open.c 24.5 KB
Newer Older
Linus Torvalds's avatar
Linus Torvalds committed
1 2 3 4 5 6 7 8 9 10
/*
 *  linux/fs/open.c
 *
 *  Copyright (C) 1991, 1992  Linus Torvalds
 */

#include <linux/string.h>
#include <linux/mm.h>
#include <linux/file.h>
#include <linux/quotaops.h>
Robert Love's avatar
Robert Love committed
11
#include <linux/fsnotify.h>
Linus Torvalds's avatar
Linus Torvalds committed
12 13 14 15 16
#include <linux/module.h>
#include <linux/slab.h>
#include <linux/tty.h>
#include <linux/namei.h>
#include <linux/backing-dev.h>
17
#include <linux/capability.h>
Linus Torvalds's avatar
Linus Torvalds committed
18 19 20
#include <linux/security.h>
#include <linux/mount.h>
#include <linux/vfs.h>
21
#include <linux/fcntl.h>
Linus Torvalds's avatar
Linus Torvalds committed
22 23
#include <asm/uaccess.h>
#include <linux/fs.h>
24
#include <linux/personality.h>
Linus Torvalds's avatar
Linus Torvalds committed
25 26
#include <linux/pagemap.h>
#include <linux/syscalls.h>
27
#include <linux/rcupdate.h>
28
#include <linux/audit.h>
Linus Torvalds's avatar
Linus Torvalds committed
29

30
int vfs_statfs(struct dentry *dentry, struct kstatfs *buf)
Linus Torvalds's avatar
Linus Torvalds committed
31 32 33
{
	int retval = -ENODEV;

34
	if (dentry) {
Linus Torvalds's avatar
Linus Torvalds committed
35
		retval = -ENOSYS;
36
		if (dentry->d_sb->s_op->statfs) {
Linus Torvalds's avatar
Linus Torvalds committed
37
			memset(buf, 0, sizeof(*buf));
38
			retval = security_sb_statfs(dentry);
Linus Torvalds's avatar
Linus Torvalds committed
39 40
			if (retval)
				return retval;
41
			retval = dentry->d_sb->s_op->statfs(dentry, buf);
Linus Torvalds's avatar
Linus Torvalds committed
42 43 44 45 46 47 48 49 50
			if (retval == 0 && buf->f_frsize == 0)
				buf->f_frsize = buf->f_bsize;
		}
	}
	return retval;
}

EXPORT_SYMBOL(vfs_statfs);

51
static int vfs_statfs_native(struct dentry *dentry, struct statfs *buf)
Linus Torvalds's avatar
Linus Torvalds committed
52 53 54 55
{
	struct kstatfs st;
	int retval;

56
	retval = vfs_statfs(dentry, &st);
Linus Torvalds's avatar
Linus Torvalds committed
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
	if (retval)
		return retval;

	if (sizeof(*buf) == sizeof(st))
		memcpy(buf, &st, sizeof(st));
	else {
		if (sizeof buf->f_blocks == 4) {
			if ((st.f_blocks | st.f_bfree | st.f_bavail) &
			    0xffffffff00000000ULL)
				return -EOVERFLOW;
			/*
			 * f_files and f_ffree may be -1; it's okay to stuff
			 * that into 32 bits
			 */
			if (st.f_files != -1 &&
			    (st.f_files & 0xffffffff00000000ULL))
				return -EOVERFLOW;
			if (st.f_ffree != -1 &&
			    (st.f_ffree & 0xffffffff00000000ULL))
				return -EOVERFLOW;
		}

		buf->f_type = st.f_type;
		buf->f_bsize = st.f_bsize;
		buf->f_blocks = st.f_blocks;
		buf->f_bfree = st.f_bfree;
		buf->f_bavail = st.f_bavail;
		buf->f_files = st.f_files;
		buf->f_ffree = st.f_ffree;
		buf->f_fsid = st.f_fsid;
		buf->f_namelen = st.f_namelen;
		buf->f_frsize = st.f_frsize;
		memset(buf->f_spare, 0, sizeof(buf->f_spare));
	}
	return 0;
}

94
static int vfs_statfs64(struct dentry *dentry, struct statfs64 *buf)
Linus Torvalds's avatar
Linus Torvalds committed
95 96 97 98
{
	struct kstatfs st;
	int retval;

99
	retval = vfs_statfs(dentry, &st);
Linus Torvalds's avatar
Linus Torvalds committed
100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128
	if (retval)
		return retval;

	if (sizeof(*buf) == sizeof(st))
		memcpy(buf, &st, sizeof(st));
	else {
		buf->f_type = st.f_type;
		buf->f_bsize = st.f_bsize;
		buf->f_blocks = st.f_blocks;
		buf->f_bfree = st.f_bfree;
		buf->f_bavail = st.f_bavail;
		buf->f_files = st.f_files;
		buf->f_ffree = st.f_ffree;
		buf->f_fsid = st.f_fsid;
		buf->f_namelen = st.f_namelen;
		buf->f_frsize = st.f_frsize;
		memset(buf->f_spare, 0, sizeof(buf->f_spare));
	}
	return 0;
}

