libfs.c 25.8 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5
/*
 *	fs/libfs.c
 *	Library for filesystems writers.
 */

6
#include <linux/export.h>
L
Linus Torvalds 已提交
7
#include <linux/pagemap.h>
8
#include <linux/slab.h>
L
Linus Torvalds 已提交
9 10
#include <linux/mount.h>
#include <linux/vfs.h>
11
#include <linux/quotaops.h>
I
Ingo Molnar 已提交
12
#include <linux/mutex.h>
C
Christoph Hellwig 已提交
13
#include <linux/exportfs.h>
A
Al Viro 已提交
14
#include <linux/writeback.h>
A
Al Viro 已提交
15
#include <linux/buffer_head.h> /* sync_mapping_buffers */
I
Ingo Molnar 已提交
16

L
Linus Torvalds 已提交
17 18
#include <asm/uaccess.h>

19 20
#include "internal.h"

N
Nick Piggin 已提交
21 22 23 24 25
static inline int simple_positive(struct dentry *dentry)
{
	return dentry->d_inode && !d_unhashed(dentry);
}

L
Linus Torvalds 已提交
26 27 28 29 30 31 32 33 34
int simple_getattr(struct vfsmount *mnt, struct dentry *dentry,
		   struct kstat *stat)
{
	struct inode *inode = dentry->d_inode;
	generic_fillattr(inode, stat);
	stat->blocks = inode->i_mapping->nrpages << (PAGE_CACHE_SHIFT - 9);
	return 0;
}

35
int simple_statfs(struct dentry *dentry, struct kstatfs *buf)
L
Linus Torvalds 已提交
36
{
37
	buf->f_type = dentry->d_sb->s_magic;
L
Linus Torvalds 已提交
38 39 40 41 42 43 44 45 46
	buf->f_bsize = PAGE_CACHE_SIZE;
	buf->f_namelen = NAME_MAX;
	return 0;
}

/*
 * Retaining negative dentries for an in-memory filesystem just wastes
 * memory and lookup time: arrange for them to be deleted immediately.
 */
N
Nick Piggin 已提交
47
static int simple_delete_dentry(const struct dentry *dentry)
L
Linus Torvalds 已提交
48 49 50 51 52 53 54 55
{
	return 1;
}

/*
 * Lookup the data. This is trivial - if the dentry didn't already
 * exist, we know it is negative.  Set d_op to delete negative dentries.
 */
A
Al Viro 已提交
56
struct dentry *simple_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags)
L
Linus Torvalds 已提交
57
{
A
Al Viro 已提交
58
	static const struct dentry_operations simple_dentry_operations = {
L
Linus Torvalds 已提交
59 60 61 62 63
		.d_delete = simple_delete_dentry,
	};

	if (dentry->d_name.len > NAME_MAX)
		return ERR_PTR(-ENAMETOOLONG);
64
	d_set_d_op(dentry, &simple_dentry_operations);
L
Linus Torvalds 已提交
65 66 67 68 69 70
	d_add(dentry, NULL);
	return NULL;
}

int dcache_dir_open(struct inode *inode, struct file *file)
{
71
	static struct qstr cursor_name = QSTR_INIT(".", 1);
L
Linus Torvalds 已提交
72

73
	file->private_data = d_alloc(file->f_path.dentry, &cursor_name);
L
Linus Torvalds 已提交
74 75 76 77 78 79 80 81 82 83 84 85

	return file->private_data ? 0 : -ENOMEM;
}

int dcache_dir_close(struct inode *inode, struct file *file)
{
	dput(file->private_data);
	return 0;
}

loff_t dcache_dir_lseek(struct file *file, loff_t offset, int origin)
{
N
Nick Piggin 已提交
86 87
	struct dentry *dentry = file->f_path.dentry;
	mutex_lock(&dentry->d_inode->i_mutex);
L
Linus Torvalds 已提交
88 89 90 91 92 93 94
	switch (origin) {
		case 1:
			offset += file->f_pos;
		case 0:
			if (offset >= 0)
				break;
		default:
N
Nick Piggin 已提交
95
			mutex_unlock(&dentry->d_inode->i_mutex);
L
Linus Torvalds 已提交
96 97 98 99 100 101 102 103 104
			return -EINVAL;
	}
	if (offset != file->f_pos) {
		file->f_pos = offset;
		if (file->f_pos >= 2) {
			struct list_head *p;
			struct dentry *cursor = file->private_data;
			loff_t n = file->f_pos - 2;

