ioctl.c 55.1 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>
C
Chris Mason 已提交
43
#include "compat.h"
C
Christoph Hellwig 已提交
44 45 46 47 48 49 50
#include "ctree.h"
#include "disk-io.h"
#include "transaction.h"
#include "btrfs_inode.h"
#include "ioctl.h"
#include "print-tree.h"
#include "volumes.h"
51
#include "locking.h"
C
Christoph Hellwig 已提交
52

53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114
/* 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;

	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.
 *
 * Unlike extN we don't have any flags we don't want to inherit currently.
 */
void btrfs_inherit_iflags(struct inode *inode, struct inode *dir)
{
115 116 117 118 119 120
	unsigned int flags;

	if (!dir)
		return;

	flags = BTRFS_I(dir)->flags;
121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156

	if (S_ISREG(inode->i_mode))
		flags &= ~BTRFS_INODE_DIRSYNC;
	else if (!S_ISDIR(inode->i_mode))
		flags &= (BTRFS_INODE_NODUMP | BTRFS_INODE_NOATIME);

	BTRFS_I(inode)->flags = flags;
	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;
}

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;

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

	if (flags & ~(FS_IMMUTABLE_FL | FS_APPEND_FL | \
		      FS_NOATIME_FL | FS_NODUMP_FL | \
		      FS_SYNC_FL | FS_DIRSYNC_FL))
		return -EOPNOTSUPP;
C
Christoph Hellwig 已提交
157

158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223
	if (!is_owner_or_cap(inode))
		return -EACCES;

	mutex_lock(&inode->i_mutex);

	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;
		}
	}

	ret = mnt_want_write(file->f_path.mnt);
	if (ret)
		goto out_unlock;

	if (flags & FS_SYNC_FL)
		ip->flags |= BTRFS_INODE_SYNC;
	else
		ip->flags &= ~BTRFS_INODE_SYNC;
	if (flags & FS_IMMUTABLE_FL)
		ip->flags |= BTRFS_INODE_IMMUTABLE;
	else
		ip->flags &= ~BTRFS_INODE_IMMUTABLE;
	if (flags & FS_APPEND_FL)
		ip->flags |= BTRFS_INODE_APPEND;
	else
		ip->flags &= ~BTRFS_INODE_APPEND;
	if (flags & FS_NODUMP_FL)
		ip->flags |= BTRFS_INODE_NODUMP;
	else
		ip->flags &= ~BTRFS_INODE_NODUMP;
	if (flags & FS_NOATIME_FL)
		ip->flags |= BTRFS_INODE_NOATIME;
	else
		ip->flags &= ~BTRFS_INODE_NOATIME;
	if (flags & FS_DIRSYNC_FL)
		ip->flags |= BTRFS_INODE_DIRSYNC;
	else
		ip->flags &= ~BTRFS_INODE_DIRSYNC;


	trans = btrfs_join_transaction(root, 1);
	BUG_ON(!trans);

	ret = btrfs_update_inode(trans, root, inode);
	BUG_ON(ret);

	btrfs_update_iflags(inode);
	inode->i_ctime = CURRENT_TIME;
	btrfs_end_transaction(trans, root);

	mnt_drop_write(file->f_path.mnt);
 out_unlock:
	mutex_unlock(&inode->i_mutex);
	return 0;
}

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
224

225 226
static noinline int create_subvol(struct btrfs_root *root,
				  struct dentry *dentry,
S
Sage Weil 已提交
227 228
				  char *name, int namelen,
				  u64 *async_transid)
C
Christoph Hellwig 已提交
229 230 231 232 233 234
{
	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;
235 236
	struct btrfs_root *new_root;
	struct inode *dir = dentry->d_parent->d_inode;
C
Christoph Hellwig 已提交
237 238 239 240
	int ret;
	int err;
	u64 objectid;
	u64 new_dirid = BTRFS_FIRST_FREE_OBJECTID;
241
	u64 index = 0;
C
Christoph Hellwig 已提交
242

243 244 245 246
	ret = btrfs_find_free_objectid(NULL, root->fs_info->tree_root,
				       0, &objectid);
	if (ret)
		return ret;
J
Josef Bacik 已提交
247 248 249 250 251 252
	/*
	 * 1 - inode item
	 * 2 - refs
	 * 1 - root item
	 * 2 - dir items
	 */
253 254 255
	trans = btrfs_start_transaction(root, 6);
	if (IS_ERR(trans))
		return PTR_ERR(trans);
C
Christoph Hellwig 已提交
256

257 258
	leaf = btrfs_alloc_free_block(trans, root, root->leafsize,
				      0, objectid, NULL, 0, 0, 0);
259 260 261 262
	if (IS_ERR(leaf)) {
		ret = PTR_ERR(leaf);
		goto fail;
	}
C
Christoph Hellwig 已提交
263

264
	memset_extent_buffer(leaf, 0, 0, sizeof(struct btrfs_header));
C
Christoph Hellwig 已提交
265 266
	btrfs_set_header_bytenr(leaf, leaf->start);
	btrfs_set_header_generation(leaf, trans->transid);
267
	btrfs_set_header_backref_rev(leaf, BTRFS_MIXED_BACKREF_REV);
C
Christoph Hellwig 已提交
268 269 270 271 272
	btrfs_set_header_owner(leaf, objectid);

	write_extent_buffer(leaf, root->fs_info->fsid,
			    (unsigned long)btrfs_header_fsid(leaf),
			    BTRFS_FSID_SIZE);
273 274 275
	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
276 277 278 279 280 281 282
	btrfs_mark_buffer_dirty(leaf);

	inode_item = &root_item.inode;
	memset(inode_item, 0, sizeof(*inode_item));
	inode_item->generation = cpu_to_le64(1);
	inode_item->size = cpu_to_le64(3);
	inode_item->nlink = cpu_to_le32(1);
283
	inode_item->nbytes = cpu_to_le64(root->leafsize);
C
Christoph Hellwig 已提交
284 285 286
	inode_item->mode = cpu_to_le32(S_IFDIR | 0755);

	btrfs_set_root_bytenr(&root_item, leaf->start);
287
	btrfs_set_root_generation(&root_item, trans->transid);
C
Christoph Hellwig 已提交
288 289
	btrfs_set_root_level(&root_item, 0);
	btrfs_set_root_refs(&root_item, 1);
290
	btrfs_set_root_used(&root_item, leaf->len);
Y
Yan Zheng 已提交
291
	btrfs_set_root_last_snapshot(&root_item, 0);
C
Christoph Hellwig 已提交
292 293 294 295

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

296
	btrfs_tree_unlock(leaf);
C
Christoph Hellwig 已提交
297 298 299 300 301 302
	free_extent_buffer(leaf);
	leaf = NULL;

	btrfs_set_root_dirid(&root_item, new_dirid);

	key.objectid = objectid;
303
	key.offset = 0;
C
Christoph Hellwig 已提交
304 305 306 307 308 309
	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;

