copy_up.c 19.9 KB
Newer Older
M
Miklos Szeredi 已提交
1 2 3 4 5 6 7 8 9
/*
 *
 * Copyright (C) 2011 Novell Inc.
 *
 * This program is free software; you can redistribute it and/or modify it
 * under the terms of the GNU General Public License version 2 as published by
 * the Free Software Foundation.
 */

10
#include <linux/module.h>
M
Miklos Szeredi 已提交
11 12 13 14 15 16 17
#include <linux/fs.h>
#include <linux/slab.h>
#include <linux/file.h>
#include <linux/splice.h>
#include <linux/xattr.h>
#include <linux/security.h>
#include <linux/uaccess.h>
18
#include <linux/sched/signal.h>
19
#include <linux/cred.h>
M
Miklos Szeredi 已提交
20
#include <linux/namei.h>
21 22
#include <linux/fdtable.h>
#include <linux/ratelimit.h>
23
#include <linux/exportfs.h>
M
Miklos Szeredi 已提交
24 25 26 27
#include "overlayfs.h"

#define OVL_COPY_UP_CHUNK_SIZE (1 << 20)

28
static int ovl_ccup_set(const char *buf, const struct kernel_param *param)
29
{
30
	pr_warn("overlayfs: \"check_copy_up\" module option is obsolete\n");
31 32 33
	return 0;
}

34
static int ovl_ccup_get(char *buf, const struct kernel_param *param)
35
{
36
	return sprintf(buf, "N\n");
37 38
}

39 40 41
module_param_call(check_copy_up, ovl_ccup_set, ovl_ccup_get, NULL, 0644);
MODULE_PARM_DESC(ovl_check_copy_up, "Obsolete; does nothing");

M
Miklos Szeredi 已提交
42 43
int ovl_copy_xattr(struct dentry *old, struct dentry *new)
{
44 45 46
	ssize_t list_size, size, value_size = 0;
	char *buf, *name, *value = NULL;
	int uninitialized_var(error);
47
	size_t slen;
M
Miklos Szeredi 已提交
48

49 50
	if (!(old->d_inode->i_opflags & IOP_XATTR) ||
	    !(new->d_inode->i_opflags & IOP_XATTR))
M
Miklos Szeredi 已提交
51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66
		return 0;

	list_size = vfs_listxattr(old, NULL, 0);
	if (list_size <= 0) {
		if (list_size == -EOPNOTSUPP)
			return 0;
		return list_size;
	}

	buf = kzalloc(list_size, GFP_KERNEL);
	if (!buf)
		return -ENOMEM;

	list_size = vfs_listxattr(old, buf, list_size);
	if (list_size <= 0) {
		error = list_size;
67
		goto out;
M
Miklos Szeredi 已提交
68 69
	}

70 71 72 73 74 75 76 77 78 79
	for (name = buf; list_size; name += slen) {
		slen = strnlen(name, list_size) + 1;

		/* underlying fs providing us with an broken xattr list? */
		if (WARN_ON(slen > list_size)) {
			error = -EIO;
			break;
		}
		list_size -= slen;

M
Miklos Szeredi 已提交
80 81
		if (ovl_is_private_xattr(name))
			continue;
82 83 84 85 86
retry:
		size = vfs_getxattr(old, name, value, value_size);
		if (size == -ERANGE)
			size = vfs_getxattr(old, name, NULL, 0);

M
Miklos Szeredi 已提交
87
		if (size < 0) {
M
Miklos Szeredi 已提交
88
			error = size;
89
			break;
M
Miklos Szeredi 已提交
90
		}
91 92 93 94 95 96 97 98 99 100 101 102 103 104

		if (size > value_size) {
			void *new;

			new = krealloc(value, size, GFP_KERNEL);
			if (!new) {
				error = -ENOMEM;
				break;
			}
			value = new;
			value_size = size;
			goto retry;
		}

