fcntl.c 13.5 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7 8 9 10 11
/*
 *  linux/fs/fcntl.c
 *
 *  Copyright (C) 1991, 1992  Linus Torvalds
 */

#include <linux/syscalls.h>
#include <linux/init.h>
#include <linux/mm.h>
#include <linux/fs.h>
#include <linux/file.h>
A
Al Viro 已提交
12
#include <linux/fdtable.h>
13
#include <linux/capability.h>
L
Linus Torvalds 已提交
14 15 16 17 18
#include <linux/dnotify.h>
#include <linux/slab.h>
#include <linux/module.h>
#include <linux/security.h>
#include <linux/ptrace.h>
19
#include <linux/signal.h>
20
#include <linux/rcupdate.h>
21
#include <linux/pid_namespace.h>
22
#include <linux/smp_lock.h>
L
Linus Torvalds 已提交
23 24 25 26 27

#include <asm/poll.h>
#include <asm/siginfo.h>
#include <asm/uaccess.h>

28
void set_close_on_exec(unsigned int fd, int flag)
L
Linus Torvalds 已提交
29 30
{
	struct files_struct *files = current->files;
31
	struct fdtable *fdt;
L
Linus Torvalds 已提交
32
	spin_lock(&files->file_lock);
33
	fdt = files_fdtable(files);
L
Linus Torvalds 已提交
34
	if (flag)
35
		FD_SET(fd, fdt->close_on_exec);
L
Linus Torvalds 已提交
36
	else
37
		FD_CLR(fd, fdt->close_on_exec);
L
Linus Torvalds 已提交
38 39 40
	spin_unlock(&files->file_lock);
}

41
static int get_close_on_exec(unsigned int fd)
L
Linus Torvalds 已提交
42 43
{
	struct files_struct *files = current->files;
44
	struct fdtable *fdt;
L
Linus Torvalds 已提交
45
	int res;
46
	rcu_read_lock();
47 48
	fdt = files_fdtable(files);
	res = FD_ISSET(fd, fdt->close_on_exec);
49
	rcu_read_unlock();
L
Linus Torvalds 已提交
50 51 52
	return res;
}

U
Ulrich Drepper 已提交
53
asmlinkage long sys_dup3(unsigned int oldfd, unsigned int newfd, int flags)
L
Linus Torvalds 已提交
54 55 56 57
{
	int err = -EBADF;
	struct file * file, *tofree;
	struct files_struct * files = current->files;
58
	struct fdtable *fdt;
L
Linus Torvalds 已提交
59

U
Ulrich Drepper 已提交
60 61 62
	if ((flags & ~O_CLOEXEC) != 0)
		return -EINVAL;

63 64 65
	if (unlikely(oldfd == newfd))
		return -EINVAL;

L
Linus Torvalds 已提交
66 67
	spin_lock(&files->file_lock);
	err = expand_files(files, newfd);
A
Al Viro 已提交
68 69 70
	file = fcheck(oldfd);
	if (unlikely(!file))
		goto Ebadf;
A
Al Viro 已提交
71 72
	if (unlikely(err < 0)) {
		if (err == -EMFILE)
A
Al Viro 已提交
73 74
			goto Ebadf;
		goto out_unlock;
A
Al Viro 已提交
75
	}
A
Al Viro 已提交
76 77 78 79 80 81 82 83 84 85 86 87 88 89
	/*
	 * We need to detect attempts to do dup2() over allocated but still
	 * not finished descriptor.  NB: OpenBSD avoids that at the price of
	 * extra work in their equivalent of fget() - they insert struct
	 * file immediately after grabbing descriptor, mark it larval if
	 * more work (e.g. actual opening) is needed and make sure that
	 * fget() treats larval files as absent.  Potentially interesting,
	 * but while extra work in fget() is trivial, locking implications
	 * and amount of surgery on open()-related paths in VFS are not.
	 * FreeBSD fails with -EBADF in the same situation, NetBSD "solution"
	 * deadlocks in rather amusing ways, AFAICS.  All of that is out of
	 * scope of POSIX or SUS, since neither considers shared descriptor
	 * tables and this condition does not arise without those.
	 */
L
Linus Torvalds 已提交
90
	err = -EBUSY;
91 92 93
	fdt = files_fdtable(files);
	tofree = fdt->fd[newfd];
	if (!tofree && FD_ISSET(newfd, fdt->open_fds))
A
Al Viro 已提交
94 95
		goto out_unlock;
	get_file(file);
96
	rcu_assign_pointer(fdt->fd[newfd], file);
97
	FD_SET(newfd, fdt->open_fds);
U
Ulrich Drepper 已提交
98 99 100 101
	if (flags & O_CLOEXEC)
		FD_SET(newfd, fdt->close_on_exec);
	else
		FD_CLR(newfd, fdt->close_on_exec);
L
Linus Torvalds 已提交
102 103 104 105 106
	spin_unlock(&files->file_lock);

	if (tofree)
		filp_close(tofree, files);

