已提交
1060 1061 1062 1063 1064 1065 1066 1067 1068 1069
	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;
	struct file_ra_state *ra = NULL;
C
Christoph Hellwig 已提交
1070
	unsigned long last_index;
1071
	u64 isize = i_size_read(inode);
1072 1073 1074
	u64 last_len = 0;
	u64 skip = 0;
	u64 defrag_end = 0;
C
Chris Mason 已提交
1075
	u64 newer_off = range->start;
C
Christoph Hellwig 已提交
1076
	unsigned long i;
1077
	unsigned long ra_index = 0;
C
Christoph Hellwig 已提交
1078
	int ret;
C
Chris Mason 已提交
1079
	int defrag_count = 0;
1080
	int compress_type = BTRFS_COMPRESS_ZLIB;
C
Chris Mason 已提交
1081
	int extent_thresh = range->extent_thresh;
1082 1083
	int max_cluster = (256 * 1024) >> PAGE_CACHE_SHIFT;
	int cluster = max_cluster;
C
Chris Mason 已提交
1084 1085 1086 1087 1088
	u64 new_align = ~((u64)128 * 1024 - 1);
	struct page **pages = NULL;

	if (extent_thresh == 0)
		extent_thresh = 256 * 1024;
1089 1090 1091 1092 1093 1094 1095

	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
1096

1097
	if (isize == 0)
1098 1099
		return 0;

C
Chris Mason 已提交
1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112
	/*
	 * 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;
	}

1113
	pages = kmalloc(sizeof(struct page *) * max_cluster,
C
Chris Mason 已提交
1114 1115 1116 1117 1118 1119 1120
			GFP_NOFS);
	if (!pages) {
		ret = -ENOMEM;
		goto out_ra;
	}

	/* find the last page to defrag */
C
Chris Mason 已提交
1121
	if (range->start + range->len > range->start) {
1122
		last_index = min_t(u64, isize - 1,
C
Chris Mason 已提交
1123 1124
			 range->start + range->len - 1) >> PAGE_CACHE_SHIFT;
	} else {
1125
		last_index = (isize - 1) >> PAGE_CACHE_SHIFT;
C
Chris Mason 已提交
1126 1127
	}

C
Chris Mason 已提交
1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143
	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
1144
		max_to_defrag = last_index + 1;
C
Chris Mason 已提交
1145

L
Li Zefan 已提交
1146 1147 1148 1149 1150 1151 1152
	/*
	 * 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;

1153 1154 1155
	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
1156 1157 1158 1159 1160 1161 1162
		/*
		 * make sure we stop running if someone unmounts
		 * the FS
		 */
		if (!(inode->i_sb->s_flags & MS_ACTIVE))
			break;

1163
		if (!should_defrag_range(inode, (u64)i << PAGE_CACHE_SHIFT,
L
Li Zefan 已提交
1164
					 extent_thresh, &last_len, &skip,
1165 1166
					 &defrag_end, range->flags &
					 BTRFS_DEFRAG_RANGE_COMPRESS)) {
1167 1168 1169 1170 1171 1172 1173 1174 1175
			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;
		}
1176 1177 1178 1179 1180 1181 1182 1183 1184

		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
1185
		if (range->flags & BTRFS_DEFRAG_RANGE_COMPRESS)
1186
			BTRFS_I(inode)->force_compress = compress_type;
1187

1188 1189 1190 1191 1192 1193
		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;
		}
1194

1195
		mutex_lock(&inode->i_mutex);
1196
		ret = cluster_pages_for_defrag(inode, pages, i, cluster);
1197 1198
		if (ret < 0) {
			mutex_unlock(&inode->i_mutex);
C
Chris Mason 已提交
1199
			goto out_ra;
1200
		}
C
Chris Mason 已提交
1201 1202 1203

		defrag_count += ret;
		balance_dirty_pages_ratelimited_nr(inode->i_mapping, ret);
1204
		mutex_unlock(&inode->i_mutex);
C
Chris Mason 已提交
1205 1206 1207 1208 1209

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

1210 1211 1212
			if (ret > 0)
				i += ret;

C
Chris Mason 已提交
1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223
			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
1224
			}
C
Chris Mason 已提交
1225
		} else {
1226
			if (ret > 0) {
L
Li Zefan 已提交
1227
				i += ret;
1228 1229
				last_len += ret << PAGE_CACHE_SHIFT;
			} else {
L
Li Zefan 已提交
1230
				i++;
1231 1232
				last_len = 0;
			}
C
Christoph Hellwig 已提交
1233 1234 1235
		}
	}

C
Chris Mason 已提交
1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253
	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);
1254
		BTRFS_I(inode)->force_compress = BTRFS_COMPRESS_NONE;
C
Chris Mason 已提交
1255 1256 1257
		mutex_unlock(&inode->i_mutex);
	}

1258
	if (range->compress_type == BTRFS_COMPRESS_LZO) {
1259
		btrfs_set_fs_incompat(root->fs_info, COMPRESS_LZO);
1260 1261
	}

1262
	ret = defrag_count;
1263

C
Chris Mason 已提交
1264 1265 1266 1267
out_ra:
	if (!file)
		kfree(ra);
	kfree(pages);
1268
	return ret;
C
Christoph Hellwig 已提交
1269 1270
}

1271 1272
static noinline int btrfs_ioctl_resize(struct btrfs_root *root,
					void __user *arg)
C
Christoph Hellwig 已提交
1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284
{
	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
1285 1286 1287
	if (root->fs_info->sb->s_flags & MS_RDONLY)
		return -EROFS;

1288 1289 1290
	if (!capable(CAP_SYS_ADMIN))
		return -EPERM;

1291 1292 1293 1294 1295 1296 1297
	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
1298
	vol_args = memdup_user(arg, sizeof(*vol_args));
1299 1300 1301 1302
	if (IS_ERR(vol_args)) {
		ret = PTR_ERR(vol_args);
		goto out;
	}
1303 1304

	vol_args->name[BTRFS_PATH_NAME_MAX] = '\0';
C
Christoph Hellwig 已提交
1305 1306 1307 1308 1309 1310 1311 1312 1313

	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);
1314
		printk(KERN_INFO "btrfs: resizing devid %llu\n",
1315
		       (unsigned long long)devid);
C
Christoph Hellwig 已提交
1316
	}
Y
Yan Zheng 已提交
1317
	device = btrfs_find_device(root, devid, NULL, NULL);
C
Christoph Hellwig 已提交
1318
	if (!device) {
1319
		printk(KERN_INFO "btrfs: resizer unable to find device %llu\n",
1320
		       (unsigned long long)devid);
C
Christoph Hellwig 已提交
1321
		ret = -EINVAL;
1322
		goto out_free;
C
Christoph Hellwig 已提交
1323
	}
L
Liu Bo 已提交
1324 1325
	if (device->fs_devices && device->fs_devices->seeding) {
		printk(KERN_INFO "btrfs: resizer unable to apply on "
1326 1327
		       "seeding device %llu\n",
		       (unsigned long long)devid);
L
Liu Bo 已提交
1328 1329 1330 1331
		ret = -EINVAL;
		goto out_free;
	}

C
Christoph Hellwig 已提交
1332 1333 1334 1335 1336 1337 1338 1339 1340 1341
	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
1342
		new_size = memparse(sizestr, NULL);
C
Christoph Hellwig 已提交
1343 1344
		if (new_size == 0) {
			ret = -EINVAL;
1345
			goto out_free;
C
Christoph Hellwig 已提交
1346 1347 1348 1349 1350 1351 1352 1353
		}
	}

	old_size = device->total_bytes;

	if (mod < 0) {
		if (new_size > old_size) {
			ret = -EINVAL;
1354
			goto out_free;
C
Christoph Hellwig 已提交
1355 1356 1357 1358 1359 1360 1361 1362
		}
		new_size = old_size - new_size;
	} else if (mod > 0) {
		new_size = old_size + new_size;
	}

	if (new_size < 256 * 1024 * 1024) {
		ret = -EINVAL;
1363
		goto out_free;
C
Christoph Hellwig 已提交
1364 1365 1366
	}
	if (new_size > device->bdev->bd_inode->i_size) {
		ret = -EFBIG;
1367
		goto out_free;
C
Christoph Hellwig 已提交
1368 1369 1370 1371 1372
	}