105 106 107 108 109 110 111
		error = security_inode_copy_up_xattr(name);
		if (error < 0 && error != -EOPNOTSUPP)
			break;
		if (error == 1) {
			error = 0;
			continue; /* Discard */
		}
M
Miklos Szeredi 已提交
112 113
		error = vfs_setxattr(new, name, value, size, 0);
		if (error)
114
			break;
M
Miklos Szeredi 已提交
115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132
	}
	kfree(value);
out:
	kfree(buf);
	return error;
}

static int ovl_copy_up_data(struct path *old, struct path *new, loff_t len)
{
	struct file *old_file;
	struct file *new_file;
	loff_t old_pos = 0;
	loff_t new_pos = 0;
	int error = 0;

	if (len == 0)
		return 0;

133
	old_file = ovl_path_open(old, O_LARGEFILE | O_RDONLY);
M
Miklos Szeredi 已提交
134 135 136
	if (IS_ERR(old_file))
		return PTR_ERR(old_file);

137
	new_file = ovl_path_open(new, O_LARGEFILE | O_WRONLY);
M
Miklos Szeredi 已提交
138 139 140 141 142
	if (IS_ERR(new_file)) {
		error = PTR_ERR(new_file);
		goto out_fput;
	}

143
	/* Try to use clone_file_range to clone up within the same fs */
144
	error = do_clone_file_range(old_file, 0, new_file, 0, len);
145 146 147 148 149
	if (!error)
		goto out;
	/* Couldn't clone, so now we try to copy the data */
	error = 0;

M
Miklos Szeredi 已提交
150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173
	/* FIXME: copy up sparse files efficiently */
	while (len) {
		size_t this_len = OVL_COPY_UP_CHUNK_SIZE;
		long bytes;

		if (len < this_len)
			this_len = len;

		if (signal_pending_state(TASK_KILLABLE, current)) {
			error = -EINTR;
			break;
		}

		bytes = do_splice_direct(old_file, &old_pos,
					 new_file, &new_pos,
					 this_len, SPLICE_F_MOVE);
		if (bytes <= 0) {
			error = bytes;
			break;
		}
		WARN_ON(old_pos != new_pos);

		len -= bytes;
	}
174
out:
M
Miklos Szeredi 已提交
175 176
	if (!error)
		error = vfs_fsync(new_file, 0);
M
Miklos Szeredi 已提交
177 178 179 180 181 182
	fput(new_file);
out_fput:
	fput(old_file);
	return error;
}

183 184 185 186 187 188 189 190 191 192
static int ovl_set_size(struct dentry *upperdentry, struct kstat *stat)
{
	struct iattr attr = {
		.ia_valid = ATTR_SIZE,
		.ia_size = stat->size,
	};

	return notify_change(upperdentry, &attr, NULL);
}

M
Miklos Szeredi 已提交
193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229
static int ovl_set_timestamps(struct dentry *upperdentry, struct kstat *stat)
{
	struct iattr attr = {
		.ia_valid =
		     ATTR_ATIME | ATTR_MTIME | ATTR_ATIME_SET | ATTR_MTIME_SET,
		.ia_atime = stat->atime,
		.ia_mtime = stat->mtime,
	};

	return notify_change(upperdentry, &attr, NULL);
}

int ovl_set_attr(struct dentry *upperdentry, struct kstat *stat)
{
	int err = 0;

	if (!S_ISLNK(stat->mode)) {
		struct iattr attr = {
			.ia_valid = ATTR_MODE,
			.ia_mode = stat->mode,
		};
		err = notify_change(upperdentry, &attr, NULL);
	}
	if (!err) {
		struct iattr attr = {
			.ia_valid = ATTR_UID | ATTR_GID,
			.ia_uid = stat->uid,
			.ia_gid = stat->gid,
		};
		err = notify_change(upperdentry, &attr, NULL);
	}
	if (!err)
		ovl_set_timestamps(upperdentry, stat);

	return err;
}

230
struct ovl_fh *ovl_encode_real_fh(struct dentry *real, bool is_upper)
231 232 233 234 235
{
	struct ovl_fh *fh;
	int fh_type, fh_len, dwords;
	void *buf;
	int buflen = MAX_HANDLE_SZ;
236
	uuid_t *uuid = &real->d_sb->s_uuid;
237

