open.c 24.4 KB
Newer Older
Linus Torvalds's avatar
Linus Torvalds committed
1 2 3 4 5 6 7 8 9 10 11
/*
 *  linux/fs/open.c
 *
 *  Copyright (C) 1991, 1992  Linus Torvalds
 */

#include <linux/string.h>
#include <linux/mm.h>
#include <linux/file.h>
#include <linux/smp_lock.h>
#include <linux/quotaops.h>
Robert Love's avatar
Robert Love committed
12
#include <linux/fsnotify.h>
Linus Torvalds's avatar
Linus Torvalds committed
13 14 15 16 17
#include <linux/module.h>
#include <linux/slab.h>
#include <linux/tty.h>
#include <linux/namei.h>
#include <linux/backing-dev.h>
18
#include <linux/capability.h>
Linus Torvalds's avatar
Linus Torvalds committed
19 20 21
#include <linux/security.h>
#include <linux/mount.h>
#include <linux/vfs.h>
22
#include <linux/fcntl.h>
Linus Torvalds's avatar
Linus Torvalds committed
23 24
#include <asm/uaccess.h>
#include <linux/fs.h>
25
#include <linux/personality.h>
Linus Torvalds's avatar
Linus Torvalds committed
26 27
#include <linux/pagemap.h>
#include <linux/syscalls.h>
28
#include <linux/rcupdate.h>
29
#include <linux/audit.h>
Linus Torvalds's avatar
Linus Torvalds committed
30

31
int vfs_statfs(struct dentry *dentry, struct kstatfs *buf)
Linus Torvalds's avatar
Linus Torvalds committed
32 33 34
{
	int retval = -ENODEV;

35
	if (dentry) {
Linus Torvalds's avatar
Linus Torvalds committed
36
		retval = -ENOSYS;
37
		if (dentry->d_sb->s_op->statfs) {
Linus Torvalds's avatar
Linus Torvalds committed
38
			memset(buf, 0, sizeof(*buf));
39
			retval = security_sb_statfs(dentry);
Linus Torvalds's avatar
Linus Torvalds committed
40 41
			if (retval)
				return retval;
42
			retval = dentry->d_sb->s_op->statfs(dentry, buf);
Linus Torvalds's avatar
Linus Torvalds committed
43 44 45 46 47 48 49 50 51
			if (retval == 0 && buf->f_frsize == 0)
				buf->f_frsize = buf->f_bsize;
		}
	}
	return retval;
}

EXPORT_SYMBOL(vfs_statfs);

52
static int vfs_statfs_native(struct dentry *dentry, struct statfs *buf)
Linus Torvalds's avatar
Linus Torvalds committed
53 54 55 56
{
	struct kstatfs st;
	int retval;

57
	retval = vfs_statfs(dentry, &st);
Linus Torvalds's avatar
Linus Torvalds committed
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
	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;
}

95
static int vfs_statfs64(struct dentry *dentry, struct statfs64 *buf)
Linus Torvalds's avatar
Linus Torvalds committed
96 97 98 99
{
	struct kstatfs st;
	int retval;

100
	retval = vfs_statfs(dentry, &st);
Linus Torvalds's avatar
Linus Torvalds committed
101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129
	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;
130
		error = vfs_statfs_native(nd.dentry, &tmp);
Linus Torvalds's avatar
Linus Torvalds committed
131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148
		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;
149
		error = vfs_statfs64(nd.dentry, &tmp);
Linus Torvalds's avatar
Linus Torvalds committed
150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167
		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;
168
	error = vfs_statfs_native(file->f_dentry, &tmp);
Linus Torvalds's avatar
Linus Torvalds committed
169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188
	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;
189
	error = vfs_statfs64(file->f_dentry, &tmp);
Linus Torvalds's avatar
Linus Torvalds committed
190 191 192 193 194 195 196
	if (!error && copy_to_user(buf, &tmp, sizeof(tmp)))
		error = -EFAULT;
	fput(file);
out:
	return error;
}