asmlinkage long sys_statfs(const char __user * path, struct statfs __user * buf)
{
	struct nameidata nd;
	int error;

	error = user_path_walk(path, &nd);
	if (!error) {
		struct statfs tmp;
129
		error = vfs_statfs_native(nd.dentry, &tmp);
Linus Torvalds's avatar
Linus Torvalds committed
130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147
		if (!error && copy_to_user(buf, &tmp, sizeof(tmp)))
			error = -EFAULT;
		path_release(&nd);
	}
	return error;
}


asmlinkage long sys_statfs64(const char __user *path, size_t sz, struct statfs64 __user *buf)
{
	struct nameidata nd;
	long error;

	if (sz != sizeof(*buf))
		return -EINVAL;
	error = user_path_walk(path, &nd);
	if (!error) {
		struct statfs64 tmp;
148
		error = vfs_statfs64(nd.dentry, &tmp);
Linus Torvalds's avatar
Linus Torvalds committed
149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166
		if (!error && copy_to_user(buf, &tmp, sizeof(tmp)))
			error = -EFAULT;
		path_release(&nd);
	}
	return error;
}


asmlinkage long sys_fstatfs(unsigned int fd, struct statfs __user * buf)
{
	struct file * file;
	struct statfs tmp;
	int error;

	error = -EBADF;
	file = fget(fd);
	if (!file)
		goto out;
167
	error = vfs_statfs_native(file->f_path.dentry, &tmp);
Linus Torvalds's avatar
Linus Torvalds committed
168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187
	if (!error && copy_to_user(buf, &tmp, sizeof(tmp)))
		error = -EFAULT;
	fput(file);
out:
	return error;
}

asmlinkage long sys_fstatfs64(unsigned int fd, size_t sz, struct statfs64 __user *buf)
{
	struct file * file;
	struct statfs64 tmp;
	int error;

	if (sz != sizeof(*buf))
		return -EINVAL;

	error = -EBADF;
	file = fget(fd);
	if (!file)
		goto out;
188
	error = vfs_statfs64(file->f_path.dentry, &tmp);
Linus Torvalds's avatar
Linus Torvalds committed
189 190 191 192 193 194 195
	if (!error && copy_to_user(buf, &tmp, sizeof(tmp)))
		error = -EFAULT;
	fput(file);
out:
	return error;
}

196 197
int do_truncate(struct dentry *dentry, loff_t length, unsigned int time_attrs,
	struct file *filp)
Linus Torvalds's avatar
Linus Torvalds committed
198 199 200 201 202 203 204 205 206
{
	int err;
	struct iattr newattrs;

	/* Not pretty: "inode->i_size" shouldn't really be signed. But it is. */
	if (length < 0)
		return -EINVAL;

	newattrs.ia_size = length;
207
	newattrs.ia_valid = ATTR_SIZE | time_attrs;
208 209 210 211
	if (filp) {
		newattrs.ia_file = filp;
		newattrs.ia_valid |= ATTR_FILE;
	}
Linus Torvalds's avatar
Linus Torvalds committed
212

213
	mutex_lock(&dentry->d_inode->i_mutex);
Linus Torvalds's avatar
Linus Torvalds committed
214
	err = notify_change(dentry, &newattrs);
215
	mutex_unlock(&dentry->d_inode->i_mutex);
Linus Torvalds's avatar
Linus Torvalds committed
216 217 218
	return err;
}

219
static long do_sys_truncate(const char __user * path, loff_t length)
Linus Torvalds's avatar
Linus Torvalds committed
220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242
{
	struct nameidata nd;
	struct inode * inode;
	int error;

	error = -EINVAL;
	if (length < 0)	/* sorry, but loff_t says... */
		goto out;

	error = user_path_walk(path, &nd);
	if (error)
		goto out;
	inode = nd.dentry->d_inode;

	/* For directories it's -EISDIR, for other non-regulars - -EINVAL */
	error = -EISDIR;
	if (S_ISDIR(inode->i_mode))
		goto dput_and_out;

	error = -EINVAL;
	if (!S_ISREG(inode->i_mode))
		goto dput_and_out;

243
	error = vfs_permission(&nd, MAY_WRITE);
Linus Torvalds's avatar
Linus Torvalds committed
244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268
	if (error)
		goto dput_and_out;

	error = -EROFS;
	if (IS_RDONLY(inode))
		goto dput_and_out;

	error = -EPERM;
	if (IS_IMMUTABLE(inode) || IS_APPEND(inode))
		goto dput_and_out;

	/*
	 * Make sure that there are no leases.
	 */
	error = break_lease(inode, FMODE_WRITE);
	if (error)
		goto dput_and_out;

	error = get_write_access(inode);
	if (error)
		goto dput_and_out;

	error = locks_verify_truncate(inode, NULL, length);
	if (!error) {
		DQUOT_INIT(inode);
269
		error = do_truncate(nd.dentry, length, 0, NULL);
Linus Torvalds's avatar
Linus Torvalds committed
270 271 272 273 274 275 276 277 278 279 280 281 282 283 284
	}
	put_write_access(inode);

dput_and_out:
	path_release(&nd);
out:
	return error;
}

asmlinkage long sys_truncate(const char __user * path, unsigned long length)
{
	/* on 32-bit boxen it will cut the range 2^31--2^32-1 off */
	return do_sys_truncate(path, (long)length);
}