A
Al Viro 已提交
107 108 109 110 111
	return newfd;

Ebadf:
	err = -EBADF;
out_unlock:
L
Linus Torvalds 已提交
112
	spin_unlock(&files->file_lock);
A
Al Viro 已提交
113
	return err;
L
Linus Torvalds 已提交
114
}
U
Ulrich Drepper 已提交
115 116 117

asmlinkage long sys_dup2(unsigned int oldfd, unsigned int newfd)
{
118 119 120 121 122 123 124 125
	if (unlikely(newfd == oldfd)) { /* corner case */
		struct files_struct *files = current->files;
		rcu_read_lock();
		if (!fcheck_files(files, oldfd))
			oldfd = -EBADF;
		rcu_read_unlock();
		return oldfd;
	}
U
Ulrich Drepper 已提交
126 127
	return sys_dup3(oldfd, newfd, 0);
}
L
Linus Torvalds 已提交
128 129 130 131

asmlinkage long sys_dup(unsigned int fildes)
{
	int ret = -EBADF;
132 133 134 135 136 137 138 139 140
	struct file *file = fget(fildes);

	if (file) {
		ret = get_unused_fd();
		if (ret >= 0)
			fd_install(ret, file);
		else
			fput(file);
	}
L
Linus Torvalds 已提交
141 142 143 144 145 146 147
	return ret;
}

#define SETFL_MASK (O_APPEND | O_NONBLOCK | O_NDELAY | FASYNC | O_DIRECT | O_NOATIME)

static int setfl(int fd, struct file * filp, unsigned long arg)
{
148
	struct inode * inode = filp->f_path.dentry->d_inode;
L
Linus Torvalds 已提交
149 150
	int error = 0;

151 152 153 154 155
	/*
	 * O_APPEND cannot be cleared if the file is marked as append-only
	 * and the file is open for write.
	 */
	if (((arg ^ filp->f_flags) & O_APPEND) && IS_APPEND(inode))
L
Linus Torvalds 已提交
156 157 158 159
		return -EPERM;

	/* O_NOATIME can only be set by the owner or superuser */
	if ((arg & O_NOATIME) && !(filp->f_flags & O_NOATIME))
160
		if (!is_owner_or_cap(inode))
L
Linus Torvalds 已提交
161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178
			return -EPERM;

	/* required for strict SunOS emulation */
	if (O_NONBLOCK != O_NDELAY)
	       if (arg & O_NDELAY)
		   arg |= O_NONBLOCK;

	if (arg & O_DIRECT) {
		if (!filp->f_mapping || !filp->f_mapping->a_ops ||
			!filp->f_mapping->a_ops->direct_IO)
				return -EINVAL;
	}

	if (filp->f_op && filp->f_op->check_flags)
		error = filp->f_op->check_flags(arg);
	if (error)
		return error;

179 180 181 182 183
	/*
	 * We still need a lock here for now to keep multiple FASYNC calls
	 * from racing with each other.
	 */
	lock_kernel();
L
Linus Torvalds 已提交
184 185 186 187 188 189 190 191 192 193
	if ((arg ^ filp->f_flags) & FASYNC) {
		if (filp->f_op && filp->f_op->fasync) {
			error = filp->f_op->fasync(fd, filp, (arg & FASYNC) != 0);
			if (error < 0)
				goto out;
		}
	}

	filp->f_flags = (arg & SETFL_MASK) | (filp->f_flags & ~SETFL_MASK);
 out:
194
	unlock_kernel();
L
Linus Torvalds 已提交
195 196 197
	return error;
}

198
static void f_modown(struct file *filp, struct pid *pid, enum pid_type type,
L
Linus Torvalds 已提交
199 200 201 202
                     uid_t uid, uid_t euid, int force)
{
	write_lock_irq(&filp->f_owner.lock);
	if (force || !filp->f_owner.pid) {
203 204 205
		put_pid(filp->f_owner.pid);
		filp->f_owner.pid = get_pid(pid);
		filp->f_owner.pid_type = type;
L
Linus Torvalds 已提交
206 207 208 209 210 211
		filp->f_owner.uid = uid;
		filp->f_owner.euid = euid;
	}
	write_unlock_irq(&filp->f_owner.lock);
}

