copy_up.c 20.5 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	pr_warn("overlayfs: \"check_copy_up\" module option is obsolete\n");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
	}
	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;
128
	loff_t cloned;
M
Miklos Szeredi 已提交
129 130 131 132 133
	int error = 0;

	if (len == 0)
		return 0;

134
	old_file = ovl_path_open(old, O_LARGEFILE | O_RDONLY);
M
Miklos Szeredi 已提交
135 136 137
	if (IS_ERR(old_file))
		return PTR_ERR(old_file);

138
	new_file = ovl_path_open(new, O_LARGEFILE | O_WRONLY);
M
Miklos Szeredi 已提交
139 140 141 142 143
	if (IS_ERR(new_file)) {
		error = PTR_ERR(new_file);
		goto out_fput;
	}

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

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
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 origin;
399
	bool indexed;
400
	bool metacopy;
401 402 403
};

static int ovl_link_up(struct ovl_copy_up_ctx *c)
404 405 406
{
	int err;
	struct dentry *upper;
407
	struct dentry *upperdir = ovl_dentry_upper(c->parent);
408 409
	struct inode *udir = d_inode(upperdir);

410 411 412 413 414 415
	/* 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);
416 417 418
	if (err)
		return err;

419
	inode_lock_nested(udir, I_MUTEX_PARENT);
420 421
	upper = lookup_one_len(c->dentry->d_name.name, upperdir,
			       c->dentry->d_name.len);
422 423
	err = PTR_ERR(upper);
	if (!IS_ERR(upper)) {
424
		err = ovl_do_link(ovl_dentry_upper(c->dentry), udir, upper);
425 426
		dput(upper);

427 428 429 430 431
		if (!err) {
			/* Restore timestamps on parent (best effort) */
			ovl_set_timestamps(upperdir, &c->pstat);
			ovl_dentry_set_upper_alias(c->dentry);
		}
432 433
	}
	inode_unlock(udir);
434 435 436 437
	if (err)
		return err;

	err = ovl_set_nlink_upper(c->dentry);
438 439 440 441

	return err;
}

442
static int ovl_copy_up_inode(struct ovl_copy_up_ctx *c, struct dentry *temp)
443 444 445
{
	int err;

446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463
	/*
	 * 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;
	}

464
	err = ovl_copy_xattr(c->lowerpath.dentry, temp);
M
Miklos Szeredi 已提交
465
	if (err)
466
		return err;
M
Miklos Szeredi 已提交
467

468 469 470
	/*
	 * Store identifier of lower inode in upper inode xattr to
	 * allow lookup of the copy up origin inode.
471 472 473
	 *
	 * Don't set origin when we are breaking the association with a lower
	 * hard link.
474
	 */
475
	if (c->origin) {
476
		err = ovl_set_origin(c->dentry, c->lowerpath.dentry, temp);
477
		if (err)
478
			return err;
479
	}
480

481 482 483 484 485 486 487
	if (c->metacopy) {
		err = ovl_check_setxattr(c->dentry, temp, OVL_XATTR_METACOPY,
					 NULL, 0, -EOPNOTSUPP);
		if (err)
			return err;
	}

488
	inode_lock(temp->d_inode);
489 490 491 492
	if (c->metacopy)
		err = ovl_set_size(temp, &c->stat);
	if (!err)
		err = ovl_set_attr(temp, &c->stat);
493 494 495
	inode_unlock(temp->d_inode);

	return err;
496 497
}

498 499 500 501 502 503
struct ovl_cu_creds {
	const struct cred *old;
	struct cred *new;
};

static int ovl_prep_cu_creds(struct dentry *dentry, struct ovl_cu_creds *cc)
A
Amir Goldstein 已提交
504 505 506
{
	int err;

507 508
	cc->old = cc->new = NULL;
	err = security_inode_copy_up(dentry, &cc->new);
A
Amir Goldstein 已提交
509
	if (err < 0)
510
		return err;
A
Amir Goldstein 已提交
511