N
Nick Piggin 已提交
105 106
			spin_lock(&dentry->d_lock);
			/* d_lock not required for cursor */
E
Eric Dumazet 已提交
107
			list_del(&cursor->d_u.d_child);
N
Nick Piggin 已提交
108 109
			p = dentry->d_subdirs.next;
			while (n && p != &dentry->d_subdirs) {
L
Linus Torvalds 已提交
110
				struct dentry *next;
E
Eric Dumazet 已提交
111
				next = list_entry(p, struct dentry, d_u.d_child);
N
Nick Piggin 已提交
112
				spin_lock_nested(&next->d_lock, DENTRY_D_LOCK_NESTED);
N
Nick Piggin 已提交
113
				if (simple_positive(next))
L
Linus Torvalds 已提交
114
					n--;
N
Nick Piggin 已提交
115
				spin_unlock(&next->d_lock);
L
Linus Torvalds 已提交
116 117
				p = p->next;
			}
E
Eric Dumazet 已提交
118
			list_add_tail(&cursor->d_u.d_child, p);
N
Nick Piggin 已提交
119
			spin_unlock(&dentry->d_lock);
L
Linus Torvalds 已提交
120 121
		}
	}
N
Nick Piggin 已提交
122
	mutex_unlock(&dentry->d_inode->i_mutex);
L
Linus Torvalds 已提交
123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139
	return offset;
}

/* Relationship between i_mode and the DT_xxx types */
static inline unsigned char dt_type(struct inode *inode)
{
	return (inode->i_mode >> 12) & 15;
}

/*
 * Directory is locked and all positive dentries in it are safe, since
 * for ramfs-type trees they can't go away without unlink() or rmdir(),
 * both impossible due to the lock on directory.
 */

int dcache_readdir(struct file * filp, void * dirent, filldir_t filldir)
{
140
	struct dentry *dentry = filp->f_path.dentry;
L
Linus Torvalds 已提交
141
	struct dentry *cursor = filp->private_data;
E
Eric Dumazet 已提交
142
	struct list_head *p, *q = &cursor->d_u.d_child;
L
Linus Torvalds 已提交
143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161
	ino_t ino;
	int i = filp->f_pos;

	switch (i) {
		case 0:
			ino = dentry->d_inode->i_ino;
			if (filldir(dirent, ".", 1, i, ino, DT_DIR) < 0)
				break;
			filp->f_pos++;
			i++;
			/* fallthrough */
		case 1:
			ino = parent_ino(dentry);
			if (filldir(dirent, "..", 2, i, ino, DT_DIR) < 0)
				break;
			filp->f_pos++;
			i++;
			/* fallthrough */
		default:
N
Nick Piggin 已提交
162
			spin_lock(&dentry->d_lock);
A
Akinobu Mita 已提交
163 164 165
			if (filp->f_pos == 2)
				list_move(q, &dentry->d_subdirs);

L
Linus Torvalds 已提交
166 167
			for (p=q->next; p != &dentry->d_subdirs; p=p->next) {
				struct dentry *next;
E
Eric Dumazet 已提交
168
				next = list_entry(p, struct dentry, d_u.d_child);
N
Nick Piggin 已提交
169 170 171
				spin_lock_nested(&next->d_lock, DENTRY_D_LOCK_NESTED);
				if (!simple_positive(next)) {
					spin_unlock(&next->d_lock);
L
Linus Torvalds 已提交
172
					continue;
N
Nick Piggin 已提交
173
				}
L
Linus Torvalds 已提交
174

N
Nick Piggin 已提交
175
				spin_unlock(&next->d_lock);
N
Nick Piggin 已提交
176
				spin_unlock(&dentry->d_lock);
177 178 179 180
				if (filldir(dirent, next->d_name.name, 
					    next->d_name.len, filp->f_pos, 
					    next->d_inode->i_ino, 
					    dt_type(next->d_inode)) < 0)
L
Linus Torvalds 已提交
181
					return 0;
N
Nick Piggin 已提交
182 183
				spin_lock(&dentry->d_lock);
				spin_lock_nested(&next->d_lock, DENTRY_D_LOCK_NESTED);
L
Linus Torvalds 已提交
184
				/* next is still alive */
A
Akinobu Mita 已提交
185
				list_move(q, p);
N
Nick Piggin 已提交
186
				spin_unlock(&next->d_lock);
L
Linus Torvalds 已提交
187 188 189
				p = q;
				filp->f_pos++;
			}
N
Nick Piggin 已提交
190
			spin_unlock(&dentry->d_lock);
L
Linus Torvalds 已提交
191 192 193 194 195 196 197 198 199
	}
	return 0;
}

ssize_t generic_read_dir(struct file *filp, char __user *buf, size_t siz, loff_t *ppos)
{
	return -EISDIR;
}

200
const struct file_operations simple_dir_operations = {
L
Linus Torvalds 已提交
201 202 203 204 205
	.open		= dcache_dir_open,
	.release	= dcache_dir_close,
	.llseek		= dcache_dir_lseek,
	.read		= generic_read_dir,
	.readdir	= dcache_readdir,
206
	.fsync		= noop_fsync,
L
Linus Torvalds 已提交
207 208
};