212 213
int __f_setown(struct file *filp, struct pid *pid, enum pid_type type,
		int force)
L
Linus Torvalds 已提交
214 215 216 217 218 219 220
{
	int err;
	
	err = security_file_set_fowner(filp);
	if (err)
		return err;

221
	f_modown(filp, pid, type, current->uid, current->euid, force);
L
Linus Torvalds 已提交
222 223
	return 0;
}
224
EXPORT_SYMBOL(__f_setown);
L
Linus Torvalds 已提交
225

226 227 228 229 230 231 232 233 234 235 236 237
int f_setown(struct file *filp, unsigned long arg, int force)
{
	enum pid_type type;
	struct pid *pid;
	int who = arg;
	int result;
	type = PIDTYPE_PID;
	if (who < 0) {
		type = PIDTYPE_PGID;
		who = -who;
	}
	rcu_read_lock();
238
	pid = find_vpid(who);
239 240 241 242
	result = __f_setown(filp, pid, type, force);
	rcu_read_unlock();
	return result;
}
L
Linus Torvalds 已提交
243 244 245 246
EXPORT_SYMBOL(f_setown);

void f_delown(struct file *filp)
{
247 248 249 250 251 252
	f_modown(filp, NULL, PIDTYPE_PID, 0, 0, 1);
}

pid_t f_getown(struct file *filp)
{
	pid_t pid;
253
	read_lock(&filp->f_owner.lock);
254
	pid = pid_vnr(filp->f_owner.pid);
255 256
	if (filp->f_owner.pid_type == PIDTYPE_PGID)
		pid = -pid;
257
	read_unlock(&filp->f_owner.lock);
258
	return pid;
L
Linus Torvalds 已提交
259 260 261 262 263 264 265 266 267
}

static long do_fcntl(int fd, unsigned int cmd, unsigned long arg,
		struct file *filp)
{
	long err = -EINVAL;

	switch (cmd) {
	case F_DUPFD:
U
Ulrich Drepper 已提交
268
	case F_DUPFD_CLOEXEC:
A
Al Viro 已提交
269 270
		if (arg >= current->signal->rlim[RLIMIT_NOFILE].rlim_cur)
			break;
271 272 273 274 275
		err = alloc_fd(arg, cmd == F_DUPFD_CLOEXEC ? O_CLOEXEC : 0);
		if (err >= 0) {
			get_file(filp);
			fd_install(err, filp);
		}
L
Linus Torvalds 已提交
276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294
		break;
	case F_GETFD:
		err = get_close_on_exec(fd) ? FD_CLOEXEC : 0;
		break;
	case F_SETFD:
		err = 0;
		set_close_on_exec(fd, arg & FD_CLOEXEC);
		break;
	case F_GETFL:
		err = filp->f_flags;
		break;
	case F_SETFL:
		err = setfl(fd, filp, arg);
		break;
	case F_GETLK:
		err = fcntl_getlk(filp, (struct flock __user *) arg);
		break;
	case F_SETLK:
	case F_SETLKW:
295
		err = fcntl_setlk(fd, filp, cmd, (struct flock __user *) arg);
L
Linus Torvalds 已提交
296 297 298 299 300 301 302 303 304
		break;
	case F_GETOWN:
		/*
		 * XXX If f_owner is a process group, the
		 * negative return value will get converted
		 * into an error.  Oops.  If we keep the
		 * current syscall conventions, the only way
		 * to fix this will be in libc.
		 */
305
		err = f_getown(filp);
L
Linus Torvalds 已提交
306 307 308 309 310 311 312 313 314 315
		force_successful_syscall_return();
		break;
	case F_SETOWN:
		err = f_setown(filp, arg, 1);
		break;
	case F_GETSIG:
		err = filp->f_owner.signum;
		break;
	case F_SETSIG:
		/* arg == 0 restores default behaviour. */
316
		if (!valid_signal(arg)) {
L
Linus Torvalds 已提交
317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382
			break;
		}
		err = 0;
		filp->f_owner.signum = arg;
		break;
	case F_GETLEASE:
		err = fcntl_getlease(filp);
		break;
	case F_SETLEASE:
		err = fcntl_setlease(fd, filp, arg);
		break;
	case F_NOTIFY:
		err = fcntl_dirnotify(fd, filp, arg);
		break;
	default:
		break;
	}
	return err;
}

asmlinkage long sys_fcntl(unsigned int fd, unsigned int cmd, unsigned long arg)
{	
	struct file *filp;
	long err = -EBADF;