512 513
	if (cc->new)
		cc->old = override_creds(cc->new);
A
Amir Goldstein 已提交
514

515
	return 0;
A
Amir Goldstein 已提交
516 517
}

518
static void ovl_revert_cu_creds(struct ovl_cu_creds *cc)
A
Amir Goldstein 已提交
519
{
520 521 522 523
	if (cc->new) {
		revert_creds(cc->old);
		put_cred(cc->new);
	}
A
Amir Goldstein 已提交
524 525 526 527 528 529 530
}

/*
 * Copyup using workdir to prepare temp file.  Used when copying up directories,
 * special files or when upper fs doesn't support O_TMPFILE.
 */
static int ovl_copy_up_workdir(struct ovl_copy_up_ctx *c)
531
{
532
	struct inode *inode;
533 534 535
	struct inode *udir = d_inode(c->destdir), *wdir = d_inode(c->workdir);
	struct dentry *temp, *upper;
	struct ovl_cu_creds cc;
536
	int err;
537 538 539 540 541 542
	struct ovl_cattr cattr = {
		/* Can't properly set mode on creation because of the umask */
		.mode = c->stat.mode & S_IFMT,
		.rdev = c->stat.rdev,
		.link = c->link
	};
543

A
Amir Goldstein 已提交
544 545 546 547
	err = ovl_lock_rename_workdir(c->workdir, c->destdir);
	if (err)
		return err;

548 549 550 551 552 553 554
	err = ovl_prep_cu_creds(c->dentry, &cc);
	if (err)
		goto unlock;

	temp = ovl_create_temp(c->workdir, &cattr);
	ovl_revert_cu_creds(&cc);

A
Amir Goldstein 已提交
555
	err = PTR_ERR(temp);
556
	if (IS_ERR(temp))
A
Amir Goldstein 已提交
557
		goto unlock;
558

559
	err = ovl_copy_up_inode(c, temp);
560
	if (err)
A
Amir Goldstein 已提交
561
		goto cleanup;
562

563 564 565
	if (S_ISDIR(c->stat.mode) && c->indexed) {
		err = ovl_create_index(c->dentry, c->lowerpath.dentry, temp);
		if (err)
A
Amir Goldstein 已提交
566
			goto cleanup;
567 568
	}

569 570 571 572 573 574 575
	upper = lookup_one_len(c->destname.name, c->destdir, c->destname.len);
	err = PTR_ERR(upper);
	if (IS_ERR(upper))
		goto cleanup;

	err = ovl_do_rename(wdir, temp, udir, upper, 0);
	dput(upper);
M
Miklos Szeredi 已提交
576
	if (err)
A
Amir Goldstein 已提交
577
		goto cleanup;
M
Miklos Szeredi 已提交
578

579 580
	if (!c->metacopy)
		ovl_set_upperdata(d_inode(c->dentry));
581
	inode = d_inode(c->dentry);
582
	ovl_inode_update(inode, temp);
583 584
	if (S_ISDIR(inode->i_mode))
		ovl_set_flag(OVL_WHITEOUTS, inode);
A
Amir Goldstein 已提交
585 586 587 588 589 590
unlock:
	unlock_rename(c->workdir, c->destdir);

	return err;

cleanup:
591 592 593
	ovl_cleanup(wdir, temp);
	dput(temp);
	goto unlock;
A
Amir Goldstein 已提交
594 595
}

596 597
/* Copyup using O_TMPFILE which does not require cross dir locking */
static int ovl_copy_up_tmpfile(struct ovl_copy_up_ctx *c)
A
Amir Goldstein 已提交
598
{
599 600 601
	struct inode *udir = d_inode(c->destdir);
	struct dentry *temp, *upper;
	struct ovl_cu_creds cc;
A
Amir Goldstein 已提交
602 603
	int err;