238
	buf = kmalloc(buflen, GFP_KERNEL);
239 240 241 242 243 244 245 246 247
	if (!buf)
		return ERR_PTR(-ENOMEM);

	/*
	 * We encode a non-connectable file handle for non-dir, because we
	 * only need to find the lower inode number and we don't want to pay
	 * the price or reconnecting the dentry.
	 */
	dwords = buflen >> 2;
248
	fh_type = exportfs_encode_fh(real, buf, &dwords, 0);
249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268
	buflen = (dwords << 2);

	fh = ERR_PTR(-EIO);
	if (WARN_ON(fh_type < 0) ||
	    WARN_ON(buflen > MAX_HANDLE_SZ) ||
	    WARN_ON(fh_type == FILEID_INVALID))
		goto out;

	BUILD_BUG_ON(MAX_HANDLE_SZ + offsetof(struct ovl_fh, fid) > 255);
	fh_len = offsetof(struct ovl_fh, fid) + buflen;
	fh = kmalloc(fh_len, GFP_KERNEL);
	if (!fh) {
		fh = ERR_PTR(-ENOMEM);
		goto out;
	}

	fh->version = OVL_FH_VERSION;
	fh->magic = OVL_FH_MAGIC;
	fh->type = fh_type;
	fh->flags = OVL_FH_FLAG_CPU_ENDIAN;
269 270 271 272 273 274 275 276
	/*
	 * When we will want to decode an overlay dentry from this handle
	 * and all layers are on the same fs, if we get a disconncted real
	 * dentry when we decode fid, the only way to tell if we should assign
	 * it to upperdentry or to lowerstack is by checking this flag.
	 */
	if (is_upper)
		fh->flags |= OVL_FH_FLAG_PATH_UPPER;
277 278 279 280 281 282 283 284 285
	fh->len = fh_len;
	fh->uuid = *uuid;
	memcpy(fh->fid, buf, buflen);

out:
	kfree(buf);
	return fh;
}

286 287
int ovl_set_origin(struct dentry *dentry, struct dentry *lower,
		   struct dentry *upper)
288 289 290 291 292 293 294 295 296
{
	const struct ovl_fh *fh = NULL;
	int err;

	/*
	 * When lower layer doesn't support export operations store a 'null' fh,
	 * so we can use the overlay.origin xattr to distignuish between a copy
	 * up and a pure upper inode.
	 */
297
	if (ovl_can_decode_fh(lower->d_sb)) {
298
		fh = ovl_encode_real_fh(lower, false);
299 300 301 302
		if (IS_ERR(fh))
			return PTR_ERR(fh);
	}

303 304 305 306 307
	/*
	 * Do not fail when upper doesn't support xattrs.
	 */
	err = ovl_check_setxattr(dentry, upper, OVL_XATTR_ORIGIN, fh,
				 fh ? fh->len : 0, 0);
308 309 310 311 312
	kfree(fh);

	return err;
}

313 314 315 316 317 318
/* Store file handle of @upper dir in @index dir entry */
static int ovl_set_upper_fh(struct dentry *upper, struct dentry *index)
{
	const struct ovl_fh *fh;
	int err;

319
	fh = ovl_encode_real_fh(upper, true);
320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362
	if (IS_ERR(fh))
		return PTR_ERR(fh);

	err = ovl_do_setxattr(index, OVL_XATTR_UPPER, fh, fh->len, 0);

	kfree(fh);
	return err;
}

/*
 * Create and install index entry.
 *
 * Caller must hold i_mutex on indexdir.
 */
static int ovl_create_index(struct dentry *dentry, struct dentry *origin,
			    struct dentry *upper)
{
	struct dentry *indexdir = ovl_indexdir(dentry->d_sb);
	struct inode *dir = d_inode(indexdir);
	struct dentry *index = NULL;
	struct dentry *temp = NULL;
	struct qstr name = { };
	int err;