310 311 312 313 314 315 316 317
	key.offset = (u64)-1;
	new_root = btrfs_read_fs_root_no_name(root->fs_info, &key);
	BUG_ON(IS_ERR(new_root));

	btrfs_record_root_in_trans(trans, new_root);

	ret = btrfs_create_subvol_root(trans, new_root, new_dirid,
				       BTRFS_I(dir)->block_group);
C
Christoph Hellwig 已提交
318 319 320
	/*
	 * insert the directory item
	 */
321 322 323 324
	ret = btrfs_set_inode_index(dir, &index);
	BUG_ON(ret);

	ret = btrfs_insert_dir_item(trans, root,
C
Christoph Hellwig 已提交
325
				    name, namelen, dir->i_ino, &key,
326
				    BTRFS_FT_DIR, index);
C
Christoph Hellwig 已提交
327 328
	if (ret)
		goto fail;
329

330 331 332 333
	btrfs_i_size_write(dir, dir->i_size + namelen * 2);
	ret = btrfs_update_inode(trans, root, dir);
	BUG_ON(ret);

334
	ret = btrfs_add_root_ref(trans, root->fs_info->tree_root,
335
				 objectid, root->root_key.objectid,
336 337
				 dir->i_ino, index, name, namelen);

338
	BUG_ON(ret);
C
Christoph Hellwig 已提交
339

340
	d_instantiate(dentry, btrfs_lookup_dentry(dir, dentry));
C
Christoph Hellwig 已提交
341
fail:
S
Sage Weil 已提交
342 343 344 345 346 347
	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
348 349 350 351 352
	if (err && !ret)
		ret = err;
	return ret;
}

S
Sage Weil 已提交
353 354
static int create_snapshot(struct btrfs_root *root, struct dentry *dentry,
			   char *name, int namelen, u64 *async_transid)
C
Christoph Hellwig 已提交
355
{
356
	struct inode *inode;
C
Christoph Hellwig 已提交
357 358
	struct btrfs_pending_snapshot *pending_snapshot;
	struct btrfs_trans_handle *trans;
359
	int ret;
C
Christoph Hellwig 已提交
360 361 362 363

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

364
	pending_snapshot = kzalloc(sizeof(*pending_snapshot), GFP_NOFS);
365 366 367 368
	if (!pending_snapshot)
		return -ENOMEM;

	btrfs_init_block_rsv(&pending_snapshot->block_rsv);
369
	pending_snapshot->dentry = dentry;
C
Christoph Hellwig 已提交
370
	pending_snapshot->root = root;
371 372 373 374 375 376 377 378 379 380

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

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

C
Christoph Hellwig 已提交
381 382
	list_add(&pending_snapshot->list,
		 &trans->transaction->pending_snapshots);
S
Sage Weil 已提交
383 384 385 386 387 388 389 390
	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);
	}
391
	BUG_ON(ret);
392 393 394 395 396 397

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

	btrfs_orphan_cleanup(pending_snapshot->snap);
C
Christoph Hellwig 已提交
398

399 400 401 402 403 404 405 406 407
	inode = btrfs_lookup_dentry(dentry->d_parent->d_inode, dentry);
	if (IS_ERR(inode)) {
		ret = PTR_ERR(inode);
		goto fail;
	}
	BUG_ON(!inode);
	d_instantiate(dentry, inode);
	ret = 0;
fail:
408
	kfree(pending_snapshot);
C
Christoph Hellwig 已提交
409 410 411
	return ret;
}

412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481
/*  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;
}

482 483 484 485 486 487 488 489 490 491 492 493 494 495 496
/* 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.
 */
497 498
static noinline int btrfs_mksubvol(struct path *parent,
				   char *name, int namelen,
S
Sage Weil 已提交
499 500
				   struct btrfs_root *snap_src,
				   u64 *async_transid)
501
{
502
	struct inode *dir  = parent->dentry->d_inode;
503 504 505
	struct dentry *dentry;
	int error;

506
	mutex_lock_nested(&dir->i_mutex, I_MUTEX_PARENT);
507 508 509 510 511 512 513 514 515 516 517 518 519 520

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

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

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

521
	error = btrfs_may_create(dir, dentry);
522 523 524
	if (error)
		goto out_drop_write;

525 526 527 528 529
	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;

530
	if (snap_src) {
S
Sage Weil 已提交
531 532
		error = create_snapshot(snap_src, dentry,
					name, namelen, async_transid);
533
	} else {
534
		error = create_subvol(BTRFS_I(dir)->root, dentry,
S
Sage Weil 已提交
535
				      name, namelen, async_transid);
536
	}
537 538 539 540
	if (!error)
		fsnotify_mkdir(dir, dentry);
out_up_read:
	up_read(&BTRFS_I(dir)->root->fs_info->subvol_sem);
541 542 543 544 545
out_drop_write:
	mnt_drop_write(parent->mnt);
out_dput:
	dput(dentry);
out_unlock:
546
	mutex_unlock(&dir->i_mutex);
547 548 549
	return error;
}

550
static int should_defrag_range(struct inode *inode, u64 start, u64 len,
C
Chris Mason 已提交
551 552
			       int thresh, u64 *last_len, u64 *skip,
			       u64 *defrag_end)
553 554 555 556 557 558
{
	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;
	int ret = 1;

C
Chris Mason 已提交
559 560 561 562

	if (thresh == 0)
		thresh = 256 * 1024;

563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585
	/*
	 * make sure that once we start defragging and extent, we keep on
	 * defragging it
	 */
	if (start < *defrag_end)
		return 1;

	*skip = 0;

	/*
	 * hopefully we have this extent in the tree already, try without
	 * the full extent lock
	 */
	read_lock(&em_tree->lock);
	em = lookup_extent_mapping(em_tree, start, len);
	read_unlock(&em_tree->lock);

	if (!em) {
		/* get the big lock and read metadata off disk */
		lock_extent(io_tree, start, start + len - 1, GFP_NOFS);
		em = btrfs_get_extent(inode, NULL, 0, start, len, 0);
		unlock_extent(io_tree, start, start + len - 1, GFP_NOFS);

586
		if (IS_ERR(em))
587 588 589 590 591 592 593 594 595 596
			return 0;
	}

	/* this will cover holes, and inline extents */
	if (em->block_start >= EXTENT_MAP_LAST_BYTE)
		ret = 0;

	/*
	 * we hit a real extent, if it is big don't bother defragging it again
	 */
C
Chris Mason 已提交
597
	if ((*last_len == 0 || *last_len >= thresh) && em->len >= thresh)
598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620
		ret = 0;

	/*
	 * 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) {
		*last_len += len;
		*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
621 622
static int btrfs_defrag_file(struct file *file,
			     struct btrfs_ioctl_defrag_range_args *range)
C
Christoph Hellwig 已提交
623 624 625 626
{
	struct inode *inode = fdentry(file)->d_inode;
	struct btrfs_root *root = BTRFS_I(inode)->root;
	struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree;
627
	struct btrfs_ordered_extent *ordered;
C
Christoph Hellwig 已提交
628 629 630 631 632 633
	struct page *page;
	unsigned long last_index;
	unsigned long ra_pages = root->fs_info->bdi.ra_pages;
	unsigned long total_read = 0;
	u64 page_start;
	u64 page_end;
634 635 636
	u64 last_len = 0;
	u64 skip = 0;
	u64 defrag_end = 0;
C
Christoph Hellwig 已提交
637 638 639
	unsigned long i;
	int ret;