285
static long do_sys_ftruncate(unsigned int fd, loff_t length, int small)
Linus Torvalds's avatar
Linus Torvalds committed
286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303
{
	struct inode * inode;
	struct dentry *dentry;
	struct file * file;
	int error;

	error = -EINVAL;
	if (length < 0)
		goto out;
	error = -EBADF;
	file = fget(fd);
	if (!file)
		goto out;

	/* explicitly opened as large or we are on 64-bit box */
	if (file->f_flags & O_LARGEFILE)
		small = 0;

304
	dentry = file->f_path.dentry;
Linus Torvalds's avatar
Linus Torvalds committed
305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320
	inode = dentry->d_inode;
	error = -EINVAL;
	if (!S_ISREG(inode->i_mode) || !(file->f_mode & FMODE_WRITE))
		goto out_putf;

	error = -EINVAL;
	/* Cannot ftruncate over 2^31 bytes without large file support */
	if (small && length > MAX_NON_LFS)
		goto out_putf;

	error = -EPERM;
	if (IS_APPEND(inode))
		goto out_putf;

	error = locks_verify_truncate(inode, file, length);
	if (!error)
321
		error = do_truncate(dentry, length, ATTR_MTIME|ATTR_CTIME, file);
Linus Torvalds's avatar
Linus Torvalds committed
322 323 324 325 326 327 328 329
out_putf:
	fput(file);
out:
	return error;
}

asmlinkage long sys_ftruncate(unsigned int fd, unsigned long length)
{
330
	long ret = do_sys_ftruncate(fd, length, 1);
331
	/* avoid REGPARM breakage on x86: */
332 333
	prevent_tail_call(ret);
	return ret;
Linus Torvalds's avatar
Linus Torvalds committed
334 335 336 337 338 339 340 341 342 343 344
}

/* LFS versions of truncate are only needed on 32 bit machines */
#if BITS_PER_LONG == 32
asmlinkage long sys_truncate64(const char __user * path, loff_t length)
{
	return do_sys_truncate(path, length);
}

asmlinkage long sys_ftruncate64(unsigned int fd, loff_t length)
{
345
	long ret = do_sys_ftruncate(fd, length, 0);
346
	/* avoid REGPARM breakage on x86: */
347 348
	prevent_tail_call(ret);
	return ret;
Linus Torvalds's avatar
Linus Torvalds committed
349 350 351 352 353 354 355 356
}
#endif

/*
 * access() needs to use the real uid/gid, not the effective uid/gid.
 * We do this by temporarily clearing all FS-related capabilities and
 * switching the fsuid/fsgid around to the real ones.
 */
357
asmlinkage long sys_faccessat(int dfd, const char __user *filename, int mode)
Linus Torvalds's avatar
Linus Torvalds committed
358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386
{
	struct nameidata nd;
	int old_fsuid, old_fsgid;
	kernel_cap_t old_cap;
	int res;

	if (mode & ~S_IRWXO)	/* where's F_OK, X_OK, W_OK, R_OK? */
		return -EINVAL;

	old_fsuid = current->fsuid;
	old_fsgid = current->fsgid;
	old_cap = current->cap_effective;

	current->fsuid = current->uid;
	current->fsgid = current->gid;

	/*
	 * Clear the capabilities if we switch to a non-root user
	 *
	 * FIXME: There is a race here against sys_capset.  The
	 * capabilities can change yet we will restore the old
	 * value below.  We should hold task_capabilities_lock,
	 * but we cannot because user_path_walk can sleep.
	 */
	if (current->uid)
		cap_clear(current->cap_effective);
	else
		current->cap_effective = current->cap_permitted;

387
	res = __user_walk_fd(dfd, filename, LOOKUP_FOLLOW|LOOKUP_ACCESS, &nd);
388 389 390 391 392 393 394 395 396 397 398
	if (res)
		goto out;

	res = vfs_permission(&nd, mode);
	/* SuS v2 requires we report a read only fs too */
	if(res || !(mode & S_IWOTH) ||
	   special_file(nd.dentry->d_inode->i_mode))
		goto out_path_release;

	if(IS_RDONLY(nd.dentry->d_inode))
		res = -EROFS;
Linus Torvalds's avatar
Linus Torvalds committed
399

400 401 402
out_path_release:
	path_release(&nd);
out:
Linus Torvalds's avatar
Linus Torvalds committed
403 404 405 406 407 408 409
	current->fsuid = old_fsuid;
	current->fsgid = old_fsgid;
	current->cap_effective = old_cap;

	return res;
}

410 411 412 413 414
asmlinkage long sys_access(const char __user *filename, int mode)
{
	return sys_faccessat(AT_FDCWD, filename, mode);
}

Linus Torvalds's avatar
Linus Torvalds committed
415 416 417 418 419
asmlinkage long sys_chdir(const char __user * filename)
{
	struct nameidata nd;
	int error;