	/*
	 * For now this is only used for creating index entry for directories,
	 * because non-dir are copied up directly to index and then hardlinked
	 * to upper dir.
	 *
	 * TODO: implement create index for non-dir, so we can call it when
	 * encoding file handle for non-dir in case index does not exist.
	 */
	if (WARN_ON(!d_is_dir(dentry)))
		return -EIO;

	/* Directory not expected to be indexed before copy up */
	if (WARN_ON(ovl_test_flag(OVL_INDEX, d_inode(dentry))))
		return -EIO;

	err = ovl_get_index_name(origin, &name);
	if (err)
		return err;

363
	temp = ovl_create_temp(indexdir, OVL_CATTR(S_IFDIR | 0));
364
	err = PTR_ERR(temp);
365
	if (IS_ERR(temp))
366
		goto free_name;
367 368 369

	err = ovl_set_upper_fh(upper, temp);
	if (err)
370
		goto out;
371 372 373 374 375 376 377 378 379

	index = lookup_one_len(name.name, indexdir, name.len);
	if (IS_ERR(index)) {
		err = PTR_ERR(index);
	} else {
		err = ovl_do_rename(dir, temp, dir, index, 0);
		dput(index);
	}
out:
380 381
	if (err)
		ovl_cleanup(dir, temp);
382
	dput(temp);
383
free_name:
384 385 386 387
	kfree(name.name);
	return err;
}

388 389 390 391 392 393 394 395 396 397 398 399
struct ovl_copy_up_ctx {
	struct dentry *parent;
	struct dentry *dentry;
	struct path lowerpath;
	struct kstat stat;
	struct kstat pstat;
	const char *link;
	struct dentry *destdir;
	struct qstr destname;
	struct dentry *workdir;
	bool tmpfile;
	bool origin;
400
	bool indexed;
401
	bool metacopy;
402 403 404
};

static int ovl_link_up(struct ovl_copy_up_ctx *c)
405 406 407
{
	int err;
	struct dentry *upper;
408
	struct dentry *upperdir = ovl_dentry_upper(c->parent);
409 410
	struct inode *udir = d_inode(upperdir);

411 412 413 414 415 416
	/* Mark parent "impure" because it may now contain non-pure upper */
	err = ovl_set_impure(c->parent, upperdir);
	if (err)
		return err;

	err = ovl_set_nlink_lower(c->dentry);
417 418 419
	if (err)
		return err;

420
	inode_lock_nested(udir, I_MUTEX_PARENT);
421 422
	upper = lookup_one_len(c->dentry->d_name.name, upperdir,
			       c->dentry->d_name.len);
423 424
	err = PTR_ERR(upper);
	if (!IS_ERR(upper)) {
425
		err = ovl_do_link(ovl_dentry_upper(c->dentry), udir, upper);
426 427
		dput(upper);

428 429 430 431 432
		if (!err) {
			/* Restore timestamps on parent (best effort) */
			ovl_set_timestamps(upperdir, &c->pstat);
			ovl_dentry_set_upper_alias(c->dentry);
		}
433 434
	}
	inode_unlock(udir);
435 436 437 438
	if (err)
		return err;

	err = ovl_set_nlink_upper(c->dentry);
439 440 441 442

	return err;
}

443 444
static int ovl_install_temp(struct ovl_copy_up_ctx *c, struct dentry *temp,
			    struct dentry **newdentry)
445 446 447
{
	int err;
	struct dentry *upper;
448
	struct inode *udir = d_inode(c->destdir);
449

450
	upper = lookup_one_len(c->destname.name, c->destdir, c->destname.len);
451 452 453
	if (IS_ERR(upper))
		return PTR_ERR(upper);

454
	if (c->tmpfile)
455
		err = ovl_do_link(temp, udir, upper);
456
	else
457
		err = ovl_do_rename(d_inode(c->workdir), temp, udir, upper, 0);
458