640 641 642
	if (inode->i_size == 0)
		return 0;

C
Chris Mason 已提交
643 644 645 646 647 648 649 650
	if (range->start + range->len > range->start) {
		last_index = min_t(u64, inode->i_size - 1,
			 range->start + range->len - 1) >> PAGE_CACHE_SHIFT;
	} else {
		last_index = (inode->i_size - 1) >> PAGE_CACHE_SHIFT;
	}

	i = range->start >> PAGE_CACHE_SHIFT;
651 652
	while (i <= last_index) {
		if (!should_defrag_range(inode, (u64)i << PAGE_CACHE_SHIFT,
C
Chris Mason 已提交
653 654 655
					PAGE_CACHE_SIZE,
					range->extent_thresh,
					&last_len, &skip,
656 657 658 659 660 661 662 663 664 665
					&defrag_end)) {
			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;
		}
C
Christoph Hellwig 已提交
666 667 668 669 670 671

		if (total_read % ra_pages == 0) {
			btrfs_force_ra(inode->i_mapping, &file->f_ra, file, i,
				       min(last_index, i + ra_pages - 1));
		}
		total_read++;
672
		mutex_lock(&inode->i_mutex);
C
Chris Mason 已提交
673 674
		if (range->flags & BTRFS_DEFRAG_RANGE_COMPRESS)
			BTRFS_I(inode)->force_compress = 1;
675

676 677 678
		ret  = btrfs_delalloc_reserve_space(inode, PAGE_CACHE_SIZE);
		if (ret)
			goto err_unlock;
679
again:
680 681 682 683 684 685
		if (inode->i_size == 0 ||
		    i > ((inode->i_size - 1) >> PAGE_CACHE_SHIFT)) {
			ret = 0;
			goto err_reservations;
		}

C
Christoph Hellwig 已提交
686
		page = grab_cache_page(inode->i_mapping, i);
687 688
		if (!page) {
			ret = -ENOMEM;
689
			goto err_reservations;
690
		}
691

C
Christoph Hellwig 已提交
692 693 694 695 696 697
		if (!PageUptodate(page)) {
			btrfs_readpage(NULL, page);
			lock_page(page);
			if (!PageUptodate(page)) {
				unlock_page(page);
				page_cache_release(page);
698
				ret = -EIO;
699
				goto err_reservations;
C
Christoph Hellwig 已提交
700 701 702
			}
		}

703 704 705 706 707 708
		if (page->mapping != inode->i_mapping) {
			unlock_page(page);
			page_cache_release(page);
			goto again;
		}

C
Christoph Hellwig 已提交
709 710
		wait_on_page_writeback(page);

711
		if (PageDirty(page)) {
712
			btrfs_delalloc_release_space(inode, PAGE_CACHE_SIZE);
713 714 715
			goto loop_unlock;
		}

C
Christoph Hellwig 已提交
716 717 718
		page_start = (u64)page->index << PAGE_CACHE_SHIFT;
		page_end = page_start + PAGE_CACHE_SIZE - 1;
		lock_extent(io_tree, page_start, page_end, GFP_NOFS);
719 720 721 722 723 724 725 726 727 728 729 730

		ordered = btrfs_lookup_ordered_extent(inode, page_start);
		if (ordered) {
			unlock_extent(io_tree, page_start, page_end, GFP_NOFS);
			unlock_page(page);
			page_cache_release(page);
			btrfs_start_ordered_extent(inode, ordered, 1);
			btrfs_put_ordered_extent(ordered);
			goto again;
		}
		set_page_extent_mapped(page);

731 732 733 734 735
		/*
		 * this makes sure page_mkwrite is called on the
		 * page if it is dirtied again later
		 */
		clear_page_dirty_for_io(page);
736 737 738
		clear_extent_bits(&BTRFS_I(inode)->io_tree, page_start,
				  page_end, EXTENT_DIRTY | EXTENT_DELALLOC |
				  EXTENT_DO_ACCOUNTING, GFP_NOFS);
739

740
		btrfs_set_extent_delalloc(inode, page_start, page_end, NULL);
741
		ClearPageChecked(page);
C
Christoph Hellwig 已提交
742
		set_page_dirty(page);
C
Chris Mason 已提交
743
		unlock_extent(io_tree, page_start, page_end, GFP_NOFS);
744 745

loop_unlock:
C
Christoph Hellwig 已提交
746 747
		unlock_page(page);
		page_cache_release(page);
748 749
		mutex_unlock(&inode->i_mutex);

C
Christoph Hellwig 已提交
750
		balance_dirty_pages_ratelimited_nr(inode->i_mapping, 1);
751
		i++;
C
Christoph Hellwig 已提交
752 753
	}

C
Chris Mason 已提交
754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775
	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);
		BTRFS_I(inode)->force_compress = 0;
		mutex_unlock(&inode->i_mutex);
	}

C
Christoph Hellwig 已提交
776
	return 0;
777 778

err_reservations:
779 780
	btrfs_delalloc_release_space(inode, PAGE_CACHE_SIZE);
err_unlock:
781 782
	mutex_unlock(&inode->i_mutex);
	return ret;
C
Christoph Hellwig 已提交
783 784
}

785 786
static noinline int btrfs_ioctl_resize(struct btrfs_root *root,
					void __user *arg)
C
Christoph Hellwig 已提交
787 788 789 790 791 792 793 794 795 796 797 798
{
	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
799 800 801
	if (root->fs_info->sb->s_flags & MS_RDONLY)
		return -EROFS;

802 803 804
	if (!capable(CAP_SYS_ADMIN))
		return -EPERM;

L
Li Zefan 已提交
805 806 807
	vol_args = memdup_user(arg, sizeof(*vol_args));
	if (IS_ERR(vol_args))
		return PTR_ERR(vol_args);
808 809

	vol_args->name[BTRFS_PATH_NAME_MAX] = '\0';
C
Christoph Hellwig 已提交
810

811
	mutex_lock(&root->fs_info->volume_mutex);
C
Christoph Hellwig 已提交
812 813 814 815 816 817 818 819
	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);
820 821
		printk(KERN_INFO "resizing devid %llu\n",
		       (unsigned long long)devid);
C
Christoph Hellwig 已提交
822
	}
Y
Yan Zheng 已提交
823
	device = btrfs_find_device(root, devid, NULL, NULL);
C
Christoph Hellwig 已提交
824
	if (!device) {
825 826
		printk(KERN_INFO "resizer unable to find device %llu\n",
		       (unsigned long long)devid);
C
Christoph Hellwig 已提交
827 828 829 830 831 832 833 834 835 836 837 838 839
		ret = -EINVAL;
		goto out_unlock;
	}
	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
840
		new_size = memparse(sizestr, NULL);
C
Christoph Hellwig 已提交
841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874
		if (new_size == 0) {
			ret = -EINVAL;
			goto out_unlock;
		}
	}

	old_size = device->total_bytes;

	if (mod < 0) {
		if (new_size > old_size) {
			ret = -EINVAL;
			goto out_unlock;
		}
		new_size = old_size - new_size;
	} else if (mod > 0) {
		new_size = old_size + new_size;
	}

	if (new_size < 256 * 1024 * 1024) {
		ret = -EINVAL;
		goto out_unlock;
	}
	if (new_size > device->bdev->bd_inode->i_size) {
		ret = -EFBIG;
		goto out_unlock;
	}