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

1373 1374 1375
	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
1376 1377

	if (new_size > old_size) {
1378
		trans = btrfs_start_transaction(root, 0);
1379 1380
		if (IS_ERR(trans)) {
			ret = PTR_ERR(trans);
1381
			goto out_free;
1382
		}
C
Christoph Hellwig 已提交
1383 1384
		ret = btrfs_grow_device(trans, device, new_size);
		btrfs_commit_transaction(trans, root);
1385
	} else if (new_size < old_size) {
C
Christoph Hellwig 已提交
1386 1387 1388
		ret = btrfs_shrink_device(device, new_size);
	}

1389
out_free:
C
Christoph Hellwig 已提交
1390
	kfree(vol_args);
1391 1392
out:
	mutex_unlock(&root->fs_info->volume_mutex);
C
Christoph Hellwig 已提交
1393 1394 1395
	return ret;
}

S
Sage Weil 已提交
1396
static noinline int btrfs_ioctl_snap_create_transid(struct file *file,
A
Arne Jansen 已提交
1397 1398 1399
				char *name, unsigned long fd, int subvol,
				u64 *transid, bool readonly,
				struct btrfs_qgroup_inherit **inherit)
C
Christoph Hellwig 已提交
1400
{
1401
	struct file *src_file;
C
Christoph Hellwig 已提交
1402
	int namelen;
1403
	int ret = 0;
C
Christoph Hellwig 已提交
1404

1405 1406 1407 1408
	ret = mnt_want_write_file(file);
	if (ret)
		goto out;

S
Sage Weil 已提交
1409 1410
	namelen = strlen(name);
	if (strchr(name, '/')) {
C
Christoph Hellwig 已提交
1411
		ret = -EINVAL;
1412
		goto out_drop_write;
C
Christoph Hellwig 已提交
1413 1414
	}

1415 1416 1417
	if (name[0] == '.' &&
	   (namelen == 1 || (name[1] == '.' && namelen == 2))) {
		ret = -EEXIST;
1418
		goto out_drop_write;
1419 1420
	}

1421
	if (subvol) {
S
Sage Weil 已提交
1422
		ret = btrfs_mksubvol(&file->f_path, name, namelen,
A
Arne Jansen 已提交
1423
				     NULL, transid, readonly, inherit);
1424
	} else {
1425
		struct inode *src_inode;
S
Sage Weil 已提交
1426
		src_file = fget(fd);
1427 1428
		if (!src_file) {
			ret = -EINVAL;
1429
			goto out_drop_write;
1430 1431 1432 1433
		}

		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
1434 1435
			printk(KERN_INFO "btrfs: Snapshot src from "
			       "another FS\n");
1436 1437
			ret = -EINVAL;
			fput(src_file);
1438
			goto out_drop_write;
1439
		}
S
Sage Weil 已提交
1440 1441
		ret = btrfs_mksubvol(&file->f_path, name, namelen,
				     BTRFS_I(src_inode)->root,
A
Arne Jansen 已提交
1442
				     transid, readonly, inherit);
1443
		fput(src_file);
1444
	}
1445 1446
out_drop_write:
	mnt_drop_write_file(file);
C
Christoph Hellwig 已提交
1447
out:
S
Sage Weil 已提交
1448 1449 1450 1451
	return ret;
}

static noinline int btrfs_ioctl_snap_create(struct file *file,
1452
					    void __user *arg, int subvol)
S
Sage Weil 已提交
1453
{
1454
	struct btrfs_ioctl_vol_args *vol_args;
S
Sage Weil 已提交
1455 1456
	int ret;

1457 1458 1459 1460
	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
1461

1462
	ret = btrfs_ioctl_snap_create_transid(file, vol_args->name,
L
Li Zefan 已提交
1463
					      vol_args->fd, subvol,
A
Arne Jansen 已提交
1464
					      NULL, false, NULL);
1465

1466 1467 1468
	kfree(vol_args);
	return ret;
}
1469

1470 1471 1472 1473 1474 1475 1476
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
1477
	bool readonly = false;
A
Arne Jansen 已提交
1478
	struct btrfs_qgroup_inherit *inherit = NULL;
1479

1480 1481 1482 1483
	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';
1484

L
Li Zefan 已提交
1485
	if (vol_args->flags &
A
Arne Jansen 已提交
1486 1487
	    ~(BTRFS_SUBVOL_CREATE_ASYNC | BTRFS_SUBVOL_RDONLY |
	      BTRFS_SUBVOL_QGROUP_INHERIT)) {
L
Li Zefan 已提交
1488
		ret = -EOPNOTSUPP;
1489
		goto out;
S
Sage Weil 已提交
1490
	}
1491 1492 1493

	if (vol_args->flags & BTRFS_SUBVOL_CREATE_ASYNC)
		ptr = &transid;
L
Li Zefan 已提交
1494 1495
	if (vol_args->flags & BTRFS_SUBVOL_RDONLY)
		readonly = true;
A
Arne Jansen 已提交
1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506
	if (vol_args->flags & BTRFS_SUBVOL_QGROUP_INHERIT) {
		if (vol_args->size > PAGE_CACHE_SIZE) {
			ret = -EINVAL;
			goto out;
		}
		inherit = memdup_user(vol_args->qgroup_inherit, vol_args->size);
		if (IS_ERR(inherit)) {
			ret = PTR_ERR(inherit);
			goto out;
		}
	}
1507 1508

	ret = btrfs_ioctl_snap_create_transid(file, vol_args->name,
A
Arne Jansen 已提交
1509 1510
					      vol_args->fd, subvol, ptr,
					      readonly, &inherit);
1511 1512 1513 1514 1515 1516

	if (ret == 0 && ptr &&
	    copy_to_user(arg +
			 offsetof(struct btrfs_ioctl_vol_args_v2,
				  transid), ptr, sizeof(*ptr)))
		ret = -EFAULT;
1517
out:
C
Christoph Hellwig 已提交
1518
	kfree(vol_args);
A
Arne Jansen 已提交
1519
	kfree(inherit);
C
Christoph Hellwig 已提交
1520 1521 1522
	return ret;
}

1523 1524 1525 1526 1527 1528 1529 1530
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
1531
	if (btrfs_ino(inode) != BTRFS_FIRST_FREE_OBJECTID)
1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554
		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;

1555 1556 1557
	ret = mnt_want_write_file(file);
	if (ret)
		goto out;
1558

1559 1560 1561 1562
	if (btrfs_ino(inode) != BTRFS_FIRST_FREE_OBJECTID) {
		ret = -EINVAL;
		goto out_drop_write;
	}
1563

1564 1565 1566 1567
	if (copy_from_user(&flags, arg, sizeof(flags))) {
		ret = -EFAULT;
		goto out_drop_write;
	}
1568

1569 1570 1571 1572
	if (flags & BTRFS_SUBVOL_CREATE_ASYNC) {
		ret = -EINVAL;
		goto out_drop_write;
	}
1573

1574 1575 1576 1577
	if (flags & ~BTRFS_SUBVOL_RDONLY) {
		ret = -EOPNOTSUPP;
		goto out_drop_write;
	}
1578

1579 1580 1581 1582
	if (!inode_owner_or_capable(inode)) {
		ret = -EACCES;
		goto out_drop_write;
	}
1583

1584 1585 1586 1587
	down_write(&root->fs_info->subvol_sem);

	/* nothing to do */
	if (!!(flags & BTRFS_SUBVOL_RDONLY) == btrfs_root_readonly(root))
1588
		goto out_drop_sem;
1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603

	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;
	}

1604
	ret = btrfs_update_root(trans, root->fs_info->tree_root,
1605 1606 1607 1608 1609 1610
				&root->root_key, &root->root_item);

	btrfs_commit_transaction(trans, root);
out_reset:
	if (ret)
		btrfs_set_root_flags(&root->root_item, root_flags);
1611
out_drop_sem:
1612
	up_write(&root->fs_info->subvol_sem);
1613 1614 1615
out_drop_write:
	mnt_drop_write_file(file);
out:
1616 1617 1618
	return ret;
}

1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654
/*
 * 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;
}

1655 1656 1657
static noinline int key_in_sk(struct btrfs_key *key,
			      struct btrfs_ioctl_search_key *sk)
{
1658 1659 1660 1661 1662 1663 1664 1665 1666
	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)
1667
		return 0;
1668 1669 1670 1671 1672 1673 1674