459
	if (!err)
460
		*newdentry = dget(c->tmpfile ? upper : temp);
461 462 463 464 465
	dput(upper);

	return err;
}

466
static struct dentry *ovl_get_tmpfile(struct ovl_copy_up_ctx *c)
M
Miklos Szeredi 已提交
467 468
{
	int err;
469
	struct dentry *temp;
470 471
	const struct cred *old_creds = NULL;
	struct cred *new_creds = NULL;
A
Amir Goldstein 已提交
472
	struct ovl_cattr cattr = {
A
Al Viro 已提交
473
		/* Can't properly set mode on creation because of the umask */
474 475 476
		.mode = c->stat.mode & S_IFMT,
		.rdev = c->stat.rdev,
		.link = c->link
A
Al Viro 已提交
477
	};
M
Miklos Szeredi 已提交
478

479
	err = security_inode_copy_up(c->dentry, &new_creds);
480
	temp = ERR_PTR(err);
481
	if (err < 0)
482
		goto out;
483 484 485 486

	if (new_creds)
		old_creds = override_creds(new_creds);

487
	if (c->tmpfile)
488
		temp = ovl_do_tmpfile(c->workdir, c->stat.mode);
489 490
	else
		temp = ovl_create_temp(c->workdir, &cattr);
491
out:
492 493 494 495 496
	if (new_creds) {
		revert_creds(old_creds);
		put_cred(new_creds);
	}

497
	return temp;
498 499
}

500
static int ovl_copy_up_inode(struct ovl_copy_up_ctx *c, struct dentry *temp)
501 502 503
{
	int err;

504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521
	/*
	 * Copy up data first and then xattrs. Writing data after
	 * xattrs will remove security.capability xattr automatically.
	 */
	if (S_ISREG(c->stat.mode) && !c->metacopy) {
		struct path upperpath, datapath;

		ovl_path_upper(c->dentry, &upperpath);
		if (WARN_ON(upperpath.dentry != NULL))
			return -EIO;
		upperpath.dentry = temp;

		ovl_path_lowerdata(c->dentry, &datapath);
		err = ovl_copy_up_data(&datapath, &upperpath, c->stat.size);
		if (err)
			return err;
	}

522
	err = ovl_copy_xattr(c->lowerpath.dentry, temp);
M
Miklos Szeredi 已提交
523
	if (err)
524
		return err;
M
Miklos Szeredi 已提交
525

526 527 528
	/*
	 * Store identifier of lower inode in upper inode xattr to
	 * allow lookup of the copy up origin inode.
529 530 531
	 *
	 * Don't set origin when we are breaking the association with a lower
	 * hard link.
532
	 */
533
	if (c->origin) {
534
		err = ovl_set_origin(c->dentry, c->lowerpath.dentry, temp);
535
		if (err)
536
			return err;
537
	}
538

539 540 541 542 543 544 545
	if (c->metacopy) {
		err = ovl_check_setxattr(c->dentry, temp, OVL_XATTR_METACOPY,
					 NULL, 0, -EOPNOTSUPP);
		if (err)
			return err;
	}

546
	inode_lock(temp->d_inode);
547 548 549 550
	if (c->metacopy)
		err = ovl_set_size(temp, &c->stat);
	if (!err)
		err = ovl_set_attr(temp, &c->stat);
551 552 553
	inode_unlock(temp->d_inode);

	return err;
554 555
}

556
static int ovl_copy_up_locked(struct ovl_copy_up_ctx *c)
557
{
558
	struct inode *udir = c->destdir->d_inode;
559
	struct inode *inode;
560
	struct dentry *newdentry = NULL;
561
	struct dentry *temp;
562 563
	int err;

564 565 566
	temp = ovl_get_tmpfile(c);
	if (IS_ERR(temp))
		return PTR_ERR(temp);
567

568
	err = ovl_copy_up_inode(c, temp);
569
	if (err)
570
		goto out;
571

572 573 574
	if (S_ISDIR(c->stat.mode) && c->indexed) {
		err = ovl_create_index(c->dentry, c->lowerpath.dentry, temp);
		if (err)
575
			goto out;
576 577
	}