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

	printk(KERN_INFO "new size for %s is %llu\n",
		device->name, (unsigned long long)new_size);

	if (new_size > old_size) {
875
		trans = btrfs_start_transaction(root, 0);
C
Christoph Hellwig 已提交
876 877 878 879 880 881 882
		ret = btrfs_grow_device(trans, device, new_size);
		btrfs_commit_transaction(trans, root);
	} else {
		ret = btrfs_shrink_device(device, new_size);
	}

out_unlock:
883
	mutex_unlock(&root->fs_info->volume_mutex);
C
Christoph Hellwig 已提交
884 885 886 887
	kfree(vol_args);
	return ret;
}

S
Sage Weil 已提交
888 889 890 891 892
static noinline int btrfs_ioctl_snap_create_transid(struct file *file,
						    char *name,
						    unsigned long fd,
						    int subvol,
						    u64 *transid)
C
Christoph Hellwig 已提交
893
{
894
	struct btrfs_root *root = BTRFS_I(fdentry(file)->d_inode)->root;
895
	struct file *src_file;
C
Christoph Hellwig 已提交
896
	int namelen;
897
	int ret = 0;
C
Christoph Hellwig 已提交
898

Y
Yan Zheng 已提交
899 900 901
	if (root->fs_info->sb->s_flags & MS_RDONLY)
		return -EROFS;

S
Sage Weil 已提交
902 903
	namelen = strlen(name);
	if (strchr(name, '/')) {
C
Christoph Hellwig 已提交
904 905 906 907
		ret = -EINVAL;
		goto out;
	}

908
	if (subvol) {
S
Sage Weil 已提交
909 910
		ret = btrfs_mksubvol(&file->f_path, name, namelen,
				     NULL, transid);
911
	} else {
912
		struct inode *src_inode;
S
Sage Weil 已提交
913
		src_file = fget(fd);
914 915 916 917 918 919 920
		if (!src_file) {
			ret = -EINVAL;
			goto out;
		}

		src_inode = src_file->f_path.dentry->d_inode;
		if (src_inode->i_sb != file->f_path.dentry->d_inode->i_sb) {
C
Chris Mason 已提交
921 922
			printk(KERN_INFO "btrfs: Snapshot src from "
			       "another FS\n");
923 924 925 926
			ret = -EINVAL;
			fput(src_file);
			goto out;
		}
S
Sage Weil 已提交
927 928 929
		ret = btrfs_mksubvol(&file->f_path, name, namelen,
				     BTRFS_I(src_inode)->root,
				     transid);
930
		fput(src_file);
931
	}
C
Christoph Hellwig 已提交
932
out:
S
Sage Weil 已提交
933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973
	return ret;
}

static noinline int btrfs_ioctl_snap_create(struct file *file,
					    void __user *arg, int subvol,
					    int async)
{
	struct btrfs_ioctl_vol_args *vol_args = NULL;
	struct btrfs_ioctl_async_vol_args *async_vol_args = NULL;
	char *name;
	u64 fd;
	u64 transid = 0;
	int ret;

	if (async) {
		async_vol_args = memdup_user(arg, sizeof(*async_vol_args));
		if (IS_ERR(async_vol_args))
			return PTR_ERR(async_vol_args);

		name = async_vol_args->name;
		fd = async_vol_args->fd;
		async_vol_args->name[BTRFS_SNAPSHOT_NAME_MAX] = '\0';
	} else {
		vol_args = memdup_user(arg, sizeof(*vol_args));
		if (IS_ERR(vol_args))
			return PTR_ERR(vol_args);
		name = vol_args->name;
		fd = vol_args->fd;
		vol_args->name[BTRFS_PATH_NAME_MAX] = '\0';
	}

	ret = btrfs_ioctl_snap_create_transid(file, name, fd,
					      subvol, &transid);

	if (!ret && async) {
		if (copy_to_user(arg +
				offsetof(struct btrfs_ioctl_async_vol_args,
				transid), &transid, sizeof(transid)))
			return -EFAULT;
	}

C
Christoph Hellwig 已提交
974
	kfree(vol_args);
S
Sage Weil 已提交
975 976
	kfree(async_vol_args);

C
Christoph Hellwig 已提交
977 978 979
	return ret;
}

980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015
/*
 * 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;
}

1016 1017 1018
static noinline int key_in_sk(struct btrfs_key *key,
			      struct btrfs_ioctl_search_key *sk)
{
1019 1020 1021 1022 1023 1024 1025 1026 1027
	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)
1028
		return 0;
1029 1030 1031 1032 1033 1034 1035

	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)
1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102
		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 found = 0;
	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;
		}
1103
		found++;
1104 1105 1106 1107 1108

		if (*num_found >= sk->nr_items)
			break;
	}
advance_key:
1109 1110
	ret = 0;
	if (key->offset < (u64)-1 && key->offset < sk->max_offset)
1111
		key->offset++;
1112 1113
	else if (key->type < (u8)-1 && key->type < sk->max_type) {
		key->offset = 0;
1114
		key->type++;
1115 1116 1117
	} else if (key->objectid < (u64)-1 && key->objectid < sk->max_objectid) {
		key->offset = 0;
		key->type = 0;
1118
		key->objectid++;
1119 1120
	} else
		ret = 1;
1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200
overflow:
	*num_found += found;
	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);
		btrfs_release_path(root, path);
		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
1201 1202 1203
	args = memdup_user(argp, sizeof(*args));
	if (IS_ERR(args))
		return PTR_ERR(args);
1204 1205 1206 1207 1208 1209 1210 1211 1212

	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;
}

1213
/*
1214 1215 1216
 * Search INODE_REFs to identify path name of 'dirid' directory
 * in a 'tree_id' tree. and sets path name to 'name'.
 */
1217 1218 1219 1220 1221
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;
1222
	char *ptr;
1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239
	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;

1240
	ptr = &name[BTRFS_INO_LOOKUP_PATH_MAX];
1241 1242 1243 1244 1245 1246 1247

	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);
1248 1249
		ret = -ENOENT;
		goto out;
1250 1251 1252 1253
	}

	key.objectid = dirid;
	key.type = BTRFS_INODE_REF_KEY;
1254
	key.offset = (u64)-1;
1255 1256 1257 1258 1259 1260 1261 1262

	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];
1263 1264
		if (ret > 0 && slot > 0)
			slot--;
1265 1266 1267
		btrfs_item_key_to_cpu(l, &key, slot);

		if (ret > 0 && (key.objectid != dirid ||
1268 1269
				key.type != BTRFS_INODE_REF_KEY)) {
			ret = -ENOENT;
1270
			goto out;
1271
		}
1272 1273 1274 1275 1276

		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;
1277
		if (ptr < name)
1278 1279 1280 1281 1282 1283 1284 1285 1286 1287
			goto out;