604 605 606
	err = ovl_prep_cu_creds(c->dentry, &cc);
	if (err)
		return err;
A
Amir Goldstein 已提交
607 608

	temp = ovl_do_tmpfile(c->workdir, c->stat.mode);
609
	ovl_revert_cu_creds(&cc);
A
Amir Goldstein 已提交
610

611 612
	if (IS_ERR(temp))
		return PTR_ERR(temp);
A
Amir Goldstein 已提交
613

614 615 616
	err = ovl_copy_up_inode(c, temp);
	if (err)
		goto out_dput;
A
Amir Goldstein 已提交
617 618 619 620 621

	inode_lock_nested(udir, I_MUTEX_PARENT);

	upper = lookup_one_len(c->destname.name, c->destdir, c->destname.len);
	err = PTR_ERR(upper);
622 623 624 625
	if (!IS_ERR(upper)) {
		err = ovl_do_link(temp, udir, upper);
		dput(upper);
	}
A
Amir Goldstein 已提交
626 627 628
	inode_unlock(udir);

	if (err)
629
		goto out_dput;
A
Amir Goldstein 已提交
630 631 632

	if (!c->metacopy)
		ovl_set_upperdata(d_inode(c->dentry));
633
	ovl_inode_update(d_inode(c->dentry), temp);
634

635 636 637
	return 0;

out_dput:
638
	dput(temp);
M
Miklos Szeredi 已提交
639 640 641 642 643 644
	return err;
}

/*
 * Copy up a single dentry
 *
M
Miklos Szeredi 已提交
645 646 647 648 649
 * 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
650
 */
M
Miklos Szeredi 已提交
651
static int ovl_do_copy_up(struct ovl_copy_up_ctx *c)
M
Miklos Szeredi 已提交
652 653
{
	int err;
654
	struct ovl_fs *ofs = c->dentry->d_sb->s_fs_info;
655
	bool to_index = false;
656

657 658 659 660 661 662 663 664 665 666 667 668 669 670 671
	/*
	 * 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)
672 673
		c->origin = true;

674
	if (to_index) {
675 676 677 678
		c->destdir = ovl_indexdir(c->dentry->d_sb);
		err = ovl_get_index_name(c->lowerpath.dentry, &c->destname);
		if (err)
			return err;
679 680 681
	} else if (WARN_ON(!c->parent)) {
		/* Disconnected dentry must be copied up to index dir */
		return -EIO;
682 683 684 685 686 687 688 689 690
	} 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
691

692
	/* Should we copyup with O_TMPFILE or with workdir? */
A
Amir Goldstein 已提交
693 694 695 696
	if (S_ISREG(c->stat.mode) && ofs->tmpfile)
		err = ovl_copy_up_tmpfile(c);
	else
		err = ovl_copy_up_workdir(c);
697 698 699 700
	if (err)
		goto out;

	if (c->indexed)
701 702 703
		ovl_set_flag(OVL_INDEX, d_inode(c->dentry));

	if (to_index) {
704 705 706
		/* Initialize nlink for copy up of disconnected dentry */
		err = ovl_set_nlink_upper(c->dentry);
	} else {
707 708 709 710 711 712 713 714
		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
715 716
	}

717 718 719
out:
	if (to_index)
		kfree(c->destname.name);
M
Miklos Szeredi 已提交
720 721 722
	return err;
}

723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739
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;
}

740 741 742
/* 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)
{
743
	struct path upperpath, datapath;
744
	int err;
745 746
	char *capability = NULL;
	ssize_t uninitialized_var(cap_size);
747 748 749 750 751

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

752 753 754 755
	ovl_path_lowerdata(c->dentry, &datapath);
	if (WARN_ON(datapath.dentry == NULL))
		return -EIO;

756 757 758 759 760 761 762
	if (c->stat.size) {
		err = cap_size = ovl_getxattr(upperpath.dentry, XATTR_NAME_CAPS,
					      &capability, 0);
		if (err < 0 && err != -ENODATA)
			goto out;
	}

763
	err = ovl_copy_up_data(&datapath, &upperpath, c->stat.size);
764
	if (err)
765 766 767 768 769 770 771 772 773 774 775 776 777
		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;
	}