578
	if (c->tmpfile) {
579
		inode_lock_nested(udir, I_MUTEX_PARENT);
580
		err = ovl_install_temp(c, temp, &newdentry);
581 582
		inode_unlock(udir);
	} else {
583
		err = ovl_install_temp(c, temp, &newdentry);
584
	}
M
Miklos Szeredi 已提交
585
	if (err)
586
		goto out;
M
Miklos Szeredi 已提交
587

588 589
	if (!c->metacopy)
		ovl_set_upperdata(d_inode(c->dentry));
590 591 592 593 594
	inode = d_inode(c->dentry);
	ovl_inode_update(inode, newdentry);
	if (S_ISDIR(inode->i_mode))
		ovl_set_flag(OVL_WHITEOUTS, inode);

595
out:
596 597
	if (err && !c->tmpfile)
		ovl_cleanup(d_inode(c->workdir), temp);
598
	dput(temp);
M
Miklos Szeredi 已提交
599 600 601 602 603 604 605
	return err;

}

/*
 * Copy up a single dentry
 *
M
Miklos Szeredi 已提交
606 607 608 609 610
 * All renames start with copy up of source if necessary.  The actual
 * rename will only proceed once the copy up was successful.  Copy up uses
 * upper parent i_mutex for exclusion.  Since rename can change d_parent it
 * is possible that the copy up will lock the old parent.  At that point
 * the file will have already been copied up anyway.
M
Miklos Szeredi 已提交
611
 */
M
Miklos Szeredi 已提交
612
static int ovl_do_copy_up(struct ovl_copy_up_ctx *c)
M
Miklos Szeredi 已提交
613 614
{
	int err;
615
	struct ovl_fs *ofs = c->dentry->d_sb->s_fs_info;
616
	bool to_index = false;
617

618 619 620 621 622 623 624 625 626 627 628 629 630 631 632
	/*
	 * Indexed non-dir is copied up directly to the index entry and then
	 * hardlinked to upper dir. Indexed dir is copied up to indexdir,
	 * then index entry is created and then copied up dir installed.
	 * Copying dir up to indexdir instead of workdir simplifies locking.
	 */
	if (ovl_need_index(c->dentry)) {
		c->indexed = true;
		if (S_ISDIR(c->stat.mode))
			c->workdir = ovl_indexdir(c->dentry->d_sb);
		else
			to_index = true;
	}

	if (S_ISDIR(c->stat.mode) || c->stat.nlink == 1 || to_index)
633 634
		c->origin = true;

635
	if (to_index) {
636 637 638 639
		c->destdir = ovl_indexdir(c->dentry->d_sb);
		err = ovl_get_index_name(c->lowerpath.dentry, &c->destname);
		if (err)
			return err;
640 641 642
	} else if (WARN_ON(!c->parent)) {
		/* Disconnected dentry must be copied up to index dir */
		return -EIO;
643 644 645 646 647 648 649 650 651
	} else {
		/*
		 * Mark parent "impure" because it may now contain non-pure
		 * upper
		 */
		err = ovl_set_impure(c->parent, c->destdir);
		if (err)
			return err;
	}
M
Miklos Szeredi 已提交
652

653
	/* Should we copyup with O_TMPFILE or with workdir? */
654 655
	if (S_ISREG(c->stat.mode) && ofs->tmpfile) {
		c->tmpfile = true;
656 657
		err = ovl_copy_up_locked(c);
	} else {
658 659
		err = ovl_lock_rename_workdir(c->workdir, c->destdir);
		if (!err) {
660 661 662
			err = ovl_copy_up_locked(c);
			unlock_rename(c->workdir, c->destdir);
		}
663 664
	}

665 666 667 668 669

	if (err)
		goto out;

	if (c->indexed)
670 671 672
		ovl_set_flag(OVL_INDEX, d_inode(c->dentry));

	if (to_index) {
673 674 675
		/* Initialize nlink for copy up of disconnected dentry */
		err = ovl_set_nlink_upper(c->dentry);
	} else {
676 677 678 679 680 681 682 683
		struct inode *udir = d_inode(c->destdir);