209
const struct inode_operations simple_dir_inode_operations = {
L
Linus Torvalds 已提交
210 211 212
	.lookup		= simple_lookup,
};

213 214 215 216
static const struct super_operations simple_super_operations = {
	.statfs		= simple_statfs,
};

L
Linus Torvalds 已提交
217 218 219 220
/*
 * Common helper for pseudo-filesystems (sockfs, pipefs, bdev - stuff that
 * will never be mountable)
 */
A
Al Viro 已提交
221
struct dentry *mount_pseudo(struct file_system_type *fs_type, char *name,
222 223
	const struct super_operations *ops,
	const struct dentry_operations *dops, unsigned long magic)
L
Linus Torvalds 已提交
224
{
D
David Howells 已提交
225
	struct super_block *s;
L
Linus Torvalds 已提交
226 227
	struct dentry *dentry;
	struct inode *root;
228
	struct qstr d_name = QSTR_INIT(name, strlen(name));
L
Linus Torvalds 已提交
229

D
David Howells 已提交
230
	s = sget(fs_type, NULL, set_anon_super, MS_NOUSER, NULL);
L
Linus Torvalds 已提交
231
	if (IS_ERR(s))
A
Al Viro 已提交
232
		return ERR_CAST(s);
L
Linus Torvalds 已提交
233

234
	s->s_maxbytes = MAX_LFS_FILESIZE;
235 236
	s->s_blocksize = PAGE_SIZE;
	s->s_blocksize_bits = PAGE_SHIFT;
L
Linus Torvalds 已提交
237
	s->s_magic = magic;
238
	s->s_op = ops ? ops : &simple_super_operations;
L
Linus Torvalds 已提交
239 240 241 242
	s->s_time_gran = 1;
	root = new_inode(s);
	if (!root)
		goto Enomem;
243 244 245 246 247 248
	/*
	 * since this is the first inode, make it number 1. New inodes created
	 * after this must take care not to collide with it (by passing
	 * max_reserved of 1 to iunique).
	 */
	root->i_ino = 1;
L
Linus Torvalds 已提交
249 250
	root->i_mode = S_IFDIR | S_IRUSR | S_IWUSR;
	root->i_atime = root->i_mtime = root->i_ctime = CURRENT_TIME;
251
	dentry = __d_alloc(s, &d_name);
L
Linus Torvalds 已提交
252 253 254 255 256 257
	if (!dentry) {
		iput(root);
		goto Enomem;
	}
	d_instantiate(dentry, root);
	s->s_root = dentry;
258
	s->s_d_op = dops;
L
Linus Torvalds 已提交
259
	s->s_flags |= MS_ACTIVE;
A
Al Viro 已提交
260
	return dget(s->s_root);
L
Linus Torvalds 已提交
261 262

Enomem:
263
	deactivate_locked_super(s);
A
Al Viro 已提交
264
	return ERR_PTR(-ENOMEM);
L
Linus Torvalds 已提交
265 266
}

S
Stephen Boyd 已提交
267 268 269 270 271 272 273
int simple_open(struct inode *inode, struct file *file)
{
	if (inode->i_private)
		file->private_data = inode->i_private;
	return 0;
}

L
Linus Torvalds 已提交
274 275 276 277 278
int simple_link(struct dentry *old_dentry, struct inode *dir, struct dentry *dentry)
{
	struct inode *inode = old_dentry->d_inode;

	inode->i_ctime = dir->i_ctime = dir->i_mtime = CURRENT_TIME;
279
	inc_nlink(inode);
A
Al Viro 已提交
280
	ihold(inode);
L
Linus Torvalds 已提交
281 282 283 284 285 286 287 288 289 290
	dget(dentry);
	d_instantiate(dentry, inode);
	return 0;
}

int simple_empty(struct dentry *dentry)
{
	struct dentry *child;
	int ret = 0;

N
Nick Piggin 已提交
291
	spin_lock(&dentry->d_lock);
N
Nick Piggin 已提交
292 293 294 295
	list_for_each_entry(child, &dentry->d_subdirs, d_u.d_child) {
		spin_lock_nested(&child->d_lock, DENTRY_D_LOCK_NESTED);
		if (simple_positive(child)) {
			spin_unlock(&child->d_lock);
L
Linus Torvalds 已提交
296
			goto out;
N
Nick Piggin 已提交
297 298 299
		}
		spin_unlock(&child->d_lock);
	}
L
Linus Torvalds 已提交
300 301
	ret = 1;
out:
N
Nick Piggin 已提交
302
	spin_unlock(&dentry->d_lock);
L
Linus Torvalds 已提交
303 304 305 306 307 308 309 310
	return ret;
}

int simple_unlink(struct inode *dir, struct dentry *dentry)
{
	struct inode *inode = dentry->d_inode;

	inode->i_ctime = dir->i_ctime = dir->i_mtime = CURRENT_TIME;
311
	drop_nlink(inode);
L
Linus Torvalds 已提交
312 313 314 315 316 317 318 319 320
	dput(dentry);
	return 0;
}

int simple_rmdir(struct inode *dir, struct dentry *dentry)
{
	if (!simple_empty(dentry))
		return -ENOTEMPTY;