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

		if (key.offset == BTRFS_FIRST_FREE_OBJECTID)
			break;

		btrfs_release_path(root, path);
		key.objectid = key.offset;
1288
		key.offset = (u64)-1;
1289 1290 1291
		dirid = key.objectid;

	}
1292
	if (ptr < name)
1293
		goto out;
1294
	memcpy(name, ptr, total_len);
1295 1296 1297 1298
	name[total_len]='\0';
	ret = 0;
out:
	btrfs_free_path(path);
1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311
	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
1312 1313 1314
	args = memdup_user(argp, sizeof(*args));
	if (IS_ERR(args))
		return PTR_ERR(args);
1315

1316 1317
	inode = fdentry(file)->d_inode;

1318 1319 1320
	if (args->treeid == 0)
		args->treeid = BTRFS_I(inode)->root->root_key.objectid;

1321 1322 1323 1324 1325 1326 1327 1328
	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);
1329 1330 1331
	return ret;
}

1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375
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;
	}

	err = mnt_want_write(file->f_path.mnt);
	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;
1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415
	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;
	}

1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431
	if (inode->i_ino != BTRFS_FIRST_FREE_OBJECTID) {
		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;

1432 1433 1434
	trans = btrfs_start_transaction(root, 0);
	if (IS_ERR(trans)) {
		err = PTR_ERR(trans);
1435
		goto out_up_write;
1436 1437 1438
	}
	trans->block_rsv = &root->fs_info->global_block_rsv;

1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451
	ret = btrfs_unlink_subvol(trans, root, dir,
				dest->root_key.objectid,
				dentry->d_name.name,
				dentry->d_name.len);
	BUG_ON(ret);

	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);

1452 1453 1454 1455 1456 1457
	if (!xchg(&dest->orphan_item_inserted, 1)) {
		ret = btrfs_insert_orphan_item(trans,
					root->fs_info->tree_root,
					dest->root_key.objectid);
		BUG_ON(ret);
	}
1458

S
Sage Weil 已提交
1459
	ret = btrfs_end_transaction(trans, root);
1460 1461 1462 1463 1464 1465 1466
	BUG_ON(ret);
	inode->i_flags |= S_DEAD;
out_up_write:
	up_write(&root->fs_info->subvol_sem);
out_unlock:
	mutex_unlock(&inode->i_mutex);
	if (!err) {
1467
		shrink_dcache_sb(root->fs_info->sb);
1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480
		btrfs_invalidate_inodes(dest);
		d_delete(dentry);
	}
out_dput:
	dput(dentry);
out_unlock_dir:
	mutex_unlock(&dir->i_mutex);
	mnt_drop_write(file->f_path.mnt);
out:
	kfree(vol_args);
	return err;
}

C
Chris Mason 已提交
1481
static int btrfs_ioctl_defrag(struct file *file, void __user *argp)
C
Christoph Hellwig 已提交
1482 1483 1484
{
	struct inode *inode = fdentry(file)->d_inode;
	struct btrfs_root *root = BTRFS_I(inode)->root;
C
Chris Mason 已提交
1485
	struct btrfs_ioctl_defrag_range_args *range;
Y
Yan Zheng 已提交
1486 1487 1488 1489 1490
	int ret;

	ret = mnt_want_write(file->f_path.mnt);
	if (ret)
		return ret;
C
Christoph Hellwig 已提交
1491 1492 1493

	switch (inode->i_mode & S_IFMT) {
	case S_IFDIR:
1494 1495 1496 1497
		if (!capable(CAP_SYS_ADMIN)) {
			ret = -EPERM;
			goto out;
		}
1498 1499 1500 1501
		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
1502 1503
		break;
	case S_IFREG:
1504 1505 1506 1507
		if (!(file->f_mode & FMODE_WRITE)) {
			ret = -EINVAL;
			goto out;
		}
C
Chris Mason 已提交
1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519

		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);
1520
				goto out;
C
Chris Mason 已提交
1521 1522 1523 1524 1525 1526 1527 1528 1529 1530
			}
			/* 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;
		}
1531
		ret = btrfs_defrag_file(file, range);
C
Chris Mason 已提交
1532
		kfree(range);
C
Christoph Hellwig 已提交
1533
		break;
1534 1535
	default:
		ret = -EINVAL;
C
Christoph Hellwig 已提交
1536
	}
1537
out:
1538
	mnt_drop_write(file->f_path.mnt);
1539
	return ret;
C
Christoph Hellwig 已提交
1540 1541
}

1542
static long btrfs_ioctl_add_dev(struct btrfs_root *root, void __user *arg)
C
Christoph Hellwig 已提交
1543 1544 1545 1546
{
	struct btrfs_ioctl_vol_args *vol_args;
	int ret;

1547 1548 1549
	if (!capable(CAP_SYS_ADMIN))
		return -EPERM;

L
Li Zefan 已提交
1550 1551 1552
	vol_args = memdup_user(arg, sizeof(*vol_args));
	if (IS_ERR(vol_args))
		return PTR_ERR(vol_args);
C
Christoph Hellwig 已提交
1553

1554
	vol_args->name[BTRFS_PATH_NAME_MAX] = '\0';
C
Christoph Hellwig 已提交
1555 1556 1557 1558 1559 1560
	ret = btrfs_init_new_device(root, vol_args->name);

	kfree(vol_args);
	return ret;
}

1561
static long btrfs_ioctl_rm_dev(struct btrfs_root *root, void __user *arg)
C
Christoph Hellwig 已提交
1562 1563 1564 1565
{
	struct btrfs_ioctl_vol_args *vol_args;
	int ret;

1566 1567 1568
	if (!capable(CAP_SYS_ADMIN))
		return -EPERM;

Y
Yan Zheng 已提交
1569 1570 1571
	if (root->fs_info->sb->s_flags & MS_RDONLY)
		return -EROFS;

L
Li Zefan 已提交
1572 1573 1574
	vol_args = memdup_user(arg, sizeof(*vol_args));
	if (IS_ERR(vol_args))
		return PTR_ERR(vol_args);
C
Christoph Hellwig 已提交
1575

1576
	vol_args->name[BTRFS_PATH_NAME_MAX] = '\0';
C
Christoph Hellwig 已提交
1577 1578 1579 1580 1581 1582
	ret = btrfs_rm_device(root, vol_args->name);

	kfree(vol_args);
	return ret;
}

1583 1584
static noinline long btrfs_ioctl_clone(struct file *file, unsigned long srcfd,
				       u64 off, u64 olen, u64 destoff)
C
Christoph Hellwig 已提交
1585 1586 1587 1588 1589 1590 1591 1592
{
	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
1593 1594
	char *buf;
	struct btrfs_key key;
C
Christoph Hellwig 已提交
1595 1596
	u32 nritems;
	int slot;
Y
Yan Zheng 已提交
1597
	int ret;
1598 1599 1600
	u64 len = olen;
	u64 bs = root->fs_info->sb->s_blocksize;
	u64 hint_byte;
1601

1602 1603 1604 1605 1606 1607 1608 1609 1610 1611
	/*
	 * 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)?
	 */

1612
	/* the destination must be opened for writing */
1613
	if (!(file->f_mode & FMODE_WRITE) || (file->f_flags & O_APPEND))
1614 1615
		return -EINVAL;