		/* Restore timestamps on parent (best effort) */
		inode_lock(udir);
		ovl_set_timestamps(c->destdir, &c->pstat);
		inode_unlock(udir);

		ovl_dentry_set_upper_alias(c->dentry);
M
Miklos Szeredi 已提交
684 685
	}

686 687 688
out:
	if (to_index)
		kfree(c->destname.name);
M
Miklos Szeredi 已提交
689 690 691
	return err;
}

692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708
static bool ovl_need_meta_copy_up(struct dentry *dentry, umode_t mode,
				  int flags)
{
	struct ovl_fs *ofs = dentry->d_sb->s_fs_info;

	if (!ofs->config.metacopy)
		return false;

	if (!S_ISREG(mode))
		return false;

	if (flags && ((OPEN_FMODE(flags) & FMODE_WRITE) || (flags & O_TRUNC)))
		return false;

	return true;
}

709 710 711
/* Copy up data of an inode which was copied up metadata only in the past. */
static int ovl_copy_up_meta_inode_data(struct ovl_copy_up_ctx *c)
{
712
	struct path upperpath, datapath;
713
	int err;
714 715
	char *capability = NULL;
	ssize_t uninitialized_var(cap_size);
716 717 718 719 720

	ovl_path_upper(c->dentry, &upperpath);
	if (WARN_ON(upperpath.dentry == NULL))
		return -EIO;

721 722 723 724
	ovl_path_lowerdata(c->dentry, &datapath);
	if (WARN_ON(datapath.dentry == NULL))
		return -EIO;

725 726 727 728 729 730 731
	if (c->stat.size) {
		err = cap_size = ovl_getxattr(upperpath.dentry, XATTR_NAME_CAPS,
					      &capability, 0);
		if (err < 0 && err != -ENODATA)
			goto out;
	}

732
	err = ovl_copy_up_data(&datapath, &upperpath, c->stat.size);
733
	if (err)
734 735 736 737 738 739 740 741 742 743 744 745 746
		goto out_free;

	/*
	 * Writing to upper file will clear security.capability xattr. We
	 * don't want that to happen for normal copy-up operation.
	 */
	if (capability) {
		err = ovl_do_setxattr(upperpath.dentry, XATTR_NAME_CAPS,
				      capability, cap_size, 0);
		if (err)
			goto out_free;
	}

747 748 749

	err = vfs_removexattr(upperpath.dentry, OVL_XATTR_METACOPY);
	if (err)
750
		goto out_free;
751 752

	ovl_set_upperdata(d_inode(c->dentry));
753 754 755
out_free:
	kfree(capability);
out:
756 757 758
	return err;
}

M
Miklos Szeredi 已提交
759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779
static int ovl_copy_up_one(struct dentry *parent, struct dentry *dentry,
			   int flags)
{
	int err;
	DEFINE_DELAYED_CALL(done);
	struct path parentpath;
	struct ovl_copy_up_ctx ctx = {
		.parent = parent,
		.dentry = dentry,
		.workdir = ovl_workdir(dentry),
	};

	if (WARN_ON(!ctx.workdir))
		return -EROFS;

	ovl_path_lower(dentry, &ctx.lowerpath);
	err = vfs_getattr(&ctx.lowerpath, &ctx.stat,
			  STATX_BASIC_STATS, AT_STATX_SYNC_AS_STAT);
	if (err)
		return err;

780 781
	ctx.metacopy = ovl_need_meta_copy_up(dentry, ctx.stat.mode, flags);

782 783 784 785
	if (parent) {
		ovl_path_upper(parent, &parentpath);
		ctx.destdir = parentpath.dentry;
		ctx.destname = dentry->d_name;
M
Miklos Szeredi 已提交
786

787 788 789 790 791 792
		err = vfs_getattr(&parentpath, &ctx.pstat,
				  STATX_ATIME | STATX_MTIME,
				  AT_STATX_SYNC_AS_STAT);
		if (err)
			return err;
	}
M
Miklos Szeredi 已提交
793 794 795 796 797 798 799 800 801 802 803