321
	drop_nlink(dentry->d_inode);
L
Linus Torvalds 已提交
322
	simple_unlink(dir, dentry);
323
	drop_nlink(dir);
L
Linus Torvalds 已提交
324 325 326 327 328 329 330 331 332 333 334 335 336 337
	return 0;
}

int simple_rename(struct inode *old_dir, struct dentry *old_dentry,
		struct inode *new_dir, struct dentry *new_dentry)
{
	struct inode *inode = old_dentry->d_inode;
	int they_are_dirs = S_ISDIR(old_dentry->d_inode->i_mode);

	if (!simple_empty(new_dentry))
		return -ENOTEMPTY;

	if (new_dentry->d_inode) {
		simple_unlink(new_dir, new_dentry);
338 339
		if (they_are_dirs) {
			drop_nlink(new_dentry->d_inode);
340
			drop_nlink(old_dir);
341
		}
L
Linus Torvalds 已提交
342
	} else if (they_are_dirs) {
343
		drop_nlink(old_dir);
344
		inc_nlink(new_dir);
L
Linus Torvalds 已提交
345 346 347 348 349 350 351 352
	}

	old_dir->i_ctime = old_dir->i_mtime = new_dir->i_ctime =
		new_dir->i_mtime = inode->i_ctime = CURRENT_TIME;

	return 0;
}

353
/**
C
Christoph Hellwig 已提交
354
 * simple_setattr - setattr for simple filesystem
355 356 357 358 359
 * @dentry: dentry
 * @iattr: iattr structure
 *
 * Returns 0 on success, -error on failure.
 *
C
Christoph Hellwig 已提交
360 361 362 363 364 365
 * simple_setattr is a simple ->setattr implementation without a proper
 * implementation of size changes.
 *
 * It can either be used for in-memory filesystems or special files
 * on simple regular filesystems.  Anything that needs to change on-disk
 * or wire state on size changes needs its own setattr method.
366 367 368 369 370 371
 */
int simple_setattr(struct dentry *dentry, struct iattr *iattr)
{
	struct inode *inode = dentry->d_inode;
	int error;

C
Christoph Hellwig 已提交
372 373
	WARN_ON_ONCE(inode->i_op->truncate);

374 375 376 377
	error = inode_change_ok(inode, iattr);
	if (error)
		return error;

378 379
	if (iattr->ia_valid & ATTR_SIZE)
		truncate_setsize(inode, iattr->ia_size);
C
Christoph Hellwig 已提交
380
	setattr_copy(inode, iattr);
C
Christoph Hellwig 已提交
381 382
	mark_inode_dirty(inode);
	return 0;
383 384 385
}
EXPORT_SYMBOL(simple_setattr);

L
Linus Torvalds 已提交
386 387
int simple_readpage(struct file *file, struct page *page)
{
388
	clear_highpage(page);
L
Linus Torvalds 已提交
389 390 391 392 393 394
	flush_dcache_page(page);
	SetPageUptodate(page);
	unlock_page(page);
	return 0;
}

395 396 397 398 399 400 401 402 403
int simple_write_begin(struct file *file, struct address_space *mapping,
			loff_t pos, unsigned len, unsigned flags,
			struct page **pagep, void **fsdata)
{
	struct page *page;
	pgoff_t index;

	index = pos >> PAGE_CACHE_SHIFT;

404
	page = grab_cache_page_write_begin(mapping, index, flags);
405 406 407 408 409
	if (!page)
		return -ENOMEM;

	*pagep = page;

410 411 412 413 414 415
	if (!PageUptodate(page) && (len != PAGE_CACHE_SIZE)) {
		unsigned from = pos & (PAGE_CACHE_SIZE - 1);

		zero_user_segments(page, 0, from, from + len, PAGE_CACHE_SIZE);
	}
	return 0;
416 417
}

418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438
/**
 * simple_write_end - .write_end helper for non-block-device FSes
 * @available: See .write_end of address_space_operations
 * @file: 		"
 * @mapping: 		"
 * @pos: 		"
 * @len: 		"
 * @copied: 		"
 * @page: 		"
 * @fsdata: 		"
 *
 * simple_write_end does the minimum needed for updating a page after writing is
 * done. It has the same API signature as the .write_end of
 * address_space_operations vector. So it can just be set onto .write_end for
 * FSes that don't need any other processing. i_mutex is assumed to be held.
 * Block based filesystems should use generic_write_end().
 * NOTE: Even though i_size might get updated by this function, mark_inode_dirty
 * is not called, so a filesystem that actually does store data in .write_inode
 * should extend on what's done here with a call to mark_inode_dirty() in the
 * case that i_size has changed.
 */
439 440 441 442
int simple_write_end(struct file *file, struct address_space *mapping,
			loff_t pos, unsigned len, unsigned copied,
			struct page *page, void *fsdata)
{
443 444
	struct inode *inode = page->mapping->host;
	loff_t last_pos = pos + copied;
445 446 447