Y
Yan Zheng 已提交
1616 1617 1618 1619
	ret = mnt_want_write(file->f_path.mnt);
	if (ret)
		return ret;

1620
	src_file = fget(srcfd);
1621 1622 1623 1624
	if (!src_file) {
		ret = -EBADF;
		goto out_drop_write;
	}
1625

C
Christoph Hellwig 已提交
1626 1627
	src = src_file->f_dentry->d_inode;

1628 1629 1630 1631
	ret = -EINVAL;
	if (src == inode)
		goto out_fput;

1632 1633 1634 1635
	/* the src must be open for reading */
	if (!(src_file->f_mode & FMODE_READ))
		goto out_fput;

Y
Yan Zheng 已提交
1636 1637 1638 1639
	ret = -EISDIR;
	if (S_ISDIR(src->i_mode) || S_ISDIR(inode->i_mode))
		goto out_fput;

C
Christoph Hellwig 已提交
1640
	ret = -EXDEV;
Y
Yan Zheng 已提交
1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651
	if (src->i_sb != inode->i_sb || BTRFS_I(src)->root != root)
		goto out_fput;

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

	path = btrfs_alloc_path();
	if (!path) {
		vfree(buf);
C
Christoph Hellwig 已提交
1652
		goto out_fput;
Y
Yan Zheng 已提交
1653 1654
	}
	path->reada = 2;
C
Christoph Hellwig 已提交
1655 1656

	if (inode < src) {
1657 1658
		mutex_lock_nested(&inode->i_mutex, I_MUTEX_PARENT);
		mutex_lock_nested(&src->i_mutex, I_MUTEX_CHILD);
C
Christoph Hellwig 已提交
1659
	} else {
1660 1661
		mutex_lock_nested(&src->i_mutex, I_MUTEX_PARENT);
		mutex_lock_nested(&inode->i_mutex, I_MUTEX_CHILD);
C
Christoph Hellwig 已提交
1662 1663
	}

1664 1665
	/* determine range to clone */
	ret = -EINVAL;
1666
	if (off + len > src->i_size || off + len < off)
C
Christoph Hellwig 已提交
1667
		goto out_unlock;
1668 1669 1670 1671
	if (len == 0)
		olen = len = src->i_size - off;
	/* if we extend to eof, continue to block boundary */
	if (off + len == src->i_size)
1672
		len = ALIGN(src->i_size, bs) - off;
1673 1674

	/* verify the end result is block aligned */
1675 1676
	if (!IS_ALIGNED(off, bs) || !IS_ALIGNED(off + len, bs) ||
	    !IS_ALIGNED(destoff, bs))
1677 1678
		goto out_unlock;

C
Christoph Hellwig 已提交
1679 1680 1681
	/* do any pending delalloc/csum calc on src, one way or
	   another, and lock file content */
	while (1) {
Z
Zheng Yan 已提交
1682
		struct btrfs_ordered_extent *ordered;
1683
		lock_extent(&BTRFS_I(src)->io_tree, off, off+len, GFP_NOFS);
1684 1685 1686 1687
		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
1688
			break;
1689
		unlock_extent(&BTRFS_I(src)->io_tree, off, off+len, GFP_NOFS);
Y
Yan Zheng 已提交
1690 1691
		if (ordered)
			btrfs_put_ordered_extent(ordered);
1692
		btrfs_wait_ordered_range(src, off, len);
C
Christoph Hellwig 已提交
1693 1694
	}

1695
	/* clone data */
C
Christoph Hellwig 已提交
1696
	key.objectid = src->i_ino;
Y
Yan Zheng 已提交
1697 1698
	key.type = BTRFS_EXTENT_DATA_KEY;
	key.offset = 0;
C
Christoph Hellwig 已提交
1699 1700 1701 1702 1703 1704

	while (1) {
		/*
		 * note the key will change type as we walk through the
		 * tree.
		 */
1705
		ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
C
Christoph Hellwig 已提交
1706 1707 1708
		if (ret < 0)
			goto out;

Y
Yan Zheng 已提交
1709 1710
		nritems = btrfs_header_nritems(path->nodes[0]);
		if (path->slots[0] >= nritems) {
C
Christoph Hellwig 已提交
1711 1712 1713 1714 1715
			ret = btrfs_next_leaf(root, path);
			if (ret < 0)
				goto out;
			if (ret > 0)
				break;
Y
Yan Zheng 已提交
1716
			nritems = btrfs_header_nritems(path->nodes[0]);
C
Christoph Hellwig 已提交
1717 1718 1719 1720
		}
		leaf = path->nodes[0];
		slot = path->slots[0];

Y
Yan Zheng 已提交
1721
		btrfs_item_key_to_cpu(leaf, &key, slot);
1722
		if (btrfs_key_type(&key) > BTRFS_EXTENT_DATA_KEY ||
C
Christoph Hellwig 已提交
1723 1724 1725
		    key.objectid != src->i_ino)
			break;

1726 1727 1728
		if (btrfs_key_type(&key) == BTRFS_EXTENT_DATA_KEY) {
			struct btrfs_file_extent_item *extent;
			int type;
Z
Zheng Yan 已提交
1729 1730
			u32 size;
			struct btrfs_key new_key;
1731 1732 1733
			u64 disko = 0, diskl = 0;
			u64 datao = 0, datal = 0;
			u8 comp;
1734
			u64 endoff;
Z
Zheng Yan 已提交
1735 1736 1737 1738 1739

			size = btrfs_item_size_nr(leaf, slot);
			read_extent_buffer(leaf, buf,
					   btrfs_item_ptr_offset(leaf, slot),
					   size);
1740 1741 1742 1743 1744

			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);
1745 1746
			if (type == BTRFS_FILE_EXTENT_REG ||
			    type == BTRFS_FILE_EXTENT_PREALLOC) {
C
Chris Mason 已提交
1747 1748 1749 1750
				disko = btrfs_file_extent_disk_bytenr(leaf,
								      extent);
				diskl = btrfs_file_extent_disk_num_bytes(leaf,
								 extent);
1751
				datao = btrfs_file_extent_offset(leaf, extent);
C
Chris Mason 已提交
1752 1753
				datal = btrfs_file_extent_num_bytes(leaf,
								    extent);
1754 1755 1756 1757 1758
			} else if (type == BTRFS_FILE_EXTENT_INLINE) {
				/* take upper bound, may be compressed */
				datal = btrfs_file_extent_ram_bytes(leaf,
								    extent);
			}
Z
Zheng Yan 已提交
1759 1760
			btrfs_release_path(root, path);

1761
			if (key.offset + datal <= off ||
1762 1763 1764
			    key.offset >= off+len)
				goto next;

Z
Zheng Yan 已提交
1765 1766
			memcpy(&new_key, &key, sizeof(new_key));
			new_key.objectid = inode->i_ino;
1767
			new_key.offset = key.offset + destoff - off;
Z
Zheng Yan 已提交
1768

1769 1770 1771 1772 1773 1774
			trans = btrfs_start_transaction(root, 1);
			if (IS_ERR(trans)) {
				ret = PTR_ERR(trans);
				goto out;
			}