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

	err = security_file_fcntl(filp, cmd, arg);
	if (err) {
		fput(filp);
		return err;
	}

	err = do_fcntl(fd, cmd, arg, filp);

 	fput(filp);
out:
	return err;
}

#if BITS_PER_LONG == 32
asmlinkage long sys_fcntl64(unsigned int fd, unsigned int cmd, unsigned long arg)
{	
	struct file * filp;
	long err;

	err = -EBADF;
	filp = fget(fd);
	if (!filp)
		goto out;

	err = security_file_fcntl(filp, cmd, arg);
	if (err) {
		fput(filp);
		return err;
	}
	err = -EBADF;
	
	switch (cmd) {
		case F_GETLK64:
			err = fcntl_getlk64(filp, (struct flock64 __user *) arg);
			break;
		case F_SETLK64:
		case F_SETLKW64:
383 384
			err = fcntl_setlk64(fd, filp, cmd,
					(struct flock64 __user *) arg);
L
Linus Torvalds 已提交
385 386 387 388 389 390 391 392 393 394 395 396 397
			break;
		default:
			err = do_fcntl(fd, cmd, arg, filp);
			break;
	}
	fput(filp);
out:
	return err;
}
#endif

/* Table to convert sigio signal codes into poll band bitmaps */

398
static const long band_table[NSIGPOLL] = {
L
Linus Torvalds 已提交
399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438
	POLLIN | POLLRDNORM,			/* POLL_IN */
	POLLOUT | POLLWRNORM | POLLWRBAND,	/* POLL_OUT */
	POLLIN | POLLRDNORM | POLLMSG,		/* POLL_MSG */
	POLLERR,				/* POLL_ERR */
	POLLPRI | POLLRDBAND,			/* POLL_PRI */
	POLLHUP | POLLERR			/* POLL_HUP */
};

static inline int sigio_perm(struct task_struct *p,
                             struct fown_struct *fown, int sig)
{
	return (((fown->euid == 0) ||
		 (fown->euid == p->suid) || (fown->euid == p->uid) ||
		 (fown->uid == p->suid) || (fown->uid == p->uid)) &&
		!security_file_send_sigiotask(p, fown, sig));
}

static void send_sigio_to_task(struct task_struct *p,
			       struct fown_struct *fown, 
			       int fd,
			       int reason)
{
	if (!sigio_perm(p, fown, fown->signum))
		return;

	switch (fown->signum) {
		siginfo_t si;
		default:
			/* Queue a rt signal with the appropriate fd as its
			   value.  We use SI_SIGIO as the source, not 
			   SI_KERNEL, since kernel signals always get 
			   delivered even if we can't queue.  Failure to
			   queue in this case _should_ be reported; we fall
			   back to SIGIO in that case. --sct */
			si.si_signo = fown->signum;
			si.si_errno = 0;
		        si.si_code  = reason;
			/* Make sure we are called with one of the POLL_*
			   reasons, otherwise we could leak kernel stack into
			   userspace.  */
439
			BUG_ON((reason & __SI_MASK) != __SI_POLL);
L
Linus Torvalds 已提交
440 441 442 443 444
			if (reason - POLL_IN >= NSIGPOLL)
				si.si_band  = ~0L;
			else
				si.si_band = band_table[reason - POLL_IN];
			si.si_fd    = fd;
445
			if (!group_send_sig_info(fown->signum, &si, p))
L
Linus Torvalds 已提交
446 447 448
				break;
		/* fall-through: fall back on the old plain SIGIO signal */
		case 0:
449
			group_send_sig_info(SIGIO, SEND_SIG_PRIV, p);
L
Linus Torvalds 已提交
450 451 452 453 454 455
	}
}

void send_sigio(struct fown_struct *fown, int fd, int band)
{
	struct task_struct *p;
456 457
	enum pid_type type;
	struct pid *pid;
L
Linus Torvalds 已提交
458 459
	
	read_lock(&fown->lock);
460
	type = fown->pid_type;
L
Linus Torvalds 已提交
461 462 463 464 465
	pid = fown->pid;
	if (!pid)
		goto out_unlock_fown;
	