	/* zero the stale part of the page if we did a short copy */
	if (copied < len) {
448 449 450
		unsigned from = pos & (PAGE_CACHE_SIZE - 1);

		zero_user(page, from + copied, len - copied);
451 452
	}

453 454 455 456 457 458 459 460
	if (!PageUptodate(page))
		SetPageUptodate(page);
	/*
	 * No need to use i_size_read() here, the i_size
	 * cannot change under us because we hold the i_mutex.
	 */
	if (last_pos > inode->i_size)
		i_size_write(inode, last_pos);
461

462
	set_page_dirty(page);
463 464 465 466 467 468
	unlock_page(page);
	page_cache_release(page);

	return copied;
}

469 470 471 472 473
/*
 * the inodes created here are not hashed. If you use iunique to generate
 * unique inode values later for this filesystem, then you must take care
 * to pass it an appropriate max_reserved value to avoid collisions.
 */
474 475
int simple_fill_super(struct super_block *s, unsigned long magic,
		      struct tree_descr *files)
L
Linus Torvalds 已提交
476 477 478 479 480 481 482 483 484
{
	struct inode *inode;
	struct dentry *root;
	struct dentry *dentry;
	int i;

	s->s_blocksize = PAGE_CACHE_SIZE;
	s->s_blocksize_bits = PAGE_CACHE_SHIFT;
	s->s_magic = magic;
485
	s->s_op = &simple_super_operations;
L
Linus Torvalds 已提交
486 487 488 489 490
	s->s_time_gran = 1;

	inode = new_inode(s);
	if (!inode)
		return -ENOMEM;
491 492 493 494 495
	/*
	 * because the root inode is 1, the files array must not contain an
	 * entry at index 1
	 */
	inode->i_ino = 1;
L
Linus Torvalds 已提交
496 497 498 499
	inode->i_mode = S_IFDIR | 0755;
	inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
	inode->i_op = &simple_dir_inode_operations;
	inode->i_fop = &simple_dir_operations;
M
Miklos Szeredi 已提交
500
	set_nlink(inode, 2);
501 502
	root = d_make_root(inode);
	if (!root)
L
Linus Torvalds 已提交
503 504 505 506
		return -ENOMEM;
	for (i = 0; !files->name || files->name[0]; i++, files++) {
		if (!files->name)
			continue;
507 508 509 510 511 512 513

		/* warn if it tries to conflict with the root inode */
		if (unlikely(i == 1))
			printk(KERN_WARNING "%s: %s passed in a files array"
				"with an index of 1!\n", __func__,
				s->s_type->name);

L
Linus Torvalds 已提交
514 515 516 517
		dentry = d_alloc_name(root, files->name);
		if (!dentry)
			goto out;
		inode = new_inode(s);
518 519
		if (!inode) {
			dput(dentry);
L
Linus Torvalds 已提交
520
			goto out;
521
		}
L
Linus Torvalds 已提交
522 523 524 525 526 527 528 529 530 531
		inode->i_mode = S_IFREG | files->mode;
		inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
		inode->i_fop = files->ops;
		inode->i_ino = i;
		d_add(dentry, inode);
	}
	s->s_root = root;
	return 0;
out:
	d_genocide(root);
532
	shrink_dcache_parent(root);
L
Linus Torvalds 已提交
533 534 535 536 537 538
	dput(root);
	return -ENOMEM;
}

static DEFINE_SPINLOCK(pin_fs_lock);

539
int simple_pin_fs(struct file_system_type *type, struct vfsmount **mount, int *count)
L
Linus Torvalds 已提交
540 541 542 543 544
{
	struct vfsmount *mnt = NULL;
	spin_lock(&pin_fs_lock);
	if (unlikely(!*mount)) {
		spin_unlock(&pin_fs_lock);
545
		mnt = vfs_kern_mount(type, MS_KERNMOUNT, type->name, NULL);
L
Linus Torvalds 已提交
546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569
		if (IS_ERR(mnt))
			return PTR_ERR(mnt);
		spin_lock(&pin_fs_lock);
		if (!*mount)
			*mount = mnt;
	}
	mntget(*mount);
	++*count;
	spin_unlock(&pin_fs_lock);
	mntput(mnt);
	return 0;
}

void simple_release_fs(struct vfsmount **mount, int *count)
{
	struct vfsmount *mnt;
	spin_lock(&pin_fs_lock);
	mnt = *mount;
	if (!--*count)
		*mount = NULL;
	spin_unlock(&pin_fs_lock);
	mntput(mnt);
}