	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)
1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 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
		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;
		}
1741
		(*num_found)++;
1742 1743 1744 1745 1746

		if (*num_found >= sk->nr_items)
			break;
	}
advance_key:
1747 1748
	ret = 0;
	if (key->offset < (u64)-1 && key->offset < sk->max_offset)
1749
		key->offset++;
1750 1751
	else if (key->type < (u8)-1 && key->type < sk->max_type) {
		key->offset = 0;
1752
		key->type++;
1753 1754 1755
	} else if (key->objectid < (u64)-1 && key->objectid < sk->max_objectid) {
		key->offset = 0;
		key->type = 0;
1756
		key->objectid++;
1757 1758
	} else
		ret = 1;
1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815
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);
1816
		btrfs_release_path(path);
1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837
		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
1838 1839 1840
	args = memdup_user(argp, sizeof(*args));
	if (IS_ERR(args))
		return PTR_ERR(args);
1841 1842 1843 1844 1845 1846 1847 1848 1849

	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;
}

1850
/*
1851 1852 1853
 * Search INODE_REFs to identify path name of 'dirid' directory
 * in a 'tree_id' tree. and sets path name to 'name'.
 */
1854 1855 1856 1857 1858
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;
1859
	char *ptr;
1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876
	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;

1877
	ptr = &name[BTRFS_INO_LOOKUP_PATH_MAX];
1878 1879 1880 1881 1882 1883 1884

	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);
1885 1886
		ret = -ENOENT;
		goto out;
1887 1888 1889 1890
	}

	key.objectid = dirid;
	key.type = BTRFS_INODE_REF_KEY;
1891
	key.offset = (u64)-1;
1892 1893 1894 1895 1896 1897 1898 1899

	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];
1900 1901
		if (ret > 0 && slot > 0)
			slot--;
1902 1903 1904
		btrfs_item_key_to_cpu(l, &key, slot);

		if (ret > 0 && (key.objectid != dirid ||
1905 1906
				key.type != BTRFS_INODE_REF_KEY)) {
			ret = -ENOENT;
1907
			goto out;
1908
		}
1909 1910 1911 1912 1913

		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;
1914
		if (ptr < name)
1915 1916 1917 1918 1919 1920 1921 1922
			goto out;

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

		if (key.offset == BTRFS_FIRST_FREE_OBJECTID)
			break;

1923
		btrfs_release_path(path);
1924
		key.objectid = key.offset;
1925
		key.offset = (u64)-1;
1926 1927
		dirid = key.objectid;
	}
1928
	if (ptr < name)
1929
		goto out;
1930
	memmove(name, ptr, total_len);
1931 1932 1933 1934
	name[total_len]='\0';
	ret = 0;
out:
	btrfs_free_path(path);
1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947
	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
1948 1949 1950
	args = memdup_user(argp, sizeof(*args));
	if (IS_ERR(args))
		return PTR_ERR(args);
1951

1952 1953
	inode = fdentry(file)->d_inode;

1954 1955 1956
	if (args->treeid == 0)
		args->treeid = BTRFS_I(inode)->root->root_key.objectid;

1957 1958 1959 1960 1961 1962 1963 1964
	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);
1965 1966 1967
	return ret;
}

1968 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
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;
	}

1995
	err = mnt_want_write_file(file);
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
	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;
2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051
	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
2052
	if (btrfs_ino(inode) != BTRFS_FIRST_FREE_OBJECTID) {
2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066 2067
		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;

2068 2069 2070
	trans = btrfs_start_transaction(root, 0);
	if (IS_ERR(trans)) {
		err = PTR_ERR(trans);
2071
		goto out_up_write;
2072 2073 2074
	}
	trans->block_rsv = &root->fs_info->global_block_rsv;

2075 2076 2077 2078
	ret = btrfs_unlink_subvol(trans, root, dir,
				dest->root_key.objectid,
				dentry->d_name.name,
				dentry->d_name.len);
2079 2080 2081 2082 2083
	if (ret) {
		err = ret;
		btrfs_abort_transaction(trans, root, ret);
		goto out_end_trans;
	}
2084 2085 2086 2087 2088 2089 2090 2091

	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);

2092 2093 2094 2095
	if (!xchg(&dest->orphan_item_inserted, 1)) {
		ret = btrfs_insert_orphan_item(trans,
					root->fs_info->tree_root,
					dest->root_key.objectid);
2096 2097 2098 2099 2100
		if (ret) {
			btrfs_abort_transaction(trans, root, ret);
			err = ret;
			goto out_end_trans;
		}
2101
	}
2102
out_end_trans:
S
Sage Weil 已提交
2103
	ret = btrfs_end_transaction(trans, root);
2104 2105
	if (ret && !err)
		err = ret;
2106 2107 2108 2109 2110 2111
	inode->i_flags |= S_DEAD;
out_up_write:
	up_write(&root->fs_info->subvol_sem);
out_unlock:
	mutex_unlock(&inode->i_mutex);
	if (!err) {
2112
		shrink_dcache_sb(root->fs_info->sb);
2113 2114 2115 2116 2117 2118 2119
		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
2120
	mnt_drop_write_file(file);
2121 2122 2123 2124 2125
out:
	kfree(vol_args);
	return err;
}

C
Chris Mason 已提交
2126
static int btrfs_ioctl_defrag(struct file *file, void __user *argp)
C
Christoph Hellwig 已提交
2127 2128 2129
{
	struct inode *inode = fdentry(file)->d_inode;
	struct btrfs_root *root = BTRFS_I(inode)->root;
C
Chris Mason 已提交
2130
	struct btrfs_ioctl_defrag_range_args *range;
Y
Yan Zheng 已提交
2131 2132
	int ret;

L
Li Zefan 已提交
2133 2134 2135
	if (btrfs_root_readonly(root))
		return -EROFS;

2136
	ret = mnt_want_write_file(file);
Y
Yan Zheng 已提交
2137 2138
	if (ret)
		return ret;
C
Christoph Hellwig 已提交
2139 2140 2141

	switch (inode->i_mode & S_IFMT) {
	case S_IFDIR:
2142 2143 2144 2145
		if (!capable(CAP_SYS_ADMIN)) {
			ret = -EPERM;
			goto out;
		}
2146 2147 2148 2149
		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
2150 2151
		break;
	case S_IFREG:
2152 2153 2154 2155
		if (!(file->f_mode & FMODE_WRITE)) {
			ret = -EINVAL;
			goto out;
		}
C
Chris Mason 已提交
2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167

		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);
2168
				goto out;
C
Chris Mason 已提交
2169 2170 2171 2172 2173 2174 2175 2176 2177 2178
			}
			/* 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
2179 2180 2181 2182
		ret = btrfs_defrag_file(fdentry(file)->d_inode, file,
					range, 0, 0);
		if (ret > 0)
			ret = 0;
C
Chris Mason 已提交
2183
		kfree(range);
C
Christoph Hellwig 已提交
2184
		break;
2185 2186
	default:
		ret = -EINVAL;
C
Christoph Hellwig 已提交
2187
	}
2188
out:
A
Al Viro 已提交
2189
	mnt_drop_write_file(file);
2190
	return ret;
C
Christoph Hellwig 已提交
2191 2192
}

2193
static long btrfs_ioctl_add_dev(struct btrfs_root *root, void __user *arg)
C
Christoph Hellwig 已提交
2194 2195 2196 2197
{
	struct btrfs_ioctl_vol_args *vol_args;
	int ret;

2198 2199 2200
	if (!capable(CAP_SYS_ADMIN))
		return -EPERM;

2201 2202 2203 2204 2205 2206 2207
	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
2208
	vol_args = memdup_user(arg, sizeof(*vol_args));
2209 2210 2211 2212
	if (IS_ERR(vol_args)) {
		ret = PTR_ERR(vol_args);
		goto out;
	}
C
Christoph Hellwig 已提交
2213

2214
	vol_args->name[BTRFS_PATH_NAME_MAX] = '\0';
C
Christoph Hellwig 已提交
2215 2216 2217
	ret = btrfs_init_new_device(root, vol_args->name);

	kfree(vol_args);
2218 2219
out:
	mutex_unlock(&root->fs_info->volume_mutex);
C
Christoph Hellwig 已提交
2220 2221 2222
	return ret;
}

2223
static long btrfs_ioctl_rm_dev(struct btrfs_root *root, void __user *arg)
C
Christoph Hellwig 已提交
2224 2225 2226 2227
{
	struct btrfs_ioctl_vol_args *vol_args;
	int ret;

2228 2229 2230
	if (!capable(CAP_SYS_ADMIN))
		return -EPERM;

Y
Yan Zheng 已提交
2231 2232 2233
	if (root->fs_info->sb->s_flags & MS_RDONLY)
		return -EROFS;