	/* maybe truncate regular file. this has no effect on dirs */
	if (flags & O_TRUNC)
		ctx.stat.size = 0;

	if (S_ISLNK(ctx.stat.mode)) {
		ctx.link = vfs_get_link(ctx.lowerpath.dentry, &done);
		if (IS_ERR(ctx.link))
			return PTR_ERR(ctx.link);
	}

804
	err = ovl_copy_up_start(dentry, flags);
M
Miklos Szeredi 已提交
805 806 807 808 809
	/* err < 0: interrupted, err > 0: raced with another copy-up */
	if (unlikely(err)) {
		if (err > 0)
			err = 0;
	} else {
810 811
		if (!ovl_dentry_upper(dentry))
			err = ovl_do_copy_up(&ctx);
812
		if (!err && parent && !ovl_dentry_has_upper_alias(dentry))
813
			err = ovl_link_up(&ctx);
814 815
		if (!err && ovl_dentry_needs_data_copy_up_locked(dentry, flags))
			err = ovl_copy_up_meta_inode_data(&ctx);
M
Miklos Szeredi 已提交
816 817
		ovl_copy_up_end(dentry);
	}
M
Miklos Szeredi 已提交
818
	do_delayed_call(&done);
M
Miklos Szeredi 已提交
819 820 821 822

	return err;
}

823
int ovl_copy_up_flags(struct dentry *dentry, int flags)
M
Miklos Szeredi 已提交
824
{
825 826
	int err = 0;
	const struct cred *old_cred = ovl_override_creds(dentry->d_sb);
827 828 829 830 831 832 833 834 835
	bool disconnected = (dentry->d_flags & DCACHE_DISCONNECTED);

	/*
	 * With NFS export, copy up can get called for a disconnected non-dir.
	 * In this case, we will copy up lower inode to index dir without
	 * linking it to upper dir.
	 */
	if (WARN_ON(disconnected && d_is_dir(dentry)))
		return -EIO;
M
Miklos Szeredi 已提交
836 837 838

	while (!err) {
		struct dentry *next;
839
		struct dentry *parent = NULL;
M
Miklos Szeredi 已提交
840

841
		if (ovl_already_copied_up(dentry, flags))
M
Miklos Szeredi 已提交
842 843 844 845
			break;

		next = dget(dentry);
		/* find the topmost dentry not yet copied up */
846
		for (; !disconnected;) {
M
Miklos Szeredi 已提交
847 848
			parent = dget_parent(next);

849
			if (ovl_dentry_upper(parent))
M
Miklos Szeredi 已提交
850 851 852 853 854 855
				break;

			dput(next);
			next = parent;
		}

M
Miklos Szeredi 已提交
856
		err = ovl_copy_up_one(parent, next, flags);
M
Miklos Szeredi 已提交
857 858 859 860

		dput(parent);
		dput(next);
	}
861
	revert_creds(old_cred);
M
Miklos Szeredi 已提交
862 863 864

	return err;
}
865

866 867 868
static bool ovl_open_need_copy_up(struct dentry *dentry, int flags)
{
	/* Copy up of disconnected dentry does not set upper alias */
869
	if (ovl_already_copied_up(dentry, flags))
870 871 872 873 874
		return false;

	if (special_file(d_inode(dentry)->i_mode))
		return false;

875
	if (!ovl_open_flags_need_copy_up(flags))
876 877 878 879 880
		return false;

	return true;
}

881
int ovl_maybe_copy_up(struct dentry *dentry, int flags)
882 883 884
{
	int err = 0;

885
	if (ovl_open_need_copy_up(dentry, flags)) {
886 887
		err = ovl_want_write(dentry);
		if (!err) {
888
			err = ovl_copy_up_flags(dentry, flags);
889 890 891 892 893 894 895
			ovl_drop_write(dentry);
		}
	}

	return err;
}

896 897 898 899 900
int ovl_copy_up_with_data(struct dentry *dentry)
{
	return ovl_copy_up_flags(dentry, O_WRONLY);
}

901 902 903 904
int ovl_copy_up(struct dentry *dentry)
{
	return ovl_copy_up_flags(dentry, 0);
}