197 198
int do_truncate(struct dentry *dentry, loff_t length, unsigned int time_attrs,
	struct file *filp)
Linus Torvalds's avatar
Linus Torvalds committed
199 200 201 202 203 204 205 206 207
{
	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;
208
	newattrs.ia_valid = ATTR_SIZE | time_attrs;
209 210 211 212
	if (filp) {
		newattrs.ia_file = filp;
		newattrs.ia_valid |= ATTR_FILE;
	}
Linus Torvalds's avatar
Linus Torvalds committed
213

214
	mutex_lock(&dentry->d_inode->i_mutex);
Linus Torvalds's avatar
Linus Torvalds committed
215
	err = notify_change(dentry, &newattrs);
216
	mutex_unlock(&dentry->d_inode->i_mutex);
Linus Torvalds's avatar
Linus Torvalds committed
217 218 219
	return err;
}

220
static long do_sys_truncate(const char __user * path, loff_t length)
Linus Torvalds's avatar
Linus Torvalds committed
221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243
{
	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;

244
	error = vfs_permission(&nd, MAY_WRITE);
Linus Torvalds's avatar
Linus Torvalds committed
245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269
	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);
270
		error = do_truncate(nd.dentry, length, 0, NULL);
Linus Torvalds's avatar
Linus Torvalds committed
271 272 273 274 275 276 277 278 279 280 281 282 283 284 285
	}
	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);
}

286
static long do_sys_ftruncate(unsigned int fd, loff_t length, int small)
Linus Torvalds's avatar
Linus Torvalds committed
287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321
{
	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;

	dentry = file->f_dentry;
	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)
322
		error = do_truncate(dentry, length, ATTR_MTIME|ATTR_CTIME, file);
Linus Torvalds's avatar
Linus Torvalds committed
323 324 325 326 327 328 329 330
out_putf:
	fput(file);
out:
	return error;
}

asmlinkage long sys_ftruncate(unsigned int fd, unsigned long length)
{
331
	long ret = do_sys_ftruncate(fd, length, 1);
332
	/* avoid REGPARM breakage on x86: */
333 334
	prevent_tail_call(ret);
	return ret;
Linus Torvalds's avatar
Linus Torvalds committed
335 336 337 338 339 340 341 342 343 344 345
}

/* 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)
{
346
	long ret = do_sys_ftruncate(fd, length, 0);
347
	/* avoid REGPARM breakage on x86: */
348 349
	prevent_tail_call(ret);
	return ret;
Linus Torvalds's avatar
Linus Torvalds committed
350 351 352 353 354 355 356 357
}
#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.
 */
358
asmlinkage long sys_faccessat(int dfd, const char __user *filename, int mode)
Linus Torvalds's avatar
Linus Torvalds committed
359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387
{
	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;

388
	res = __user_walk_fd(dfd, filename, LOOKUP_FOLLOW|LOOKUP_ACCESS, &nd);
Linus Torvalds's avatar
Linus Torvalds committed
389
	if (!res) {
390
		res = vfs_permission(&nd, mode);
Linus Torvalds's avatar
Linus Torvalds committed
391 392 393 394 395 396 397 398 399 400 401 402 403 404
		/* SuS v2 requires we report a read only fs too */
		if(!res && (mode & S_IWOTH) && IS_RDONLY(nd.dentry->d_inode)
		   && !special_file(nd.dentry->d_inode->i_mode))
			res = -EROFS;
		path_release(&nd);
	}

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

	return res;
}

405 406 407 408 409
asmlinkage long sys_access(const char __user *filename, int mode)
{
	return sys_faccessat(AT_FDCWD, filename, mode);
}

Linus Torvalds's avatar
Linus Torvalds committed
410 411 412 413 414
asmlinkage long sys_chdir(const char __user * filename)
{
	struct nameidata nd;
	int error;

415 416
	error = __user_walk(filename,
			    LOOKUP_FOLLOW|LOOKUP_DIRECTORY|LOOKUP_CHDIR, &nd);
Linus Torvalds's avatar
Linus Torvalds committed
417 418 419
	if (error)
		goto out;

420
	error = vfs_permission(&nd, MAY_EXEC);
Linus Torvalds's avatar
Linus Torvalds committed
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
	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;

	dentry = file->f_dentry;
	mnt = file->f_vfsmnt;
	inode = dentry->d_inode;