2234 2235 2236 2237 2238 2239 2240
	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
2241
	vol_args = memdup_user(arg, sizeof(*vol_args));
2242 2243 2244 2245
	if (IS_ERR(vol_args)) {
		ret = PTR_ERR(vol_args);
		goto out;
	}
C
Christoph Hellwig 已提交
2246

2247
	vol_args->name[BTRFS_PATH_NAME_MAX] = '\0';
C
Christoph Hellwig 已提交
2248 2249 2250
	ret = btrfs_rm_device(root, vol_args->name);

	kfree(vol_args);
2251 2252
out:
	mutex_unlock(&root->fs_info->volume_mutex);
C
Christoph Hellwig 已提交
2253 2254 2255
	return ret;
}

J
Jan Schmidt 已提交
2256 2257
static long btrfs_ioctl_fs_info(struct btrfs_root *root, void __user *arg)
{
2258
	struct btrfs_ioctl_fs_info_args *fi_args;
J
Jan Schmidt 已提交
2259 2260 2261
	struct btrfs_device *device;
	struct btrfs_device *next;
	struct btrfs_fs_devices *fs_devices = root->fs_info->fs_devices;
2262
	int ret = 0;
J
Jan Schmidt 已提交
2263 2264 2265 2266

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

2267 2268 2269 2270 2271 2272
	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
2273 2274 2275

	mutex_lock(&fs_devices->device_list_mutex);
	list_for_each_entry_safe(device, next, &fs_devices->devices, dev_list) {
2276 2277
		if (device->devid > fi_args->max_id)
			fi_args->max_id = device->devid;
J
Jan Schmidt 已提交
2278 2279 2280
	}
	mutex_unlock(&fs_devices->device_list_mutex);

2281 2282
	if (copy_to_user(arg, fi_args, sizeof(*fi_args)))
		ret = -EFAULT;
J
Jan Schmidt 已提交
2283

2284 2285
	kfree(fi_args);
	return ret;
J
Jan Schmidt 已提交
2286 2287 2288 2289 2290 2291 2292 2293 2294 2295 2296 2297 2298 2299 2300 2301 2302 2303 2304 2305 2306 2307 2308 2309 2310 2311 2312 2313 2314 2315 2316 2317 2318 2319
}

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));
2320
	if (dev->name) {
2321 2322 2323 2324 2325 2326
		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();
2327 2328
		di_args->path[sizeof(di_args->path) - 1] = 0;
	} else {
2329
		di_args->path[0] = '\0';
2330
	}
J
Jan Schmidt 已提交
2331 2332 2333 2334 2335 2336 2337 2338 2339

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

	kfree(di_args);
	return ret;
}

2340 2341
static noinline long btrfs_ioctl_clone(struct file *file, unsigned long srcfd,
				       u64 off, u64 olen, u64 destoff)
C
Christoph Hellwig 已提交
2342 2343 2344 2345 2346 2347 2348 2349
{
	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
2350 2351
	char *buf;
	struct btrfs_key key;
C
Christoph Hellwig 已提交
2352 2353
	u32 nritems;
	int slot;
Y
Yan Zheng 已提交
2354
	int ret;
2355 2356
	u64 len = olen;
	u64 bs = root->fs_info->sb->s_blocksize;
2357

2358 2359 2360 2361 2362 2363 2364 2365 2366 2367
	/*
	 * 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)?
	 */

2368
	/* the destination must be opened for writing */
2369
	if (!(file->f_mode & FMODE_WRITE) || (file->f_flags & O_APPEND))
2370 2371
		return -EINVAL;

L
Li Zefan 已提交
2372 2373 2374
	if (btrfs_root_readonly(root))
		return -EROFS;

2375
	ret = mnt_want_write_file(file);
Y
Yan Zheng 已提交
2376 2377 2378
	if (ret)
		return ret;

2379
	src_file = fget(srcfd);
2380 2381 2382 2383
	if (!src_file) {
		ret = -EBADF;
		goto out_drop_write;
	}
2384

2385 2386 2387 2388
	ret = -EXDEV;
	if (src_file->f_path.mnt != file->f_path.mnt)
		goto out_fput;

C
Christoph Hellwig 已提交
2389 2390
	src = src_file->f_dentry->d_inode;

2391 2392 2393 2394
	ret = -EINVAL;
	if (src == inode)
		goto out_fput;

2395 2396 2397 2398
	/* the src must be open for reading */
	if (!(src_file->f_mode & FMODE_READ))
		goto out_fput;

2399 2400 2401 2402 2403
	/* 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
2404 2405 2406 2407
	ret = -EISDIR;
	if (S_ISDIR(src->i_mode) || S_ISDIR(inode->i_mode))
		goto out_fput;

C
Christoph Hellwig 已提交
2408
	ret = -EXDEV;
2409
	if (src->i_sb != inode->i_sb)
Y
Yan Zheng 已提交
2410 2411 2412 2413 2414 2415 2416 2417 2418 2419
		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
2420
		goto out_fput;
Y
Yan Zheng 已提交
2421 2422
	}
	path->reada = 2;
C
Christoph Hellwig 已提交
2423 2424

	if (inode < src) {
2425 2426
		mutex_lock_nested(&inode->i_mutex, I_MUTEX_PARENT);
		mutex_lock_nested(&src->i_mutex, I_MUTEX_CHILD);
C
Christoph Hellwig 已提交
2427
	} else {
2428 2429
		mutex_lock_nested(&src->i_mutex, I_MUTEX_PARENT);
		mutex_lock_nested(&inode->i_mutex, I_MUTEX_CHILD);
C
Christoph Hellwig 已提交
2430 2431
	}

2432 2433
	/* determine range to clone */
	ret = -EINVAL;
2434
	if (off + len > src->i_size || off + len < off)
C
Christoph Hellwig 已提交
2435
		goto out_unlock;
2436 2437 2438 2439
	if (len == 0)
		olen = len = src->i_size - off;
	/* if we extend to eof, continue to block boundary */
	if (off + len == src->i_size)
2440
		len = ALIGN(src->i_size, bs) - off;
2441 2442

	/* verify the end result is block aligned */
2443 2444
	if (!IS_ALIGNED(off, bs) || !IS_ALIGNED(off + len, bs) ||
	    !IS_ALIGNED(destoff, bs))
2445 2446
		goto out_unlock;

2447 2448 2449 2450 2451 2452
	if (destoff > inode->i_size) {
		ret = btrfs_cont_expand(inode, inode->i_size, destoff);
		if (ret)
			goto out_unlock;
	}

2453 2454 2455 2456
	/* 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
2457 2458 2459
	/* do any pending delalloc/csum calc on src, one way or
	   another, and lock file content */
	while (1) {
Z
Zheng Yan 已提交
2460
		struct btrfs_ordered_extent *ordered;
2461
		lock_extent(&BTRFS_I(src)->io_tree, off, off+len);
2462 2463 2464 2465
		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
2466
			break;
2467
		unlock_extent(&BTRFS_I(src)->io_tree, off, off+len);
Y
Yan Zheng 已提交
2468 2469
		if (ordered)
			btrfs_put_ordered_extent(ordered);
2470
		btrfs_wait_ordered_range(src, off, len);
C
Christoph Hellwig 已提交
2471 2472
	}

2473
	/* clone data */
L
Li Zefan 已提交
2474
	key.objectid = btrfs_ino(src);
Y
Yan Zheng 已提交
2475 2476
	key.type = BTRFS_EXTENT_DATA_KEY;
	key.offset = 0;
C
Christoph Hellwig 已提交
2477 2478 2479 2480 2481 2482

	while (1) {
		/*
		 * note the key will change type as we walk through the
		 * tree.
		 */
2483 2484
		ret = btrfs_search_slot(NULL, BTRFS_I(src)->root, &key, path,
				0, 0);
C
Christoph Hellwig 已提交
2485 2486 2487
		if (ret < 0)
			goto out;

Y
Yan Zheng 已提交
2488 2489
		nritems = btrfs_header_nritems(path->nodes[0]);
		if (path->slots[0] >= nritems) {
2490
			ret = btrfs_next_leaf(BTRFS_I(src)->root, path);
C
Christoph Hellwig 已提交
2491 2492 2493 2494
			if (ret < 0)
				goto out;
			if (ret > 0)
				break;
Y
Yan Zheng 已提交
2495
			nritems = btrfs_header_nritems(path->nodes[0]);
C
Christoph Hellwig 已提交
2496 2497 2498 2499
		}
		leaf = path->nodes[0];
		slot = path->slots[0];