778 779 780

	err = vfs_removexattr(upperpath.dentry, OVL_XATTR_METACOPY);
	if (err)
781
		goto out_free;
782 783

	ovl_set_upperdata(d_inode(c->dentry));
784 785 786
out_free:
	kfree(capability);
out:
787 788 789
	return err;
}

M
Miklos Szeredi 已提交
790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810
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;

811 812
	ctx.metacopy = ovl_need_meta_copy_up(dentry, ctx.stat.mode, flags);

813 814 815 816
	if (parent) {
		ovl_path_upper(parent, &parentpath);
		ctx.destdir = parentpath.dentry;
		ctx.destname = dentry->d_name;
M
Miklos Szeredi 已提交
817

818 819 820 821 822 823
		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
824 825 826 827 828 829 830 831 832 833 834

	/* 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);
	}

835
	err = ovl_copy_up_start(dentry, flags);
M
Miklos Szeredi 已提交
836 837 838 839 840
	/* err < 0: interrupted, err > 0: raced with another copy-up */
	if (unlikely(err)) {
		if (err > 0)
			err = 0;
	} else {
841 842
		if (!ovl_dentry_upper(dentry))
			err = ovl_do_copy_up(&ctx);
843
		if (!err && parent && !ovl_dentry_has_upper_alias(dentry))
844
			err = ovl_link_up(&ctx);
845 846
		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
847 848
		ovl_copy_up_end(dentry);
	}
M
Miklos Szeredi 已提交
849
	do_delayed_call(&done);
M
Miklos Szeredi 已提交
850 851 852 853

	return err;
}

854
int ovl_copy_up_flags(struct dentry *dentry, int flags)
M
Miklos Szeredi 已提交
855
{
856 857
	int err = 0;
	const struct cred *old_cred = ovl_override_creds(dentry->d_sb);
858 859 860 861 862 863 864 865 866
	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
867 868 869

	while (!err) {
		struct dentry *next;
870
		struct dentry *parent = NULL;
M
Miklos Szeredi 已提交
871

872
		if (ovl_already_copied_up(dentry, flags))
M
Miklos Szeredi 已提交
873 874 875 876
			break;

		next = dget(dentry);
		/* find the topmost dentry not yet copied up */
877
		for (; !disconnected;) {
M
Miklos Szeredi 已提交
878 879
			parent = dget_parent(next);

880
			if (ovl_dentry_upper(parent))
M
Miklos Szeredi 已提交
881 882 883 884 885 886
				break;

			dput(next);
			next = parent;
		}

M
Miklos Szeredi 已提交
887
		err = ovl_copy_up_one(parent, next, flags);
M
Miklos Szeredi 已提交
888 889 890 891

		dput(parent);
		dput(next);
	}
892
	revert_creds(old_cred);
M
Miklos Szeredi 已提交
893 894 895

	return err;
}
896

897 898 899
static bool ovl_open_need_copy_up(struct dentry *dentry, int flags)
{
	/* Copy up of disconnected dentry does not set upper alias */
900
	if (ovl_already_copied_up(dentry, flags))
901 902 903 904 905
		return false;

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

906
	if (!ovl_open_flags_need_copy_up(flags))
907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926
		return false;

	return true;
}

int ovl_open_maybe_copy_up(struct dentry *dentry, unsigned int file_flags)
{
	int err = 0;

	if (ovl_open_need_copy_up(dentry, file_flags)) {
		err = ovl_want_write(dentry);
		if (!err) {
			err = ovl_copy_up_flags(dentry, file_flags);
			ovl_drop_write(dentry);
		}
	}

	return err;
}

927 928 929 930 931
int ovl_copy_up_with_data(struct dentry *dentry)
{
	return ovl_copy_up_flags(dentry, O_WRONLY);
}

932 933 934 935
int ovl_copy_up(struct dentry *dentry)
{
	return ovl_copy_up_flags(dentry, 0);
}