420 421
	error = __user_walk(filename,
			    LOOKUP_FOLLOW|LOOKUP_DIRECTORY|LOOKUP_CHDIR, &nd);
Linus Torvalds's avatar
Linus Torvalds committed
422 423 424
	if (error)
		goto out;

425
	error = vfs_permission(&nd, MAY_EXEC);
Linus Torvalds's avatar
Linus Torvalds committed
426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449
	if (error)
		goto dput_and_out;

	set_fs_pwd(current->fs, nd.mnt, nd.dentry);

dput_and_out:
	path_release(&nd);
out:
	return error;
}

asmlinkage long sys_fchdir(unsigned int fd)
{
	struct file *file;
	struct dentry *dentry;
	struct inode *inode;
	struct vfsmount *mnt;
	int error;

	error = -EBADF;
	file = fget(fd);
	if (!file)
		goto out;

450 451
	dentry = file->f_path.dentry;
	mnt = file->f_path.mnt;
Linus Torvalds's avatar
Linus Torvalds committed
452 453 454 455 456 457
	inode = dentry->d_inode;

	error = -ENOTDIR;
	if (!S_ISDIR(inode->i_mode))
		goto out_putf;

458
	error = file_permission(file, MAY_EXEC);
Linus Torvalds's avatar
Linus Torvalds committed
459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475
	if (!error)
		set_fs_pwd(current->fs, mnt, dentry);
out_putf:
	fput(file);
out:
	return error;
}

asmlinkage long sys_chroot(const char __user * filename)
{
	struct nameidata nd;
	int error;

	error = __user_walk(filename, LOOKUP_FOLLOW | LOOKUP_DIRECTORY | LOOKUP_NOALT, &nd);
	if (error)
		goto out;

476
	error = vfs_permission(&nd, MAY_EXEC);
Linus Torvalds's avatar
Linus Torvalds committed
477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504
	if (error)
		goto dput_and_out;

	error = -EPERM;
	if (!capable(CAP_SYS_CHROOT))
		goto dput_and_out;

	set_fs_root(current->fs, nd.mnt, nd.dentry);
	set_fs_altroot();
	error = 0;
dput_and_out:
	path_release(&nd);
out:
	return error;
}

asmlinkage long sys_fchmod(unsigned int fd, mode_t mode)
{
	struct inode * inode;
	struct dentry * dentry;
	struct file * file;
	int err = -EBADF;
	struct iattr newattrs;

	file = fget(fd);
	if (!file)
		goto out;

505
	dentry = file->f_path.dentry;
Linus Torvalds's avatar
Linus Torvalds committed
506 507
	inode = dentry->d_inode;

508
	audit_inode(NULL, inode);
509

Linus Torvalds's avatar
Linus Torvalds committed
510 511 512 513 514 515
	err = -EROFS;
	if (IS_RDONLY(inode))
		goto out_putf;
	err = -EPERM;
	if (IS_IMMUTABLE(inode) || IS_APPEND(inode))
		goto out_putf;
516
	mutex_lock(&inode->i_mutex);
Linus Torvalds's avatar
Linus Torvalds committed
517 518 519 520 521
	if (mode == (mode_t) -1)
		mode = inode->i_mode;
	newattrs.ia_mode = (mode & S_IALLUGO) | (inode->i_mode & ~S_IALLUGO);
	newattrs.ia_valid = ATTR_MODE | ATTR_CTIME;
	err = notify_change(dentry, &newattrs);
522
	mutex_unlock(&inode->i_mutex);
Linus Torvalds's avatar
Linus Torvalds committed
523 524 525 526 527 528 529

out_putf:
	fput(file);
out:
	return err;
}

530 531
asmlinkage long sys_fchmodat(int dfd, const char __user *filename,
			     mode_t mode)
Linus Torvalds's avatar
Linus Torvalds committed
532 533 534 535 536 537
{
	struct nameidata nd;
	struct inode * inode;
	int error;
	struct iattr newattrs;

538
	error = __user_walk_fd(dfd, filename, LOOKUP_FOLLOW, &nd);
Linus Torvalds's avatar
Linus Torvalds committed
539 540 541 542 543 544 545 546 547 548 549 550
	if (error)
		goto out;
	inode = nd.dentry->d_inode;

	error = -EROFS;
	if (IS_RDONLY(inode))
		goto dput_and_out;

	error = -EPERM;
	if (IS_IMMUTABLE(inode) || IS_APPEND(inode))
		goto dput_and_out;

551
	mutex_lock(&inode->i_mutex);
Linus Torvalds's avatar
Linus Torvalds committed
552 553 554 555 556
	if (mode == (mode_t) -1)
		mode = inode->i_mode;
	newattrs.ia_mode = (mode & S_IALLUGO) | (inode->i_mode & ~S_IALLUGO);
	newattrs.ia_valid = ATTR_MODE | ATTR_CTIME;
	error = notify_change(nd.dentry, &newattrs);
557
	mutex_unlock(&inode->i_mutex);
Linus Torvalds's avatar
Linus Torvalds committed
558 559 560 561 562 563 564

dput_and_out:
	path_release(&nd);
out:
	return error;
}

565 566 567 568 569
asmlinkage long sys_chmod(const char __user *filename, mode_t mode)
{
	return sys_fchmodat(AT_FDCWD, filename, mode);
}