1775 1776
			if (type == BTRFS_FILE_EXTENT_REG ||
			    type == BTRFS_FILE_EXTENT_PREALLOC) {
1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790
				if (off > key.offset) {
					datao += off - key.offset;
					datal -= off - key.offset;
				}

				if (key.offset + datal > off + len)
					datal = off + len - key.offset;

				ret = btrfs_drop_extents(trans, inode,
							 new_key.offset,
							 new_key.offset + datal,
							 &hint_byte, 1);
				BUG_ON(ret);

1791 1792
				ret = btrfs_insert_empty_item(trans, root, path,
							      &new_key, size);
1793
				BUG_ON(ret);
1794 1795 1796 1797

				leaf = path->nodes[0];
				slot = path->slots[0];
				write_extent_buffer(leaf, buf,
Z
Zheng Yan 已提交
1798 1799
					    btrfs_item_ptr_offset(leaf, slot),
					    size);
Y
Yan Zheng 已提交
1800

1801
				extent = btrfs_item_ptr(leaf, slot,
C
Christoph Hellwig 已提交
1802
						struct btrfs_file_extent_item);
1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813

				/* 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
1814
					ret = btrfs_inc_extent_ref(trans, root,
1815 1816 1817 1818
							disko, diskl, 0,
							root->root_key.objectid,
							inode->i_ino,
							new_key.offset - datao);
Z
Zheng Yan 已提交
1819
					BUG_ON(ret);
C
Christoph Hellwig 已提交
1820
				}
1821 1822 1823 1824 1825 1826 1827
			} 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
1828

1829 1830
				if (key.offset + datal > off+len)
					trim = key.offset + datal - (off+len);
C
Chris Mason 已提交
1831

1832 1833
				if (comp && (skip || trim)) {
					ret = -EINVAL;
1834
					btrfs_end_transaction(trans, root);
1835 1836 1837 1838
					goto out;
				}
				size -= skip + trim;
				datal -= skip + trim;
1839 1840 1841 1842 1843 1844 1845

				ret = btrfs_drop_extents(trans, inode,
							 new_key.offset,
							 new_key.offset + datal,
							 &hint_byte, 1);
				BUG_ON(ret);

1846 1847
				ret = btrfs_insert_empty_item(trans, root, path,
							      &new_key, size);
1848
				BUG_ON(ret);
1849 1850

				if (skip) {
C
Chris Mason 已提交
1851 1852
					u32 start =
					  btrfs_file_extent_calc_inline_size(0);
1853 1854 1855 1856 1857 1858 1859 1860 1861 1862
					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
1863
			}
1864 1865

			btrfs_mark_buffer_dirty(leaf);
1866
			btrfs_release_path(root, path);
1867

1868
			inode->i_mtime = inode->i_ctime = CURRENT_TIME;
1869 1870 1871 1872 1873 1874 1875

			/*
			 * 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;
1876 1877
			if (endoff > destoff+olen)
				endoff = destoff+olen;
1878 1879 1880
			if (endoff > inode->i_size)
				btrfs_i_size_write(inode, endoff);

1881 1882 1883 1884 1885
			BTRFS_I(inode)->flags = BTRFS_I(src)->flags;
			ret = btrfs_update_inode(trans, root, inode);
			BUG_ON(ret);
			btrfs_end_transaction(trans, root);
		}
C
Chris Mason 已提交
1886
next:
Z
Zheng Yan 已提交
1887
		btrfs_release_path(root, path);
C
Christoph Hellwig 已提交
1888 1889 1890 1891
		key.offset++;
	}
	ret = 0;
out:
Y
Yan Zheng 已提交
1892
	btrfs_release_path(root, path);
1893
	unlock_extent(&BTRFS_I(src)->io_tree, off, off+len, GFP_NOFS);
C
Christoph Hellwig 已提交
1894 1895 1896
out_unlock:
	mutex_unlock(&src->i_mutex);
	mutex_unlock(&inode->i_mutex);
Y
Yan Zheng 已提交
1897 1898
	vfree(buf);
	btrfs_free_path(path);
C
Christoph Hellwig 已提交
1899 1900
out_fput:
	fput(src_file);
1901 1902
out_drop_write:
	mnt_drop_write(file->f_path.mnt);
C
Christoph Hellwig 已提交
1903 1904 1905
	return ret;
}

1906
static long btrfs_ioctl_clone_range(struct file *file, void __user *argp)
1907 1908 1909
{
	struct btrfs_ioctl_clone_range_args args;

1910
	if (copy_from_user(&args, argp, sizeof(args)))
1911 1912 1913 1914 1915
		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
1916 1917 1918 1919 1920 1921
/*
 * 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.
 */
1922
static long btrfs_ioctl_trans_start(struct file *file)
C
Christoph Hellwig 已提交
1923 1924 1925 1926
{
	struct inode *inode = fdentry(file)->d_inode;
	struct btrfs_root *root = BTRFS_I(inode)->root;
	struct btrfs_trans_handle *trans;
1927
	int ret;
C
Christoph Hellwig 已提交
1928

1929
	ret = -EPERM;
1930
	if (!capable(CAP_SYS_ADMIN))
1931
		goto out;
1932

1933 1934
	ret = -EINPROGRESS;
	if (file->private_data)
C
Christoph Hellwig 已提交
1935
		goto out;
1936

Y
Yan Zheng 已提交
1937 1938 1939 1940
	ret = mnt_want_write(file->f_path.mnt);
	if (ret)
		goto out;

1941 1942 1943 1944
	mutex_lock(&root->fs_info->trans_mutex);
	root->fs_info->open_ioctl_trans++;
	mutex_unlock(&root->fs_info->trans_mutex);

1945
	ret = -ENOMEM;
1946
	trans = btrfs_start_ioctl_transaction(root, 0);
1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957
	if (!trans)
		goto out_drop;

	file->private_data = trans;
	return 0;

out_drop:
	mutex_lock(&root->fs_info->trans_mutex);
	root->fs_info->open_ioctl_trans--;
	mutex_unlock(&root->fs_info->trans_mutex);
	mnt_drop_write(file->f_path.mnt);
C
Christoph Hellwig 已提交
1958 1959 1960 1961
out:
	return ret;
}

1962 1963 1964 1965 1966 1967 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 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
static long btrfs_ioctl_default_subvol(struct file *file, void __user *argp)
{
	struct inode *inode = fdentry(file)->d_inode;
	struct btrfs_root *root = BTRFS_I(inode)->root;
	struct btrfs_root *new_root;
	struct btrfs_dir_item *di;
	struct btrfs_trans_handle *trans;
	struct btrfs_path *path;
	struct btrfs_key location;
	struct btrfs_disk_key disk_key;
	struct btrfs_super_block *disk_super;
	u64 features;
	u64 objectid = 0;
	u64 dir_id;