Y
Yan Zheng 已提交
2500
		btrfs_item_key_to_cpu(leaf, &key, slot);
2501
		if (btrfs_key_type(&key) > BTRFS_EXTENT_DATA_KEY ||
L
Li Zefan 已提交
2502
		    key.objectid != btrfs_ino(src))
C
Christoph Hellwig 已提交
2503 2504
			break;

2505 2506 2507
		if (btrfs_key_type(&key) == BTRFS_EXTENT_DATA_KEY) {
			struct btrfs_file_extent_item *extent;
			int type;
Z
Zheng Yan 已提交
2508 2509
			u32 size;
			struct btrfs_key new_key;
2510 2511 2512
			u64 disko = 0, diskl = 0;
			u64 datao = 0, datal = 0;
			u8 comp;
2513
			u64 endoff;
Z
Zheng Yan 已提交
2514 2515 2516 2517 2518

			size = btrfs_item_size_nr(leaf, slot);
			read_extent_buffer(leaf, buf,
					   btrfs_item_ptr_offset(leaf, slot),
					   size);
2519 2520 2521 2522 2523

			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);
2524 2525
			if (type == BTRFS_FILE_EXTENT_REG ||
			    type == BTRFS_FILE_EXTENT_PREALLOC) {
C
Chris Mason 已提交
2526 2527 2528 2529
				disko = btrfs_file_extent_disk_bytenr(leaf,
								      extent);
				diskl = btrfs_file_extent_disk_num_bytes(leaf,
								 extent);
2530
				datao = btrfs_file_extent_offset(leaf, extent);
C
Chris Mason 已提交
2531 2532
				datal = btrfs_file_extent_num_bytes(leaf,
								    extent);
2533 2534 2535 2536 2537
			} else if (type == BTRFS_FILE_EXTENT_INLINE) {
				/* take upper bound, may be compressed */
				datal = btrfs_file_extent_ram_bytes(leaf,
								    extent);
			}
2538
			btrfs_release_path(path);
Z
Zheng Yan 已提交
2539

2540
			if (key.offset + datal <= off ||
2541 2542 2543
			    key.offset >= off+len)
				goto next;

Z
Zheng Yan 已提交
2544
			memcpy(&new_key, &key, sizeof(new_key));
L
Li Zefan 已提交
2545
			new_key.objectid = btrfs_ino(inode);
2546 2547 2548 2549
			if (off <= key.offset)
				new_key.offset = key.offset + destoff - off;
			else
				new_key.offset = destoff;
Z
Zheng Yan 已提交
2550

2551 2552 2553 2554 2555 2556
			/*
			 * 1 - adjusting old extent (we may have to split it)
			 * 1 - add new extent
			 * 1 - inode update
			 */
			trans = btrfs_start_transaction(root, 3);
2557 2558 2559 2560 2561
			if (IS_ERR(trans)) {
				ret = PTR_ERR(trans);
				goto out;
			}

2562 2563
			if (type == BTRFS_FILE_EXTENT_REG ||
			    type == BTRFS_FILE_EXTENT_PREALLOC) {
2564 2565 2566 2567 2568 2569 2570 2571 2572 2573
				/*
				 *    a  | --- range to clone ---|  b
				 * | ------------- extent ------------- |
				 */

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

				/* substract range a */
2574 2575 2576 2577 2578
				if (off > key.offset) {
					datao += off - key.offset;
					datal -= off - key.offset;
				}

J
Josef Bacik 已提交
2579
				ret = btrfs_drop_extents(trans, root, inode,
2580 2581
							 new_key.offset,
							 new_key.offset + datal,
2582
							 1);
2583 2584 2585 2586 2587 2588
				if (ret) {
					btrfs_abort_transaction(trans, root,
								ret);
					btrfs_end_transaction(trans, root);
					goto out;
				}
2589

2590 2591
				ret = btrfs_insert_empty_item(trans, root, path,
							      &new_key, size);
2592 2593 2594 2595 2596 2597
				if (ret) {
					btrfs_abort_transaction(trans, root,
								ret);
					btrfs_end_transaction(trans, root);
					goto out;
				}
2598 2599 2600 2601

				leaf = path->nodes[0];
				slot = path->slots[0];
				write_extent_buffer(leaf, buf,
Z
Zheng Yan 已提交
2602 2603
					    btrfs_item_ptr_offset(leaf, slot),
					    size);
Y
Yan Zheng 已提交
2604

2605
				extent = btrfs_item_ptr(leaf, slot,
C
Christoph Hellwig 已提交
2606
						struct btrfs_file_extent_item);
2607 2608 2609 2610 2611 2612 2613 2614 2615 2616 2617

				/* 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
2618
					ret = btrfs_inc_extent_ref(trans, root,
2619 2620
							disko, diskl, 0,
							root->root_key.objectid,
L
Li Zefan 已提交
2621
							btrfs_ino(inode),
A
Arne Jansen 已提交
2622 2623
							new_key.offset - datao,
							0);
2624 2625 2626 2627 2628 2629 2630 2631 2632
					if (ret) {
						btrfs_abort_transaction(trans,
									root,
									ret);
						btrfs_end_transaction(trans,
								      root);
						goto out;

					}
C
Christoph Hellwig 已提交
2633
				}
2634 2635 2636 2637 2638 2639 2640
			} 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
2641

2642 2643
				if (key.offset + datal > off+len)
					trim = key.offset + datal - (off+len);
C
Chris Mason 已提交
2644

2645 2646
				if (comp && (skip || trim)) {
					ret = -EINVAL;
2647
					btrfs_end_transaction(trans, root);
2648 2649 2650 2651
					goto out;
				}
				size -= skip + trim;
				datal -= skip + trim;
2652

J
Josef Bacik 已提交
2653
				ret = btrfs_drop_extents(trans, root, inode,
2654 2655
							 new_key.offset,
							 new_key.offset + datal,
2656
							 1);
2657 2658 2659 2660 2661 2662
				if (ret) {
					btrfs_abort_transaction(trans, root,
								ret);
					btrfs_end_transaction(trans, root);
					goto out;
				}
2663

2664 2665
				ret = btrfs_insert_empty_item(trans, root, path,
							      &new_key, size);
2666 2667 2668 2669 2670 2671
				if (ret) {
					btrfs_abort_transaction(trans, root,
								ret);
					btrfs_end_transaction(trans, root);
					goto out;
				}
2672 2673

				if (skip) {
C
Chris Mason 已提交
2674 2675
					u32 start =
					  btrfs_file_extent_calc_inline_size(0);
2676 2677 2678 2679 2680 2681 2682 2683 2684 2685
					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
2686
			}
2687 2688

			btrfs_mark_buffer_dirty(leaf);
2689
			btrfs_release_path(path);
2690

2691
			inode_inc_iversion(inode);
2692
			inode->i_mtime = inode->i_ctime = CURRENT_TIME;
2693 2694 2695 2696 2697 2698 2699

			/*
			 * 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;
2700 2701
			if (endoff > destoff+olen)
				endoff = destoff+olen;
2702 2703 2704
			if (endoff > inode->i_size)
				btrfs_i_size_write(inode, endoff);

2705
			ret = btrfs_update_inode(trans, root, inode);
2706 2707 2708 2709 2710 2711
			if (ret) {
				btrfs_abort_transaction(trans, root, ret);
				btrfs_end_transaction(trans, root);
				goto out;
			}
			ret = btrfs_end_transaction(trans, root);
2712
		}
C
Chris Mason 已提交
2713
next:
2714
		btrfs_release_path(path);
C
Christoph Hellwig 已提交
2715 2716 2717 2718
		key.offset++;
	}
	ret = 0;
out:
2719
	btrfs_release_path(path);
2720
	unlock_extent(&BTRFS_I(src)->io_tree, off, off+len);
C
Christoph Hellwig 已提交
2721 2722 2723
out_unlock:
	mutex_unlock(&src->i_mutex);
	mutex_unlock(&inode->i_mutex);
Y
Yan Zheng 已提交
2724 2725
	vfree(buf);
	btrfs_free_path(path);
C
Christoph Hellwig 已提交
2726 2727
out_fput:
	fput(src_file);
2728
out_drop_write:
A
Al Viro 已提交
2729
	mnt_drop_write_file(file);
C
Christoph Hellwig 已提交
2730 2731 2732
	return ret;
}

2733
static long btrfs_ioctl_clone_range(struct file *file, void __user *argp)
2734 2735 2736
{
	struct btrfs_ioctl_clone_range_args args;