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

453
	error = file_permission(file, MAY_EXEC);
Linus Torvalds's avatar
Linus Torvalds committed
454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470
	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;

471
	error = vfs_permission(&nd, MAY_EXEC);
Linus Torvalds's avatar
Linus Torvalds committed
472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502
	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;

	dentry = file->f_dentry;
	inode = dentry->d_inode;

503
	audit_inode(NULL, inode);
504

Linus Torvalds's avatar
Linus Torvalds committed
505 506 507 508 509 510
	err = -EROFS;
	if (IS_RDONLY(inode))
		goto out_putf;
	err = -EPERM;
	if (IS_IMMUTABLE(inode) || IS_APPEND(inode))
		goto out_putf;
511
	mutex_lock(&inode->i_mutex);
Linus Torvalds's avatar
Linus Torvalds committed
512 513 514 515 516
	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);
517
	mutex_unlock(&inode->i_mutex);
Linus Torvalds's avatar
Linus Torvalds committed
518 519 520 521 522 523 524

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

525 526
asmlinkage long sys_fchmodat(int dfd, const char __user *filename,
			     mode_t mode)
Linus Torvalds's avatar
Linus Torvalds committed
527 528 529 530 531 532
{
	struct nameidata nd;
	struct inode * inode;
	int error;
	struct iattr newattrs;

533
	error = __user_walk_fd(dfd, filename, LOOKUP_FOLLOW, &nd);
Linus Torvalds's avatar
Linus Torvalds committed
534 535 536 537 538 539 540 541 542 543 544 545
	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;

546
	mutex_lock(&inode->i_mutex);
Linus Torvalds's avatar
Linus Torvalds committed
547 548 549 550 551
	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);
552
	mutex_unlock(&inode->i_mutex);
Linus Torvalds's avatar
Linus Torvalds committed
553 554 555 556 557 558 559

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

560 561 562 563 564
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
565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592
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;
593
	mutex_lock(&inode->i_mutex);
Linus Torvalds's avatar
Linus Torvalds committed
594
	error = notify_change(dentry, &newattrs);
595
	mutex_unlock(&inode->i_mutex);
Linus Torvalds's avatar
Linus Torvalds committed
596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612
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);
	if (!error) {
		error = chown_common(nd.dentry, user, group);
		path_release(&nd);
	}
	return error;
}

613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632
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);
	if (!error) {
		error = chown_common(nd.dentry, user, group);
		path_release(&nd);
	}
out:
	return error;
}

Linus Torvalds's avatar
Linus Torvalds committed
633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653
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);
	if (!error) {
		error = chown_common(nd.dentry, user, group);
		path_release(&nd);
	}
	return error;
}


asmlinkage long sys_fchown(unsigned int fd, uid_t user, gid_t group)
{
	struct file * file;
	int error = -EBADF;

	file = fget(fd);
	if (file) {
654 655
		struct dentry * dentry;
		dentry = file->f_dentry;
656
		audit_inode(NULL, dentry->d_inode);
657
		error = chown_common(dentry, user, group);
Linus Torvalds's avatar
Linus Torvalds committed
658 659 660 661 662
		fput(file);
	}
	return error;
}

663
static struct file *__dentry_open(struct dentry *dentry, struct vfsmount *mnt,
664 665
					int flags, struct file *f,
					int (*open)(struct inode *, struct file *))
Linus Torvalds's avatar
Linus Torvalds committed
666 667 668 669 670
{
	struct inode *inode;
	int error;

	f->f_flags = flags;
671 672
	f->f_mode = ((flags+1) & O_ACCMODE) | FMODE_LSEEK |
				FMODE_PREAD | FMODE_PWRITE;
Linus Torvalds's avatar
Linus Torvalds committed
673 674 675 676 677 678 679 680 681 682 683 684 685 686
	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;
	f->f_dentry = dentry;
	f->f_vfsmnt = mnt;
	f->f_pos = 0;
	f->f_op = fops_get(inode->i_fop);
	file_move(f, &inode->i_sb->s_files);