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

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

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

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

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

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

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

	trans = btrfs_start_transaction(root, 1);
	if (!trans) {
		btrfs_free_path(path);
		return -ENOMEM;
	}

	dir_id = btrfs_super_root_dir(&root->fs_info->super_copy);
	di = btrfs_lookup_dir_item(trans, root->fs_info->tree_root, path,
				   dir_id, "default", 7, 1);
2011
	if (IS_ERR_OR_NULL(di)) {
2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034
		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);

	disk_super = &root->fs_info->super_copy;
	features = btrfs_super_incompat_flags(disk_super);
	if (!(features & BTRFS_FEATURE_INCOMPAT_DEFAULT_SUBVOL)) {
		features |= BTRFS_FEATURE_INCOMPAT_DEFAULT_SUBVOL;
		btrfs_set_super_incompat_flags(disk_super, features);
	}
	btrfs_end_transaction(trans, root);

	return 0;
}

2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050
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
2051 2052 2053 2054 2055
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;
2056 2057
	struct btrfs_ioctl_space_info *dest_orig;
	struct btrfs_ioctl_space_info *user_dest;
J
Josef Bacik 已提交
2058
	struct btrfs_space_info *info;
2059 2060 2061 2062 2063
	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;
2064
	int alloc_size;
J
Josef Bacik 已提交
2065
	int ret = 0;
2066
	int slot_count = 0;
2067
	int i, c;
J
Josef Bacik 已提交
2068 2069 2070 2071 2072 2073

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

2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097
	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);
	}
2098 2099 2100 2101 2102 2103

	/* space_slots == 0 means they are asking for a count */
	if (space_args.space_slots == 0) {
		space_args.total_spaces = slot_count;
		goto out;
	}
2104 2105 2106

	slot_count = min_t(int, space_args.space_slots, slot_count);

2107
	alloc_size = sizeof(*dest) * slot_count;
2108

2109 2110 2111 2112 2113 2114
	/* 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
2115
	space_args.total_spaces = 0;
2116 2117 2118 2119
	dest = kmalloc(alloc_size, GFP_NOFS);
	if (!dest)
		return -ENOMEM;
	dest_orig = dest;
J
Josef Bacik 已提交
2120

2121
	/* now we have a buffer to copy into */
2122 2123 2124 2125 2126 2127 2128 2129 2130 2131 2132 2133 2134
	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();
2135

2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148
		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++;
			}
		}
		up_read(&info->groups_sem);
J
Josef Bacik 已提交
2149 2150
	}

2151 2152 2153 2154 2155 2156 2157 2158 2159
	user_dest = (struct btrfs_ioctl_space_info *)
		(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
2160 2161 2162 2163 2164
		ret = -EFAULT;

	return ret;
}

C
Christoph Hellwig 已提交
2165 2166 2167 2168 2169 2170 2171 2172 2173 2174 2175 2176 2177
/*
 * 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;
2178 2179
	if (!trans)
		return -EINVAL;
C
Christoph Hellwig 已提交
2180
	file->private_data = NULL;
2181

2182 2183
	btrfs_end_transaction(trans, root);

2184 2185 2186 2187
	mutex_lock(&root->fs_info->trans_mutex);
	root->fs_info->open_ioctl_trans--;
	mutex_unlock(&root->fs_info->trans_mutex);

2188
	mnt_drop_write(file->f_path.mnt);
2189
	return 0;
C
Christoph Hellwig 已提交
2190 2191
}

2192 2193 2194 2195 2196 2197 2198 2199 2200 2201 2202 2203 2204 2205 2206 2207 2208 2209 2210 2211 2212 2213 2214 2215 2216 2217 2218 2219 2220 2221
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;

	trans = btrfs_start_transaction(root, 0);
	transid = trans->transid;
	btrfs_commit_transaction_async(trans, root, 0);

	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);
}

C
Christoph Hellwig 已提交
2222 2223 2224 2225
long btrfs_ioctl(struct file *file, unsigned int
		cmd, unsigned long arg)
{
	struct btrfs_root *root = BTRFS_I(fdentry(file)->d_inode)->root;
2226
	void __user *argp = (void __user *)arg;
C
Christoph Hellwig 已提交
2227 2228

	switch (cmd) {
2229 2230 2231 2232 2233 2234
	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);
C
Christoph Hellwig 已提交
2235
	case BTRFS_IOC_SNAP_CREATE:
S
Sage Weil 已提交
2236 2237 2238
		return btrfs_ioctl_snap_create(file, argp, 0, 0);
	case BTRFS_IOC_SNAP_CREATE_ASYNC:
		return btrfs_ioctl_snap_create(file, argp, 0, 1);
2239
	case BTRFS_IOC_SUBVOL_CREATE:
S
Sage Weil 已提交
2240
		return btrfs_ioctl_snap_create(file, argp, 1, 0);
2241 2242
	case BTRFS_IOC_SNAP_DESTROY:
		return btrfs_ioctl_snap_destroy(file, argp);
2243 2244
	case BTRFS_IOC_DEFAULT_SUBVOL:
		return btrfs_ioctl_default_subvol(file, argp);
C
Christoph Hellwig 已提交
2245
	case BTRFS_IOC_DEFRAG:
C
Chris Mason 已提交
2246 2247 2248
		return btrfs_ioctl_defrag(file, NULL);
	case BTRFS_IOC_DEFRAG_RANGE:
		return btrfs_ioctl_defrag(file, argp);
C
Christoph Hellwig 已提交
2249
	case BTRFS_IOC_RESIZE:
2250
		return btrfs_ioctl_resize(root, argp);
C
Christoph Hellwig 已提交
2251
	case BTRFS_IOC_ADD_DEV:
2252
		return btrfs_ioctl_add_dev(root, argp);
C
Christoph Hellwig 已提交
2253
	case BTRFS_IOC_RM_DEV:
2254
		return btrfs_ioctl_rm_dev(root, argp);
C
Christoph Hellwig 已提交
2255 2256 2257
	case BTRFS_IOC_BALANCE:
		return btrfs_balance(root->fs_info->dev_root);
	case BTRFS_IOC_CLONE:
2258 2259
		return btrfs_ioctl_clone(file, arg, 0, 0, 0);
	case BTRFS_IOC_CLONE_RANGE:
2260
		return btrfs_ioctl_clone_range(file, argp);
C
Christoph Hellwig 已提交
2261 2262 2263 2264
	case BTRFS_IOC_TRANS_START:
		return btrfs_ioctl_trans_start(file);
	case BTRFS_IOC_TRANS_END:
		return btrfs_ioctl_trans_end(file);
2265 2266 2267 2268
	case BTRFS_IOC_TREE_SEARCH:
		return btrfs_ioctl_tree_search(file, argp);
	case BTRFS_IOC_INO_LOOKUP:
		return btrfs_ioctl_ino_lookup(file, argp);
J
Josef Bacik 已提交
2269 2270
	case BTRFS_IOC_SPACE_INFO:
		return btrfs_ioctl_space_info(root, argp);
C
Christoph Hellwig 已提交
2271 2272 2273
	case BTRFS_IOC_SYNC:
		btrfs_sync_fs(file->f_dentry->d_sb, 1);
		return 0;
2274 2275 2276 2277
	case BTRFS_IOC_START_SYNC:
		return btrfs_ioctl_start_sync(file, argp);
	case BTRFS_IOC_WAIT_SYNC:
		return btrfs_ioctl_wait_sync(file, argp);
C
Christoph Hellwig 已提交
2278 2279 2280 2281
	}

	return -ENOTTY;
}