570 571 572 573 574 575 576 577 578 579 580 581 582 583
/**
 * simple_read_from_buffer - copy data from the buffer to user space
 * @to: the user space buffer to read to
 * @count: the maximum number of bytes to read
 * @ppos: the current position in the buffer
 * @from: the buffer to read from
 * @available: the size of the buffer
 *
 * The simple_read_from_buffer() function reads up to @count bytes from the
 * buffer @from at offset @ppos into the user space address starting at @to.
 *
 * On success, the number of bytes read is returned and the offset @ppos is
 * advanced by this number, or negative value is returned on error.
 **/
L
Linus Torvalds 已提交
584 585 586 587
ssize_t simple_read_from_buffer(void __user *to, size_t count, loff_t *ppos,
				const void *from, size_t available)
{
	loff_t pos = *ppos;
588 589
	size_t ret;

L
Linus Torvalds 已提交
590 591
	if (pos < 0)
		return -EINVAL;
592
	if (pos >= available || !count)
L
Linus Torvalds 已提交
593 594 595
		return 0;
	if (count > available - pos)
		count = available - pos;
596 597
	ret = copy_to_user(to, from + pos, count);
	if (ret == count)
L
Linus Torvalds 已提交
598
		return -EFAULT;
599
	count -= ret;
L
Linus Torvalds 已提交
600 601 602 603
	*ppos = pos + count;
	return count;
}

604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637
/**
 * simple_write_to_buffer - copy data from user space to the buffer
 * @to: the buffer to write to
 * @available: the size of the buffer
 * @ppos: the current position in the buffer
 * @from: the user space buffer to read from
 * @count: the maximum number of bytes to read
 *
 * The simple_write_to_buffer() function reads up to @count bytes from the user
 * space address starting at @from into the buffer @to at offset @ppos.
 *
 * On success, the number of bytes written is returned and the offset @ppos is
 * advanced by this number, or negative value is returned on error.
 **/
ssize_t simple_write_to_buffer(void *to, size_t available, loff_t *ppos,
		const void __user *from, size_t count)
{
	loff_t pos = *ppos;
	size_t res;

	if (pos < 0)
		return -EINVAL;
	if (pos >= available || !count)
		return 0;
	if (count > available - pos)
		count = available - pos;
	res = copy_from_user(to + pos, from, count);
	if (res == count)
		return -EFAULT;
	count -= res;
	*ppos = pos + count;
	return count;
}

638 639 640 641 642 643 644 645 646 647 648 649 650 651
/**
 * memory_read_from_buffer - copy data from the buffer
 * @to: the kernel space buffer to read to
 * @count: the maximum number of bytes to read
 * @ppos: the current position in the buffer
 * @from: the buffer to read from
 * @available: the size of the buffer
 *
 * The memory_read_from_buffer() function reads up to @count bytes from the
 * buffer @from at offset @ppos into the kernel space address starting at @to.
 *
 * On success, the number of bytes read is returned and the offset @ppos is
 * advanced by this number, or negative value is returned on error.
 **/
652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668
ssize_t memory_read_from_buffer(void *to, size_t count, loff_t *ppos,
				const void *from, size_t available)
{
	loff_t pos = *ppos;

	if (pos < 0)
		return -EINVAL;
	if (pos >= available)
		return 0;
	if (count > available - pos)
		count = available - pos;
	memcpy(to, from + pos, count);
	*ppos = pos + count;

	return count;
}

L
Linus Torvalds 已提交
669 670 671 672 673 674
/*
 * Transaction based IO.
 * The file expects a single write which triggers the transaction, and then
 * possibly a read which collects the result - which is stored in a
 * file-local buffer.
 */
675 676 677 678 679 680 681 682 683 684 685 686 687 688 689

void simple_transaction_set(struct file *file, size_t n)
{
	struct simple_transaction_argresp *ar = file->private_data;

	BUG_ON(n > SIMPLE_TRANSACTION_LIMIT);

	/*
	 * The barrier ensures that ar->size will really remain zero until
	 * ar->data is ready for reading.
	 */
	smp_mb();
	ar->size = n;
}

L
Linus Torvalds 已提交
690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735
char *simple_transaction_get(struct file *file, const char __user *buf, size_t size)
{
	struct simple_transaction_argresp *ar;
	static DEFINE_SPINLOCK(simple_transaction_lock);

	if (size > SIMPLE_TRANSACTION_LIMIT - 1)
		return ERR_PTR(-EFBIG);

	ar = (struct simple_transaction_argresp *)get_zeroed_page(GFP_KERNEL);
	if (!ar)
		return ERR_PTR(-ENOMEM);

	spin_lock(&simple_transaction_lock);