Linus Torvalds's avatar
Linus Torvalds committed
570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597
static int chown_common(struct dentry * dentry, uid_t user, gid_t group)
{
	struct inode * inode;
	int error;
	struct iattr newattrs;

	error = -ENOENT;
	if (!(inode = dentry->d_inode)) {
		printk(KERN_ERR "chown_common: NULL inode\n");
		goto out;
	}
	error = -EROFS;
	if (IS_RDONLY(inode))
		goto out;
	error = -EPERM;
	if (IS_IMMUTABLE(inode) || IS_APPEND(inode))
		goto out;
	newattrs.ia_valid =  ATTR_CTIME;
	if (user != (uid_t) -1) {
		newattrs.ia_valid |= ATTR_UID;
		newattrs.ia_uid = user;
	}
	if (group != (gid_t) -1) {
		newattrs.ia_valid |= ATTR_GID;
		newattrs.ia_gid = group;
	}
	if (!S_ISDIR(inode->i_mode))
		newattrs.ia_valid |= ATTR_KILL_SUID|ATTR_KILL_SGID;
598
	mutex_lock(&inode->i_mutex);
Linus Torvalds's avatar
Linus Torvalds committed
599
	error = notify_change(dentry, &newattrs);
600
	mutex_unlock(&inode->i_mutex);
Linus Torvalds's avatar
Linus Torvalds committed
601 602 603 604 605 606 607 608 609 610
out:
	return error;
}

asmlinkage long sys_chown(const char __user * filename, uid_t user, gid_t group)
{
	struct nameidata nd;
	int error;

	error = user_path_walk(filename, &nd);
611 612 613 614 615
	if (error)
		goto out;
	error = chown_common(nd.dentry, user, group);
	path_release(&nd);
out:
Linus Torvalds's avatar
Linus Torvalds committed
616 617 618
	return error;
}

619 620 621 622 623 624 625 626 627 628 629 630
asmlinkage long sys_fchownat(int dfd, const char __user *filename, uid_t user,
			     gid_t group, int flag)
{
	struct nameidata nd;
	int error = -EINVAL;
	int follow;

	if ((flag & ~AT_SYMLINK_NOFOLLOW) != 0)
		goto out;

	follow = (flag & AT_SYMLINK_NOFOLLOW) ? 0 : LOOKUP_FOLLOW;
	error = __user_walk_fd(dfd, filename, follow, &nd);
631 632 633 634
	if (error)
		goto out;
	error = chown_common(nd.dentry, user, group);
	path_release(&nd);
635 636 637 638
out:
	return error;
}

Linus Torvalds's avatar
Linus Torvalds committed
639 640 641 642 643 644
asmlinkage long sys_lchown(const char __user * filename, uid_t user, gid_t group)
{
	struct nameidata nd;
	int error;

	error = user_path_walk_link(filename, &nd);
645 646 647 648 649
	if (error)
		goto out;
	error = chown_common(nd.dentry, user, group);
	path_release(&nd);
out:
Linus Torvalds's avatar
Linus Torvalds committed
650 651 652 653 654 655 656 657
	return error;
}


asmlinkage long sys_fchown(unsigned int fd, uid_t user, gid_t group)
{
	struct file * file;
	int error = -EBADF;
658
	struct dentry * dentry;
Linus Torvalds's avatar
Linus Torvalds committed
659 660

	file = fget(fd);
661 662 663
	if (!file)
		goto out;

664
	dentry = file->f_path.dentry;
665 666 667 668
	audit_inode(NULL, dentry->d_inode);
	error = chown_common(dentry, user, group);
	fput(file);
out:
Linus Torvalds's avatar
Linus Torvalds committed
669 670 671
	return error;
}

672
static struct file *__dentry_open(struct dentry *dentry, struct vfsmount *mnt,
673 674
					int flags, struct file *f,
					int (*open)(struct inode *, struct file *))
Linus Torvalds's avatar
Linus Torvalds committed
675 676 677 678 679
{
	struct inode *inode;
	int error;

	f->f_flags = flags;
680 681
	f->f_mode = ((flags+1) & O_ACCMODE) | FMODE_LSEEK |
				FMODE_PREAD | FMODE_PWRITE;
Linus Torvalds's avatar
Linus Torvalds committed
682 683 684 685 686 687 688 689
	inode = dentry->d_inode;
	if (f->f_mode & FMODE_WRITE) {
		error = get_write_access(inode);
		if (error)
			goto cleanup_file;
	}

	f->f_mapping = inode->i_mapping;
690 691
	f->f_path.dentry = dentry;
	f->f_path.mnt = mnt;
Linus Torvalds's avatar
Linus Torvalds committed
692 693 694 695
	f->f_pos = 0;
	f->f_op = fops_get(inode->i_fop);
	file_move(f, &inode->i_sb->s_files);

696 697 698 699
	if (!open && f->f_op)
		open = f->f_op->open;
	if (open) {
		error = open(inode, f);
Linus Torvalds's avatar
Linus Torvalds committed
700 701 702
		if (error)
			goto cleanup_all;
	}
703