2737
	if (copy_from_user(&args, argp, sizeof(args)))
2738 2739 2740 2741 2742
		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
2743 2744 2745 2746 2747 2748
/*
 * 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.
 */
2749
static long btrfs_ioctl_trans_start(struct file *file)
C
Christoph Hellwig 已提交
2750 2751 2752 2753
{
	struct inode *inode = fdentry(file)->d_inode;
	struct btrfs_root *root = BTRFS_I(inode)->root;
	struct btrfs_trans_handle *trans;
2754
	int ret;
C
Christoph Hellwig 已提交
2755

2756
	ret = -EPERM;
2757
	if (!capable(CAP_SYS_ADMIN))
2758
		goto out;
2759

2760 2761
	ret = -EINPROGRESS;
	if (file->private_data)
C
Christoph Hellwig 已提交
2762
		goto out;
2763

L
Li Zefan 已提交
2764 2765 2766 2767
	ret = -EROFS;
	if (btrfs_root_readonly(root))
		goto out;

2768
	ret = mnt_want_write_file(file);
Y
Yan Zheng 已提交
2769 2770 2771
	if (ret)
		goto out;

J
Josef Bacik 已提交
2772
	atomic_inc(&root->fs_info->open_ioctl_trans);
2773

2774
	ret = -ENOMEM;
2775
	trans = btrfs_start_ioctl_transaction(root);
2776
	if (IS_ERR(trans))
2777 2778 2779 2780 2781 2782
		goto out_drop;

	file->private_data = trans;
	return 0;

out_drop:
J
Josef Bacik 已提交
2783
	atomic_dec(&root->fs_info->open_ioctl_trans);
A
Al Viro 已提交
2784
	mnt_drop_write_file(file);
C
Christoph Hellwig 已提交
2785 2786 2787 2788
out:
	return ret;
}

2789 2790 2791 2792 2793 2794 2795 2796 2797 2798 2799 2800 2801 2802 2803 2804 2805 2806 2807 2808 2809 2810 2811 2812 2813 2814 2815 2816 2817 2818 2819 2820 2821 2822 2823 2824 2825 2826 2827
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;
	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);
2828
	if (IS_ERR(trans)) {
2829
		btrfs_free_path(path);
2830
		return PTR_ERR(trans);
2831 2832
	}

2833
	dir_id = btrfs_super_root_dir(root->fs_info->super_copy);
2834 2835
	di = btrfs_lookup_dir_item(trans, root->fs_info->tree_root, path,
				   dir_id, "default", 7, 1);
2836
	if (IS_ERR_OR_NULL(di)) {
2837 2838 2839 2840 2841 2842 2843 2844 2845 2846 2847 2848
		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);

2849
	btrfs_set_fs_incompat(root->fs_info, DEFAULT_SUBVOL);
2850 2851 2852 2853 2854
	btrfs_end_transaction(trans, root);

	return 0;
}

2855 2856 2857 2858 2859 2860 2861 2862 2863 2864 2865 2866 2867 2868 2869 2870
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
2871 2872 2873 2874 2875
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;
2876
	struct btrfs_ioctl_space_info *dest_orig;
2877
	struct btrfs_ioctl_space_info __user *user_dest;
J
Josef Bacik 已提交
2878
	struct btrfs_space_info *info;
2879 2880 2881 2882 2883
	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;
2884
	int alloc_size;
J
Josef Bacik 已提交
2885
	int ret = 0;
2886
	u64 slot_count = 0;
2887
	int i, c;
J
Josef Bacik 已提交
2888 2889 2890 2891 2892 2893

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

2894 2895 2896 2897 2898 2899 2900 2901 2902 2903 2904 2905 2906 2907 2908 2909 2910 2911 2912 2913 2914 2915 2916 2917
	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);
	}
2918 2919 2920 2921 2922 2923

	/* space_slots == 0 means they are asking for a count */
	if (space_args.space_slots == 0) {
		space_args.total_spaces = slot_count;
		goto out;
	}
2924

2925
	slot_count = min_t(u64, space_args.space_slots, slot_count);
2926

2927
	alloc_size = sizeof(*dest) * slot_count;
2928

2929 2930 2931 2932 2933 2934
	/* 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
2935
	space_args.total_spaces = 0;
2936 2937 2938 2939
	dest = kmalloc(alloc_size, GFP_NOFS);
	if (!dest)
		return -ENOMEM;
	dest_orig = dest;
J
Josef Bacik 已提交
2940

2941
	/* now we have a buffer to copy into */
2942 2943 2944
	for (i = 0; i < num_types; i++) {
		struct btrfs_space_info *tmp;

2945 2946 2947
		if (!slot_count)
			break;

2948 2949 2950 2951 2952 2953 2954 2955 2956 2957
		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();
2958

2959 2960 2961 2962 2963 2964 2965 2966 2967 2968
		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++;
2969
				slot_count--;
2970
			}
2971 2972
			if (!slot_count)
				break;
2973 2974
		}
		up_read(&info->groups_sem);
J
Josef Bacik 已提交
2975 2976
	}

D
Daniel J Blueman 已提交
2977
	user_dest = (struct btrfs_ioctl_space_info __user *)
2978 2979 2980 2981 2982 2983 2984 2985
		(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
2986 2987 2988 2989 2990
		ret = -EFAULT;

	return ret;
}

C
Christoph Hellwig 已提交
2991 2992 2993 2994 2995 2996 2997 2998 2999 3000 3001 3002 3003
/*
 * 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;
3004 3005
	if (!trans)
		return -EINVAL;
C
Christoph Hellwig 已提交
3006
	file->private_data = NULL;
3007

3008 3009
	btrfs_end_transaction(trans, root);

J
Josef Bacik 已提交
3010
	atomic_dec(&root->fs_info->open_ioctl_trans);
3011

A
Al Viro 已提交
3012
	mnt_drop_write_file(file);
3013
	return 0;
C
Christoph Hellwig 已提交
3014 3015
}

3016 3017 3018 3019 3020
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
3021
	int ret;
3022 3023

	trans = btrfs_start_transaction(root, 0);
3024 3025
	if (IS_ERR(trans))
		return PTR_ERR(trans);
3026
	transid = trans->transid;
T
Tsutomu Itoh 已提交
3027
	ret = btrfs_commit_transaction_async(trans, root, 0);
3028 3029
	if (ret) {
		btrfs_end_transaction(trans, root);
T
Tsutomu Itoh 已提交
3030
		return ret;
3031
	}
3032 3033 3034 3035 3036 3037 3038 3039 3040 3041 3042 3043 3044 3045 3046 3047 3048 3049 3050 3051 3052

	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
3053 3054 3055 3056 3057 3058 3059 3060 3061 3062 3063 3064 3065
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
3066
			      &sa->progress, sa->flags & BTRFS_SCRUB_READONLY);
J
Jan Schmidt 已提交
3067 3068 3069 3070 3071 3072 3073 3074 3075 3076 3077 3078 3079 3080 3081 3082 3083 3084 3085 3086 3087 3088 3089 3090 3091 3092 3093 3094 3095 3096 3097 3098 3099 3100 3101 3102 3103 3104

	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;
}

3105
static long btrfs_ioctl_get_dev_stats(struct btrfs_root *root,
3106
				      void __user *arg)
3107 3108 3109 3110 3111 3112 3113 3114
{
	struct btrfs_ioctl_get_dev_stats *sa;
	int ret;

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

3115 3116 3117 3118 3119 3120
	if ((sa->flags & BTRFS_DEV_STATS_RESET) && !capable(CAP_SYS_ADMIN)) {
		kfree(sa);
		return -EPERM;
	}

	ret = btrfs_get_dev_stats(root, sa);
3121 3122 3123 3124 3125 3126 3127 3128

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

	kfree(sa);
	return ret;
}

3129 3130 3131 3132
static long btrfs_ioctl_ino_to_path(struct btrfs_root *root, void __user *arg)
{
	int ret = 0;
	int i;
3133
	u64 rel_ptr;
3134
	int size;
3135
	struct btrfs_ioctl_ino_path_args *ipa = NULL;
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
	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) {
3168 3169
		rel_ptr = ipath->fspath->val[i] -
			  (u64)(unsigned long)ipath->fspath->val;
3170
		ipath->fspath->val[i] = rel_ptr;
3171 3172
	}

3173 3174
	ret = copy_to_user((void *)(unsigned long)ipa->fspath,
			   (void *)(unsigned long)ipath->fspath, size);
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 3202 3203 3204 3205 3206 3207 3208 3209 3210 3211 3212
	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
3213
	u64 extent_item_pos;
3214 3215 3216 3217 3218 3219 3220 3221 3222 3223 3224 3225 3226 3227 3228 3229 3230 3231 3232 3233 3234 3235 3236 3237 3238 3239 3240 3241 3242 3243
	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
3244
	btrfs_release_path(path);
3245 3246 3247 3248 3249 3250