687 688 689 690
	if (!open && f->f_op)
		open = f->f_op->open;
	if (open) {
		error = open(inode, f);
Linus Torvalds's avatar
Linus Torvalds committed
691 692 693
		if (error)
			goto cleanup_all;
	}
694

Linus Torvalds's avatar
Linus Torvalds committed
695 696 697 698 699 700
	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) {
701 702 703
		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
704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724
			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);
	f->f_dentry = NULL;
	f->f_vfsmnt = NULL;
cleanup_file:
	put_filp(f);
	dput(dentry);
	mntput(mnt);
	return ERR_PTR(error);
}

725 726 727 728 729 730 731 732 733 734 735 736 737 738
/*
 * 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.
 */
739 740
static struct file *do_filp_open(int dfd, const char *filename, int flags,
				 int mode)
741 742 743 744 745 746 747 748
{
	int namei_flags, error;
	struct nameidata nd;

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

749
	error = open_namei(dfd, filename, namei_flags, mode, &nd);
750
	if (!error)
751
		return nameidata_to_filp(&nd, flags);
752 753 754

	return ERR_PTR(error);
}
755 756 757 758 759

struct file *filp_open(const char *filename, int flags, int mode)
{
	return do_filp_open(AT_FDCWD, filename, flags, mode);
}
760 761
EXPORT_SYMBOL(filp_open);

762 763 764 765 766 767 768 769 770 771
/**
 * 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.
772 773 774 775
 * 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).
776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821
 * 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? */
	if (filp->f_dentry == NULL)
		filp = __dentry_open(nd->dentry, nd->mnt, flags, filp, NULL);
	else
		path_release(nd);
	return filp;
}

822 823 824 825
/*
 * dentry_open() will have done dput(dentry) and mntput(mnt) if it returns an
 * error.
 */
826 827 828 829 830 831 832
struct file *dentry_open(struct dentry *dentry, struct vfsmount *mnt, int flags)
{
	int error;
	struct file *f;

	error = -ENFILE;
	f = get_empty_filp();
833 834 835
	if (f == NULL) {
		dput(dentry);
		mntput(mnt);
836
		return ERR_PTR(error);
837
	}
838

839
	return __dentry_open(dentry, mnt, flags, f, NULL);
840
}
Linus Torvalds's avatar
Linus Torvalds committed
841 842 843 844 845 846 847 848 849
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;
850
	struct fdtable *fdt;
Linus Torvalds's avatar
Linus Torvalds committed
851 852 853 854 855

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

repeat:
856 857 858
	fdt = files_fdtable(files);
 	fd = find_next_zero_bit(fdt->open_fds->fds_bits,
				fdt->max_fdset,
859
				files->next_fd);
Linus Torvalds's avatar
Linus Torvalds committed
860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881

	/*
	 * 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;
	}

882 883
	FD_SET(fd, fdt->open_fds);
	FD_CLR(fd, fdt->close_on_exec);
884
	files->next_fd = fd + 1;
Linus Torvalds's avatar
Linus Torvalds committed
885 886
#if 1
	/* Sanity check */
887
	if (fdt->fd[fd] != NULL) {
Linus Torvalds's avatar
Linus Torvalds committed
888
		printk(KERN_WARNING "get_unused_fd: slot %d not NULL!\n", fd);
889
		fdt->fd[fd] = NULL;
Linus Torvalds's avatar
Linus Torvalds committed
890 891 892 893 894 895 896 897 898 899 900
	}
#endif
	error = fd;

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

EXPORT_SYMBOL(get_unused_fd);

901
static void __put_unused_fd(struct files_struct *files, unsigned int fd)
Linus Torvalds's avatar
Linus Torvalds committed
902
{
903 904
	struct fdtable *fdt = files_fdtable(files);
	__FD_CLR(fd, fdt->open_fds);
905 906
	if (fd < files->next_fd)
		files->next_fd = fd;
Linus Torvalds's avatar
Linus Torvalds committed
907 908 909 910 911 912 913 914 915 916 917 918 919
}

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);