Linus Torvalds's avatar
Linus Torvalds committed
704 705 706 707 708 709
	f->f_flags &= ~(O_CREAT | O_EXCL | O_NOCTTY | O_TRUNC);

	file_ra_state_init(&f->f_ra, f->f_mapping->host->i_mapping);

	/* NB: we're sure to have correct a_ops only after f_op->open */
	if (f->f_flags & O_DIRECT) {
710 711 712
		if (!f->f_mapping->a_ops ||
		    ((!f->f_mapping->a_ops->direct_IO) &&
		    (!f->f_mapping->a_ops->get_xip_page))) {
Linus Torvalds's avatar
Linus Torvalds committed
713 714 715 716 717 718 719 720 721 722 723 724
			fput(f);
			f = ERR_PTR(-EINVAL);
		}
	}

	return f;

cleanup_all:
	fops_put(f->f_op);
	if (f->f_mode & FMODE_WRITE)
		put_write_access(inode);
	file_kill(f);
725 726
	f->f_path.dentry = NULL;
	f->f_path.mnt = NULL;
Linus Torvalds's avatar
Linus Torvalds committed
727 728 729 730 731 732 733
cleanup_file:
	put_filp(f);
	dput(dentry);
	mntput(mnt);
	return ERR_PTR(error);
}

734 735 736 737 738 739 740 741 742 743 744 745 746 747
/*
 * Note that while the flag value (low two bits) for sys_open means:
 *	00 - read-only
 *	01 - write-only
 *	10 - read-write
 *	11 - special
 * it is changed into
 *	00 - no permissions needed
 *	01 - read-permission
 *	10 - write-permission
 *	11 - read-write
 * for the internal routines (ie open_namei()/follow_link() etc). 00 is
 * used by symlinks.
 */
748 749
static struct file *do_filp_open(int dfd, const char *filename, int flags,
				 int mode)
750 751 752 753 754 755 756 757
{
	int namei_flags, error;
	struct nameidata nd;

	namei_flags = flags;
	if ((namei_flags+1) & O_ACCMODE)
		namei_flags++;

758
	error = open_namei(dfd, filename, namei_flags, mode, &nd);
759
	if (!error)
760
		return nameidata_to_filp(&nd, flags);
761 762 763

	return ERR_PTR(error);
}
764 765 766 767 768

struct file *filp_open(const char *filename, int flags, int mode)
{
	return do_filp_open(AT_FDCWD, filename, flags, mode);
}
769 770
EXPORT_SYMBOL(filp_open);

771 772 773 774 775 776 777 778 779 780
/**
 * lookup_instantiate_filp - instantiates the open intent filp
 * @nd: pointer to nameidata
 * @dentry: pointer to dentry
 * @open: open callback
 *
 * Helper for filesystems that want to use lookup open intents and pass back
 * a fully instantiated struct file to the caller.
 * This function is meant to be called from within a filesystem's
 * lookup method.
781 782 783 784
 * Beware of calling it for non-regular files! Those ->open methods might block
 * (e.g. in fifo_open), leaving you with parent locked (and in case of fifo,
 * leading to a deadlock, as nobody can open that fifo anymore, because
 * another process to open fifo will block on locked parent when doing lookup).
785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823
 * Note that in case of error, nd->intent.open.file is destroyed, but the
 * path information remains valid.
 * If the open callback is set to NULL, then the standard f_op->open()
 * filesystem callback is substituted.
 */
struct file *lookup_instantiate_filp(struct nameidata *nd, struct dentry *dentry,
		int (*open)(struct inode *, struct file *))
{
	if (IS_ERR(nd->intent.open.file))
		goto out;
	if (IS_ERR(dentry))
		goto out_err;
	nd->intent.open.file = __dentry_open(dget(dentry), mntget(nd->mnt),
					     nd->intent.open.flags - 1,
					     nd->intent.open.file,
					     open);
out:
	return nd->intent.open.file;
out_err:
	release_open_intent(nd);
	nd->intent.open.file = (struct file *)dentry;
	goto out;
}
EXPORT_SYMBOL_GPL(lookup_instantiate_filp);

/**
 * nameidata_to_filp - convert a nameidata to an open filp.
 * @nd: pointer to nameidata
 * @flags: open flags
 *
 * Note that this function destroys the original nameidata
 */
struct file *nameidata_to_filp(struct nameidata *nd, int flags)
{
	struct file *filp;

	/* Pick up the filp from the open intent */
	filp = nd->intent.open.file;
	/* Has the filesystem initialised the file for us? */
824
	if (filp->f_path.dentry == NULL)
825 826 827 828 829 830
		filp = __dentry_open(nd->dentry, nd->mnt, flags, filp, NULL);
	else
		path_release(nd);
	return filp;
}

831 832 833 834
/*
 * dentry_open() will have done dput(dentry) and mntput(mnt) if it returns an
 * error.
 */
835 836 837 838 839 840 841
struct file *dentry_open(struct dentry *dentry, struct vfsmount *mnt, int flags)
{
	int error;
	struct file *f;

	error = -ENFILE;
	f = get_empty_filp();
842 843 844
	if (f == NULL) {
		dput(dentry);
		mntput(mnt);
845
		return ERR_PTR(error);
846
	}
847