	/* only one write allowed per open */
	if (file->private_data) {
		spin_unlock(&simple_transaction_lock);
		free_page((unsigned long)ar);
		return ERR_PTR(-EBUSY);
	}

	file->private_data = ar;

	spin_unlock(&simple_transaction_lock);

	if (copy_from_user(ar->data, buf, size))
		return ERR_PTR(-EFAULT);

	return ar->data;
}

ssize_t simple_transaction_read(struct file *file, char __user *buf, size_t size, loff_t *pos)
{
	struct simple_transaction_argresp *ar = file->private_data;

	if (!ar)
		return 0;
	return simple_read_from_buffer(buf, size, pos, ar->data, ar->size);
}

int simple_transaction_release(struct inode *inode, struct file *file)
{
	free_page((unsigned long)file->private_data);
	return 0;
}

736 737 738
/* Simple attribute files */

struct simple_attr {
739 740
	int (*get)(void *, u64 *);
	int (*set)(void *, u64);
741 742 743 744
	char get_buf[24];	/* enough to store a u64 and "\n\0" */
	char set_buf[24];
	void *data;
	const char *fmt;	/* format for read operation */
I
Ingo Molnar 已提交
745
	struct mutex mutex;	/* protects access to these buffers */
746 747 748 749 750
};

/* simple_attr_open is called by an actual attribute open file operation
 * to set the attribute specific access operations. */
int simple_attr_open(struct inode *inode, struct file *file,
751
		     int (*get)(void *, u64 *), int (*set)(void *, u64),
752 753 754 755 756 757 758 759 760 761
		     const char *fmt)
{
	struct simple_attr *attr;

	attr = kmalloc(sizeof(*attr), GFP_KERNEL);
	if (!attr)
		return -ENOMEM;

	attr->get = get;
	attr->set = set;
762
	attr->data = inode->i_private;
763
	attr->fmt = fmt;
I
Ingo Molnar 已提交
764
	mutex_init(&attr->mutex);
765 766 767 768 769 770

	file->private_data = attr;

	return nonseekable_open(inode, file);
}

771
int simple_attr_release(struct inode *inode, struct file *file)
772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789
{
	kfree(file->private_data);
	return 0;
}

/* read from the buffer that is filled with the get function */
ssize_t simple_attr_read(struct file *file, char __user *buf,
			 size_t len, loff_t *ppos)
{
	struct simple_attr *attr;
	size_t size;
	ssize_t ret;

	attr = file->private_data;

	if (!attr->get)
		return -EACCES;

790 791 792 793
	ret = mutex_lock_interruptible(&attr->mutex);
	if (ret)
		return ret;

794
	if (*ppos) {		/* continued read */
795
		size = strlen(attr->get_buf);
796 797 798 799 800 801
	} else {		/* first read */
		u64 val;
		ret = attr->get(attr->data, &val);
		if (ret)
			goto out;

802
		size = scnprintf(attr->get_buf, sizeof(attr->get_buf),
803 804
				 attr->fmt, (unsigned long long)val);
	}
805 806

	ret = simple_read_from_buffer(buf, len, ppos, attr->get_buf, size);
807
out:
I
Ingo Molnar 已提交
808
	mutex_unlock(&attr->mutex);
809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824
	return ret;
}

/* interpret the buffer as a number to call the set function with */
ssize_t simple_attr_write(struct file *file, const char __user *buf,
			  size_t len, loff_t *ppos)
{
	struct simple_attr *attr;
	u64 val;
	size_t size;
	ssize_t ret;

	attr = file->private_data;
	if (!attr->set)
		return -EACCES;

825 826 827 828
	ret = mutex_lock_interruptible(&attr->mutex);
	if (ret)
		return ret;

829 830 831 832 833 834
	ret = -EFAULT;
	size = min(sizeof(attr->set_buf) - 1, len);
	if (copy_from_user(attr->set_buf, buf, size))
		goto out;

	attr->set_buf[size] = '\0';
835
	val = simple_strtoll(attr->set_buf, NULL, 0);
836 837 838
	ret = attr->set(attr->data, val);
	if (ret == 0)
		ret = len; /* on success, claim we got the whole input */
839
out:
I
Ingo Molnar 已提交
840
	mutex_unlock(&attr->mutex);
841 842 843
	return ret;
}