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

J
Jan Schmidt 已提交
3251
	extent_item_pos = loi->logical - key.objectid;
3252 3253
	ret = iterate_extent_inodes(root->fs_info, key.objectid,
					extent_item_pos, 0, build_ino_list,
J
Jan Schmidt 已提交
3254
					inodes);
3255 3256 3257 3258

	if (ret < 0)
		goto out;

3259 3260
	ret = copy_to_user((void *)(unsigned long)loi->inodes,
			   (void *)(unsigned long)inodes, size);
3261 3262 3263 3264 3265 3266 3267 3268 3269 3270 3271
	if (ret)
		ret = -EFAULT;

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

	return ret;
}

3272
void update_ioctl_balance_args(struct btrfs_fs_info *fs_info, int lock,
3273 3274 3275 3276 3277 3278
			       struct btrfs_ioctl_balance_args *bargs)
{
	struct btrfs_balance_control *bctl = fs_info->balance_ctl;

	bargs->flags = bctl->flags;

3279 3280 3281 3282
	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;
3283 3284
	if (atomic_read(&fs_info->balance_cancel_req))
		bargs->state |= BTRFS_BALANCE_STATE_CANCEL_REQ;
3285

3286 3287 3288
	memcpy(&bargs->data, &bctl->data, sizeof(bargs->data));
	memcpy(&bargs->meta, &bctl->meta, sizeof(bargs->meta));
	memcpy(&bargs->sys, &bctl->sys, sizeof(bargs->sys));
3289 3290 3291 3292 3293 3294 3295 3296

	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));
	}
3297 3298
}

3299
static long btrfs_ioctl_balance(struct file *file, void __user *arg)
3300
{
3301
	struct btrfs_root *root = BTRFS_I(fdentry(file)->d_inode)->root;
3302 3303 3304 3305 3306 3307 3308 3309
	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;

3310
	ret = mnt_want_write_file(file);
3311 3312 3313
	if (ret)
		return ret;

3314 3315 3316 3317 3318 3319 3320 3321 3322
	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;
		}
3323 3324 3325 3326 3327 3328 3329 3330 3331 3332 3333 3334 3335 3336

		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;
		}
3337 3338 3339 3340
	} else {
		bargs = NULL;
	}

3341 3342 3343 3344 3345
	if (fs_info->balance_ctl) {
		ret = -EINPROGRESS;
		goto out_bargs;
	}

3346 3347 3348 3349 3350 3351 3352 3353 3354 3355 3356 3357 3358
	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;
3359 3360 3361
	} else {
		/* balance everything - no filters */
		bctl->flags |= BTRFS_BALANCE_TYPE_MASK;
3362 3363
	}

3364
do_balance:
3365 3366
	ret = btrfs_balance(bctl, bargs);
	/*
3367 3368
	 * bctl is freed in __cancel_balance or in free_fs_info if
	 * restriper was paused all the way until unmount
3369 3370 3371 3372 3373 3374 3375 3376 3377 3378 3379
	 */
	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);
3380
	mnt_drop_write_file(file);
3381 3382 3383
	return ret;
}

3384 3385 3386 3387 3388 3389 3390 3391
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);
3392 3393
	case BTRFS_BALANCE_CTL_CANCEL:
		return btrfs_cancel_balance(root->fs_info);
3394 3395 3396 3397 3398
	}

	return -EINVAL;
}

3399 3400 3401 3402 3403 3404 3405 3406 3407 3408 3409 3410 3411 3412 3413 3414 3415 3416 3417 3418 3419 3420 3421 3422 3423 3424 3425 3426 3427 3428 3429 3430 3431
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;
}

A
Arne Jansen 已提交
3432 3433 3434 3435 3436 3437 3438 3439 3440 3441 3442 3443 3444 3445 3446 3447 3448 3449 3450 3451 3452 3453 3454 3455 3456 3457 3458 3459 3460 3461 3462 3463 3464 3465 3466 3467 3468 3469 3470 3471 3472 3473 3474 3475 3476 3477 3478 3479 3480 3481 3482 3483 3484 3485 3486 3487 3488 3489 3490 3491 3492 3493 3494 3495 3496 3497 3498 3499 3500 3501 3502 3503 3504 3505 3506 3507 3508 3509 3510 3511 3512 3513 3514 3515 3516 3517 3518 3519 3520 3521 3522 3523 3524 3525 3526 3527 3528 3529 3530 3531 3532 3533 3534 3535 3536 3537 3538 3539 3540 3541 3542 3543 3544 3545 3546 3547 3548 3549 3550 3551 3552 3553 3554 3555 3556 3557 3558 3559 3560 3561 3562 3563 3564 3565 3566 3567 3568 3569 3570 3571 3572 3573 3574 3575 3576 3577 3578 3579 3580 3581 3582 3583 3584 3585 3586 3587 3588 3589 3590 3591 3592 3593 3594 3595 3596 3597 3598 3599 3600 3601 3602 3603 3604 3605 3606 3607 3608
static long btrfs_ioctl_quota_ctl(struct btrfs_root *root, void __user *arg)
{
	struct btrfs_ioctl_quota_ctl_args *sa;
	struct btrfs_trans_handle *trans = NULL;
	int ret;
	int err;

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

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

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

	if (sa->cmd != BTRFS_QUOTA_CTL_RESCAN) {
		trans = btrfs_start_transaction(root, 2);
		if (IS_ERR(trans)) {
			ret = PTR_ERR(trans);
			goto out;
		}
	}

	switch (sa->cmd) {
	case BTRFS_QUOTA_CTL_ENABLE:
		ret = btrfs_quota_enable(trans, root->fs_info);
		break;
	case BTRFS_QUOTA_CTL_DISABLE:
		ret = btrfs_quota_disable(trans, root->fs_info);
		break;
	case BTRFS_QUOTA_CTL_RESCAN:
		ret = btrfs_quota_rescan(root->fs_info);
		break;
	default:
		ret = -EINVAL;
		break;
	}

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

	if (trans) {
		err = btrfs_commit_transaction(trans, root);
		if (err && !ret)
			ret = err;
	}

out:
	kfree(sa);
	return ret;
}

static long btrfs_ioctl_qgroup_assign(struct btrfs_root *root, void __user *arg)
{
	struct btrfs_ioctl_qgroup_assign_args *sa;
	struct btrfs_trans_handle *trans;
	int ret;
	int err;

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

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

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

	trans = btrfs_join_transaction(root);
	if (IS_ERR(trans)) {
		ret = PTR_ERR(trans);
		goto out;
	}

	/* FIXME: check if the IDs really exist */
	if (sa->assign) {
		ret = btrfs_add_qgroup_relation(trans, root->fs_info,
						sa->src, sa->dst);
	} else {
		ret = btrfs_del_qgroup_relation(trans, root->fs_info,
						sa->src, sa->dst);
	}

	err = btrfs_end_transaction(trans, root);
	if (err && !ret)
		ret = err;

out:
	kfree(sa);
	return ret;
}

static long btrfs_ioctl_qgroup_create(struct btrfs_root *root, void __user *arg)
{
	struct btrfs_ioctl_qgroup_create_args *sa;
	struct btrfs_trans_handle *trans;
	int ret;
	int err;

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

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

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

	trans = btrfs_join_transaction(root);
	if (IS_ERR(trans)) {
		ret = PTR_ERR(trans);
		goto out;
	}

	/* FIXME: check if the IDs really exist */
	if (sa->create) {
		ret = btrfs_create_qgroup(trans, root->fs_info, sa->qgroupid,
					  NULL);
	} else {
		ret = btrfs_remove_qgroup(trans, root->fs_info, sa->qgroupid);
	}

	err = btrfs_end_transaction(trans, root);
	if (err && !ret)
		ret = err;

out:
	kfree(sa);
	return ret;
}

static long btrfs_ioctl_qgroup_limit(struct btrfs_root *root, void __user *arg)
{
	struct btrfs_ioctl_qgroup_limit_args *sa;
	struct btrfs_trans_handle *trans;
	int ret;
	int err;
	u64 qgroupid;

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

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

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

	trans = btrfs_join_transaction(root);
	if (IS_ERR(trans)) {
		ret = PTR_ERR(trans);
		goto out;
	}

	qgroupid = sa->qgroupid;
	if (!qgroupid) {
		/* take the current subvol as qgroup */
		qgroupid = root->root_key.objectid;
	}