/*
920
 * Install a file pointer in the fd array.
Linus Torvalds's avatar
Linus Torvalds committed
921 922 923 924 925 926 927 928 929 930 931 932 933 934
 *
 * 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;
935
	struct fdtable *fdt;
Linus Torvalds's avatar
Linus Torvalds committed
936
	spin_lock(&files->file_lock);
937
	fdt = files_fdtable(files);
938 939
	BUG_ON(fdt->fd[fd] != NULL);
	rcu_assign_pointer(fdt->fd[fd], file);
Linus Torvalds's avatar
Linus Torvalds committed
940 941 942 943 944
	spin_unlock(&files->file_lock);
}

EXPORT_SYMBOL(fd_install);

945
long do_sys_open(int dfd, const char __user *filename, int flags, int mode)
Linus Torvalds's avatar
Linus Torvalds committed
946
{
947 948
	char *tmp = getname(filename);
	int fd = PTR_ERR(tmp);
Linus Torvalds's avatar
Linus Torvalds committed
949 950 951 952

	if (!IS_ERR(tmp)) {
		fd = get_unused_fd();
		if (fd >= 0) {
953
			struct file *f = do_filp_open(dfd, tmp, flags, mode);
954 955 956 957
			if (IS_ERR(f)) {
				put_unused_fd(fd);
				fd = PTR_ERR(f);
			} else {
Robert Love's avatar
Robert Love committed
958
				fsnotify_open(f->f_dentry);
959 960
				fd_install(fd, f);
			}
Linus Torvalds's avatar
Linus Torvalds committed
961 962 963 964 965
		}
		putname(tmp);
	}
	return fd;
}
966 967 968

asmlinkage long sys_open(const char __user *filename, int flags, int mode)
{
969 970
	long ret;

971 972 973
	if (force_o_largefile())
		flags |= O_LARGEFILE;

974 975 976 977
	ret = do_sys_open(AT_FDCWD, filename, flags, mode);
	/* avoid REGPARM breakage on x86: */
	prevent_tail_call(ret);
	return ret;
978
}
Linus Torvalds's avatar
Linus Torvalds committed
979 980
EXPORT_SYMBOL_GPL(sys_open);

981 982 983
asmlinkage long sys_openat(int dfd, const char __user *filename, int flags,
			   int mode)
{
984 985
	long ret;

986 987 988
	if (force_o_largefile())
		flags |= O_LARGEFILE;

989 990 991 992
	ret = do_sys_open(dfd, filename, flags, mode);
	/* avoid REGPARM breakage on x86: */
	prevent_tail_call(ret);
	return ret;
993 994
}

Linus Torvalds's avatar
Linus Torvalds committed
995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013
#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)
{
1014
	int retval = 0;
Linus Torvalds's avatar
Linus Torvalds committed
1015 1016 1017

	if (!file_count(filp)) {
		printk(KERN_ERR "VFS: Close: file count is 0\n");
1018
		return 0;
Linus Torvalds's avatar
Linus Torvalds committed
1019 1020
	}

1021
	if (filp->f_op && filp->f_op->flush)
1022
		retval = filp->f_op->flush(filp, id);
Linus Torvalds's avatar
Linus Torvalds committed
1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040

	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;
1041
	struct fdtable *fdt;
1042
	int retval;
Linus Torvalds's avatar
Linus Torvalds committed
1043 1044

	spin_lock(&files->file_lock);
1045 1046
	fdt = files_fdtable(files);
	if (fd >= fdt->max_fds)
Linus Torvalds's avatar
Linus Torvalds committed
1047
		goto out_unlock;
1048
	filp = fdt->fd[fd];
Linus Torvalds's avatar
Linus Torvalds committed
1049 1050
	if (!filp)
		goto out_unlock;
1051
	rcu_assign_pointer(fdt->fd[fd], NULL);
1052
	FD_CLR(fd, fdt->close_on_exec);
Linus Torvalds's avatar
Linus Torvalds committed
1053 1054
	__put_unused_fd(files, fd);
	spin_unlock(&files->file_lock);
1055 1056 1057 1058 1059 1060 1061 1062 1063 1064
	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
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 1103 1104 1105 1106 1107 1108 1109 1110 1111

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)) {
		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);