848
	return __dentry_open(dentry, mnt, flags, f, NULL);
849
}
Linus Torvalds's avatar
Linus Torvalds committed
850 851 852 853 854 855 856 857 858
EXPORT_SYMBOL(dentry_open);

/*
 * Find an empty file descriptor entry, and mark it busy.
 */
int get_unused_fd(void)
{
	struct files_struct * files = current->files;
	int fd, error;
859
	struct fdtable *fdt;
Linus Torvalds's avatar
Linus Torvalds committed
860 861 862 863 864

  	error = -EMFILE;
	spin_lock(&files->file_lock);

repeat:
865
	fdt = files_fdtable(files);
866
	fd = find_next_zero_bit(fdt->open_fds->fds_bits, fdt->max_fds,
867
				files->next_fd);
Linus Torvalds's avatar
Linus Torvalds committed
868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889

	/*
	 * N.B. For clone tasks sharing a files structure, this test
	 * will limit the total number of files that can be opened.
	 */
	if (fd >= current->signal->rlim[RLIMIT_NOFILE].rlim_cur)
		goto out;

	/* Do we need to expand the fd array or fd set?  */
	error = expand_files(files, fd);
	if (error < 0)
		goto out;

	if (error) {
		/*
	 	 * If we needed to expand the fs array we
		 * might have blocked - try again.
		 */
		error = -EMFILE;
		goto repeat;
	}

890 891
	FD_SET(fd, fdt->open_fds);
	FD_CLR(fd, fdt->close_on_exec);
892
	files->next_fd = fd + 1;
Linus Torvalds's avatar
Linus Torvalds committed
893 894
#if 1
	/* Sanity check */
895
	if (fdt->fd[fd] != NULL) {
Linus Torvalds's avatar
Linus Torvalds committed
896
		printk(KERN_WARNING "get_unused_fd: slot %d not NULL!\n", fd);
897
		fdt->fd[fd] = NULL;
Linus Torvalds's avatar
Linus Torvalds committed
898 899 900 901 902 903 904 905 906 907 908
	}
#endif
	error = fd;

out:
	spin_unlock(&files->file_lock);
	return error;
}

EXPORT_SYMBOL(get_unused_fd);

909
static void __put_unused_fd(struct files_struct *files, unsigned int fd)
Linus Torvalds's avatar
Linus Torvalds committed
910
{
911 912
	struct fdtable *fdt = files_fdtable(files);
	__FD_CLR(fd, fdt->open_fds);
913 914
	if (fd < files->next_fd)
		files->next_fd = fd;
Linus Torvalds's avatar
Linus Torvalds committed
915 916 917 918 919 920 921 922 923 924 925 926 927
}

void fastcall put_unused_fd(unsigned int fd)
{
	struct files_struct *files = current->files;
	spin_lock(&files->file_lock);
	__put_unused_fd(files, fd);
	spin_unlock(&files->file_lock);
}

EXPORT_SYMBOL(put_unused_fd);

/*
928
 * Install a file pointer in the fd array.
Linus Torvalds's avatar
Linus Torvalds committed
929 930 931 932 933 934 935 936 937 938 939 940 941 942
 *
 * The VFS is full of places where we drop the files lock between
 * setting the open_fds bitmap and installing the file in the file
 * array.  At any such point, we are vulnerable to a dup2() race
 * installing a file in the array before us.  We need to detect this and
 * fput() the struct file we are about to overwrite in this case.
 *
 * It should never happen - if we allow dup2() do it, _really_ bad things
 * will follow.
 */

void fastcall fd_install(unsigned int fd, struct file * file)
{
	struct files_struct *files = current->files;
943
	struct fdtable *fdt;
Linus Torvalds's avatar
Linus Torvalds committed
944
	spin_lock(&files->file_lock);
945
	fdt = files_fdtable(files);
946 947
	BUG_ON(fdt->fd[fd] != NULL);
	rcu_assign_pointer(fdt->fd[fd], file);
Linus Torvalds's avatar
Linus Torvalds committed
948 949 950 951 952
	spin_unlock(&files->file_lock);
}

EXPORT_SYMBOL(fd_install);

953
long do_sys_open(int dfd, const char __user *filename, int flags, int mode)
Linus Torvalds's avatar
Linus Torvalds committed
954
{
955 956
	char *tmp = getname(filename);
	int fd = PTR_ERR(tmp);
Linus Torvalds's avatar
Linus Torvalds committed
957 958 959 960

	if (!IS_ERR(tmp)) {
		fd = get_unused_fd();
		if (fd >= 0) {
961
			struct file *f = do_filp_open(dfd, tmp, flags, mode);
962 963 964 965
			if (IS_ERR(f)) {
				put_unused_fd(fd);
				fd = PTR_ERR(f);
			} else {
966
				fsnotify_open(f->f_path.dentry);
967 968
				fd_install(fd, f);
			}
Linus Torvalds's avatar
Linus Torvalds committed
969 970 971 972 973
		}
		putname(tmp);
	}
	return fd;
}
974 975 976

asmlinkage long sys_open(const char __user *filename, int flags, int mode)
{
977 978
	long ret;