	/* FIXME: check if the IDs really exist */
	ret = btrfs_limit_qgroup(trans, root->fs_info, qgroupid, &sa->lim);

	err = btrfs_end_transaction(trans, root);
	if (err && !ret)
		ret = err;

out:
	kfree(sa);
	return ret;
}

3609 3610 3611 3612 3613 3614 3615 3616 3617 3618 3619 3620 3621 3622 3623 3624 3625 3626 3627 3628 3629 3630 3631 3632 3633 3634 3635 3636 3637 3638 3639 3640 3641 3642 3643 3644 3645 3646 3647 3648 3649 3650 3651 3652 3653 3654 3655 3656 3657 3658 3659 3660 3661 3662 3663 3664 3665 3666 3667 3668 3669 3670 3671 3672 3673 3674 3675 3676 3677 3678 3679 3680 3681 3682 3683 3684 3685 3686 3687 3688 3689
static long btrfs_ioctl_set_received_subvol(struct file *file,
					    void __user *arg)
{
	struct btrfs_ioctl_received_subvol_args *sa = NULL;
	struct inode *inode = fdentry(file)->d_inode;
	struct btrfs_root *root = BTRFS_I(inode)->root;
	struct btrfs_root_item *root_item = &root->root_item;
	struct btrfs_trans_handle *trans;
	struct timespec ct = CURRENT_TIME;
	int ret = 0;

	ret = mnt_want_write_file(file);
	if (ret < 0)
		return ret;

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

	if (btrfs_ino(inode) != BTRFS_FIRST_FREE_OBJECTID) {
		ret = -EINVAL;
		goto out;
	}

	if (btrfs_root_readonly(root)) {
		ret = -EROFS;
		goto out;
	}

	if (!inode_owner_or_capable(inode)) {
		ret = -EACCES;
		goto out;
	}

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

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

	sa->rtransid = trans->transid;
	sa->rtime.sec = ct.tv_sec;
	sa->rtime.nsec = ct.tv_nsec;

	memcpy(root_item->received_uuid, sa->uuid, BTRFS_UUID_SIZE);
	btrfs_set_root_stransid(root_item, sa->stransid);
	btrfs_set_root_rtransid(root_item, sa->rtransid);
	root_item->stime.sec = cpu_to_le64(sa->stime.sec);
	root_item->stime.nsec = cpu_to_le32(sa->stime.nsec);
	root_item->rtime.sec = cpu_to_le64(sa->rtime.sec);
	root_item->rtime.nsec = cpu_to_le32(sa->rtime.nsec);

	ret = btrfs_update_root(trans, root->fs_info->tree_root,
				&root->root_key, &root->root_item);
	if (ret < 0) {
		btrfs_end_transaction(trans, root);
		trans = NULL;
		goto out;
	} else {
		ret = btrfs_commit_transaction(trans, root);
		if (ret < 0)
			goto out;
	}

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

out:
	kfree(sa);
	up_write(&root->fs_info->subvol_sem);
	mnt_drop_write_file(file);
	return ret;
}

C
Christoph Hellwig 已提交
3690 3691 3692 3693
long btrfs_ioctl(struct file *file, unsigned int
		cmd, unsigned long arg)
{
	struct btrfs_root *root = BTRFS_I(fdentry(file)->d_inode)->root;
3694
	void __user *argp = (void __user *)arg;
C
Christoph Hellwig 已提交
3695 3696

	switch (cmd) {
3697 3698 3699 3700 3701 3702
	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);
3703 3704
	case FITRIM:
		return btrfs_ioctl_fitrim(file, argp);
C
Christoph Hellwig 已提交
3705
	case BTRFS_IOC_SNAP_CREATE:
3706
		return btrfs_ioctl_snap_create(file, argp, 0);
3707
	case BTRFS_IOC_SNAP_CREATE_V2:
3708
		return btrfs_ioctl_snap_create_v2(file, argp, 0);
3709
	case BTRFS_IOC_SUBVOL_CREATE:
3710
		return btrfs_ioctl_snap_create(file, argp, 1);
A
Arne Jansen 已提交
3711 3712
	case BTRFS_IOC_SUBVOL_CREATE_V2:
		return btrfs_ioctl_snap_create_v2(file, argp, 1);
3713 3714
	case BTRFS_IOC_SNAP_DESTROY:
		return btrfs_ioctl_snap_destroy(file, argp);
3715 3716 3717 3718
	case BTRFS_IOC_SUBVOL_GETFLAGS:
		return btrfs_ioctl_subvol_getflags(file, argp);
	case BTRFS_IOC_SUBVOL_SETFLAGS:
		return btrfs_ioctl_subvol_setflags(file, argp);
3719 3720
	case BTRFS_IOC_DEFAULT_SUBVOL:
		return btrfs_ioctl_default_subvol(file, argp);
C
Christoph Hellwig 已提交
3721
	case BTRFS_IOC_DEFRAG:
C
Chris Mason 已提交
3722 3723 3724
		return btrfs_ioctl_defrag(file, NULL);
	case BTRFS_IOC_DEFRAG_RANGE:
		return btrfs_ioctl_defrag(file, argp);
C
Christoph Hellwig 已提交
3725
	case BTRFS_IOC_RESIZE:
3726
		return btrfs_ioctl_resize(root, argp);
C
Christoph Hellwig 已提交
3727
	case BTRFS_IOC_ADD_DEV:
3728
		return btrfs_ioctl_add_dev(root, argp);
C
Christoph Hellwig 已提交
3729
	case BTRFS_IOC_RM_DEV:
3730
		return btrfs_ioctl_rm_dev(root, argp);
J
Jan Schmidt 已提交
3731 3732 3733 3734
	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
3735
	case BTRFS_IOC_BALANCE:
3736
		return btrfs_ioctl_balance(file, NULL);
C
Christoph Hellwig 已提交
3737
	case BTRFS_IOC_CLONE:
3738 3739
		return btrfs_ioctl_clone(file, arg, 0, 0, 0);
	case BTRFS_IOC_CLONE_RANGE:
3740
		return btrfs_ioctl_clone_range(file, argp);
C
Christoph Hellwig 已提交
3741 3742 3743 3744
	case BTRFS_IOC_TRANS_START:
		return btrfs_ioctl_trans_start(file);
	case BTRFS_IOC_TRANS_END:
		return btrfs_ioctl_trans_end(file);
3745 3746 3747 3748
	case BTRFS_IOC_TREE_SEARCH:
		return btrfs_ioctl_tree_search(file, argp);
	case BTRFS_IOC_INO_LOOKUP:
		return btrfs_ioctl_ino_lookup(file, argp);
3749 3750 3751 3752
	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
3753 3754
	case BTRFS_IOC_SPACE_INFO:
		return btrfs_ioctl_space_info(root, argp);
C
Christoph Hellwig 已提交
3755 3756 3757
	case BTRFS_IOC_SYNC:
		btrfs_sync_fs(file->f_dentry->d_sb, 1);
		return 0;
3758 3759 3760 3761
	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
3762 3763 3764 3765 3766 3767
	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);
3768
	case BTRFS_IOC_BALANCE_V2:
3769
		return btrfs_ioctl_balance(file, argp);
3770 3771
	case BTRFS_IOC_BALANCE_CTL:
		return btrfs_ioctl_balance_ctl(root, arg);
3772 3773
	case BTRFS_IOC_BALANCE_PROGRESS:
		return btrfs_ioctl_balance_progress(root, argp);
3774 3775
	case BTRFS_IOC_SET_RECEIVED_SUBVOL:
		return btrfs_ioctl_set_received_subvol(file, argp);
3776 3777
	case BTRFS_IOC_SEND:
		return btrfs_ioctl_send(file, argp);
3778
	case BTRFS_IOC_GET_DEV_STATS:
3779
		return btrfs_ioctl_get_dev_stats(root, argp);
A
Arne Jansen 已提交
3780 3781 3782 3783 3784 3785 3786 3787
	case BTRFS_IOC_QUOTA_CTL:
		return btrfs_ioctl_quota_ctl(root, argp);
	case BTRFS_IOC_QGROUP_ASSIGN:
		return btrfs_ioctl_qgroup_assign(root, argp);
	case BTRFS_IOC_QGROUP_CREATE:
		return btrfs_ioctl_qgroup_create(root, argp);
	case BTRFS_IOC_QGROUP_LIMIT:
		return btrfs_ioctl_qgroup_limit(root, argp);
C
Christoph Hellwig 已提交
3788 3789 3790 3791
	}

	return -ENOTTY;
}