	read_lock(&tasklist_lock);
466 467 468
	do_each_pid_task(pid, type, p) {
		send_sigio_to_task(p, fown, fd, band);
	} while_each_pid_task(pid, type, p);
L
Linus Torvalds 已提交
469 470 471 472 473 474 475 476 477
	read_unlock(&tasklist_lock);
 out_unlock_fown:
	read_unlock(&fown->lock);
}

static void send_sigurg_to_task(struct task_struct *p,
                                struct fown_struct *fown)
{
	if (sigio_perm(p, fown, SIGURG))
478
		group_send_sig_info(SIGURG, SEND_SIG_PRIV, p);
L
Linus Torvalds 已提交
479 480 481 482 483
}

int send_sigurg(struct fown_struct *fown)
{
	struct task_struct *p;
484 485 486
	enum pid_type type;
	struct pid *pid;
	int ret = 0;
L
Linus Torvalds 已提交
487 488
	
	read_lock(&fown->lock);
489
	type = fown->pid_type;
L
Linus Torvalds 已提交
490 491 492 493 494 495 496
	pid = fown->pid;
	if (!pid)
		goto out_unlock_fown;

	ret = 1;
	
	read_lock(&tasklist_lock);
497 498 499
	do_each_pid_task(pid, type, p) {
		send_sigurg_to_task(p, fown);
	} while_each_pid_task(pid, type, p);
L
Linus Torvalds 已提交
500 501 502 503 504 505 506
	read_unlock(&tasklist_lock);
 out_unlock_fown:
	read_unlock(&fown->lock);
	return ret;
}

static DEFINE_RWLOCK(fasync_lock);
507
static struct kmem_cache *fasync_cache __read_mostly;
L
Linus Torvalds 已提交
508 509 510 511 512 513 514 515 516 517 518 519 520

/*
 * fasync_helper() is used by some character device drivers (mainly mice)
 * to set up the fasync queue. It returns negative on error, 0 if it did
 * no changes and positive if it added/deleted the entry.
 */
int fasync_helper(int fd, struct file * filp, int on, struct fasync_struct **fapp)
{
	struct fasync_struct *fa, **fp;
	struct fasync_struct *new = NULL;
	int result = 0;

	if (on) {
521
		new = kmem_cache_alloc(fasync_cache, GFP_KERNEL);
L
Linus Torvalds 已提交
522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592
		if (!new)
			return -ENOMEM;
	}
	write_lock_irq(&fasync_lock);
	for (fp = fapp; (fa = *fp) != NULL; fp = &fa->fa_next) {
		if (fa->fa_file == filp) {
			if(on) {
				fa->fa_fd = fd;
				kmem_cache_free(fasync_cache, new);
			} else {
				*fp = fa->fa_next;
				kmem_cache_free(fasync_cache, fa);
				result = 1;
			}
			goto out;
		}
	}

	if (on) {
		new->magic = FASYNC_MAGIC;
		new->fa_file = filp;
		new->fa_fd = fd;
		new->fa_next = *fapp;
		*fapp = new;
		result = 1;
	}
out:
	write_unlock_irq(&fasync_lock);
	return result;
}

EXPORT_SYMBOL(fasync_helper);

void __kill_fasync(struct fasync_struct *fa, int sig, int band)
{
	while (fa) {
		struct fown_struct * fown;
		if (fa->magic != FASYNC_MAGIC) {
			printk(KERN_ERR "kill_fasync: bad magic number in "
			       "fasync_struct!\n");
			return;
		}
		fown = &fa->fa_file->f_owner;
		/* Don't send SIGURG to processes which have not set a
		   queued signum: SIGURG has its own default signalling
		   mechanism. */
		if (!(sig == SIGURG && fown->signum == 0))
			send_sigio(fown, fa->fa_fd, band);
		fa = fa->fa_next;
	}
}

EXPORT_SYMBOL(__kill_fasync);

void kill_fasync(struct fasync_struct **fp, int sig, int band)
{
	/* First a quick test without locking: usually
	 * the list is empty.
	 */
	if (*fp) {
		read_lock(&fasync_lock);
		/* reread *fp after obtaining the lock */
		__kill_fasync(*fp, sig, band);
		read_unlock(&fasync_lock);
	}
}
EXPORT_SYMBOL(kill_fasync);

static int __init fasync_init(void)
{
	fasync_cache = kmem_cache_create("fasync_cache",
593
		sizeof(struct fasync_struct), 0, SLAB_PANIC, NULL);
L
Linus Torvalds 已提交
594 595 596 597
	return 0;
}

module_init(fasync_init)