979 980 981
	if (force_o_largefile())
		flags |= O_LARGEFILE;

982 983 984 985
	ret = do_sys_open(AT_FDCWD, filename, flags, mode);
	/* avoid REGPARM breakage on x86: */
	prevent_tail_call(ret);
	return ret;
986
}
Linus Torvalds's avatar
Linus Torvalds committed
987 988
EXPORT_SYMBOL_GPL(sys_open);

989 990 991
asmlinkage long sys_openat(int dfd, const char __user *filename, int flags,
			   int mode)
{
992 993
	long ret;

994 995 996
	if (force_o_largefile())
		flags |= O_LARGEFILE;

997 998 999 1000
	ret = do_sys_open(dfd, filename, flags, mode);
	/* avoid REGPARM breakage on x86: */
	prevent_tail_call(ret);
	return ret;
1001 1002
}

Linus Torvalds's avatar
Linus Torvalds committed
1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021
#ifndef __alpha__

/*
 * For backward compatibility?  Maybe this should be moved
 * into arch/i386 instead?
 */
asmlinkage long sys_creat(const char __user * pathname, int mode)
{
	return sys_open(pathname, O_CREAT | O_WRONLY | O_TRUNC, mode);
}

#endif

/*
 * "id" is the POSIX thread ID. We use the
 * files pointer for this..
 */
int filp_close(struct file *filp, fl_owner_t id)
{
1022
	int retval = 0;
Linus Torvalds's avatar
Linus Torvalds committed
1023 1024 1025

	if (!file_count(filp)) {
		printk(KERN_ERR "VFS: Close: file count is 0\n");
1026
		return 0;
Linus Torvalds's avatar
Linus Torvalds committed
1027 1028
	}

1029
	if (filp->f_op && filp->f_op->flush)
1030
		retval = filp->f_op->flush(filp, id);
Linus Torvalds's avatar
Linus Torvalds committed
1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048

	dnotify_flush(filp, id);
	locks_remove_posix(filp, id);
	fput(filp);
	return retval;
}

EXPORT_SYMBOL(filp_close);

/*
 * Careful here! We test whether the file pointer is NULL before
 * releasing the fd. This ensures that one clone task can't release
 * an fd while another clone is opening it.
 */
asmlinkage long sys_close(unsigned int fd)
{
	struct file * filp;
	struct files_struct *files = current->files;
1049
	struct fdtable *fdt;
1050
	int retval;
Linus Torvalds's avatar
Linus Torvalds committed
1051 1052

	spin_lock(&files->file_lock);
1053 1054
	fdt = files_fdtable(files);
	if (fd >= fdt->max_fds)
Linus Torvalds's avatar
Linus Torvalds committed
1055
		goto out_unlock;
1056
	filp = fdt->fd[fd];
Linus Torvalds's avatar
Linus Torvalds committed
1057 1058
	if (!filp)
		goto out_unlock;
1059
	rcu_assign_pointer(fdt->fd[fd], NULL);
1060
	FD_CLR(fd, fdt->close_on_exec);
Linus Torvalds's avatar
Linus Torvalds committed
1061 1062
	__put_unused_fd(files, fd);
	spin_unlock(&files->file_lock);
1063 1064 1065 1066 1067 1068 1069 1070 1071 1072
	retval = filp_close(filp, files);

	/* can't restart close syscall because file table entry was cleared */
	if (unlikely(retval == -ERESTARTSYS ||
		     retval == -ERESTARTNOINTR ||
		     retval == -ERESTARTNOHAND ||
		     retval == -ERESTART_RESTARTBLOCK))
		retval = -EINTR;

	return retval;
Linus Torvalds's avatar
Linus Torvalds committed
1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087

out_unlock:
	spin_unlock(&files->file_lock);
	return -EBADF;
}

EXPORT_SYMBOL(sys_close);

/*
 * This routine simulates a hangup on the tty, to arrange that users
 * are given clean terminals at login time.
 */
asmlinkage long sys_vhangup(void)
{
	if (capable(CAP_SYS_TTY_CONFIG)) {
1088
		/* XXX: this needs locking */
Linus Torvalds's avatar
Linus Torvalds committed
1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120
		tty_vhangup(current->signal->tty);
		return 0;
	}
	return -EPERM;
}

/*
 * Called when an inode is about to be open.
 * We use this to disallow opening large files on 32bit systems if
 * the caller didn't specify O_LARGEFILE.  On 64bit systems we force
 * on this flag in sys_open.
 */
int generic_file_open(struct inode * inode, struct file * filp)
{
	if (!(filp->f_flags & O_LARGEFILE) && i_size_read(inode) > MAX_NON_LFS)
		return -EFBIG;
	return 0;
}

EXPORT_SYMBOL(generic_file_open);

/*
 * This is used by subsystems that don't want seekable
 * file descriptors
 */
int nonseekable_open(struct inode *inode, struct file *filp)
{
	filp->f_mode &= ~(FMODE_LSEEK | FMODE_PREAD | FMODE_PWRITE);
	return 0;
}

EXPORT_SYMBOL(nonseekable_open);