C
Christoph Hellwig 已提交
844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871
/**
 * generic_fh_to_dentry - generic helper for the fh_to_dentry export operation
 * @sb:		filesystem to do the file handle conversion on
 * @fid:	file handle to convert
 * @fh_len:	length of the file handle in bytes
 * @fh_type:	type of file handle
 * @get_inode:	filesystem callback to retrieve inode
 *
 * This function decodes @fid as long as it has one of the well-known
 * Linux filehandle types and calls @get_inode on it to retrieve the
 * inode for the object specified in the file handle.
 */
struct dentry *generic_fh_to_dentry(struct super_block *sb, struct fid *fid,
		int fh_len, int fh_type, struct inode *(*get_inode)
			(struct super_block *sb, u64 ino, u32 gen))
{
	struct inode *inode = NULL;

	if (fh_len < 2)
		return NULL;

	switch (fh_type) {
	case FILEID_INO32_GEN:
	case FILEID_INO32_GEN_PARENT:
		inode = get_inode(sb, fid->i32.ino, fid->i32.gen);
		break;
	}

872
	return d_obtain_alias(inode);
C
Christoph Hellwig 已提交
873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904
}
EXPORT_SYMBOL_GPL(generic_fh_to_dentry);

/**
 * generic_fh_to_dentry - generic helper for the fh_to_parent export operation
 * @sb:		filesystem to do the file handle conversion on
 * @fid:	file handle to convert
 * @fh_len:	length of the file handle in bytes
 * @fh_type:	type of file handle
 * @get_inode:	filesystem callback to retrieve inode
 *
 * This function decodes @fid as long as it has one of the well-known
 * Linux filehandle types and calls @get_inode on it to retrieve the
 * inode for the _parent_ object specified in the file handle if it
 * is specified in the file handle, or NULL otherwise.
 */
struct dentry *generic_fh_to_parent(struct super_block *sb, struct fid *fid,
		int fh_len, int fh_type, struct inode *(*get_inode)
			(struct super_block *sb, u64 ino, u32 gen))
{
	struct inode *inode = NULL;

	if (fh_len <= 2)
		return NULL;

	switch (fh_type) {
	case FILEID_INO32_GEN_PARENT:
		inode = get_inode(sb, fid->i32.parent_ino,
				  (fh_len > 3 ? fid->i32.parent_gen : 0));
		break;
	}

905
	return d_obtain_alias(inode);
C
Christoph Hellwig 已提交
906 907 908
}
EXPORT_SYMBOL_GPL(generic_fh_to_parent);

909 910 911 912 913 914 915 916 917
/**
 * generic_file_fsync - generic fsync implementation for simple filesystems
 * @file:	file to synchronize
 * @datasync:	only synchronize essential metadata if true
 *
 * This is a generic implementation of the fsync method for simple
 * filesystems which track all non-inode metadata in the buffers list
 * hanging off the address_space structure.
 */
918 919
int generic_file_fsync(struct file *file, loff_t start, loff_t end,
		       int datasync)
A
Al Viro 已提交
920
{
921
	struct inode *inode = file->f_mapping->host;
A
Al Viro 已提交
922 923 924
	int err;
	int ret;

925 926 927 928 929
	err = filemap_write_and_wait_range(inode->i_mapping, start, end);
	if (err)
		return err;

	mutex_lock(&inode->i_mutex);
A
Al Viro 已提交
930 931
	ret = sync_mapping_buffers(inode->i_mapping);
	if (!(inode->i_state & I_DIRTY))
932
		goto out;
A
Al Viro 已提交
933
	if (datasync && !(inode->i_state & I_DIRTY_DATASYNC))
934
		goto out;
A
Al Viro 已提交
935

C
Christoph Hellwig 已提交
936
	err = sync_inode_metadata(inode, 1);
A
Al Viro 已提交
937 938
	if (ret == 0)
		ret = err;
939 940
out:
	mutex_unlock(&inode->i_mutex);
A
Al Viro 已提交
941 942
	return ret;
}
943 944
EXPORT_SYMBOL(generic_file_fsync);

945 946 947 948 949 950 951 952 953 954 955 956
/**
 * generic_check_addressable - Check addressability of file system
 * @blocksize_bits:	log of file system block size
 * @num_blocks:		number of blocks in file system
 *
 * Determine whether a file system with @num_blocks blocks (and a
 * block size of 2**@blocksize_bits) is addressable by the sector_t
 * and page cache of the system.  Return 0 if so and -EFBIG otherwise.
 */
int generic_check_addressable(unsigned blocksize_bits, u64 num_blocks)
{
	u64 last_fs_block = num_blocks - 1;
957 958
	u64 last_fs_page =
		last_fs_block >> (PAGE_CACHE_SHIFT - blocksize_bits);
959 960 961 962 963 964 965

	if (unlikely(num_blocks == 0))
		return 0;

	if ((blocksize_bits < 9) || (blocksize_bits > PAGE_CACHE_SHIFT))
		return -EINVAL;

966 967
	if ((last_fs_block > (sector_t)(~0ULL) >> (blocksize_bits - 9)) ||
	    (last_fs_page > (pgoff_t)(~0ULL))) {
968 969 970 971 972 973
		return -EFBIG;
	}
	return 0;
}
EXPORT_SYMBOL(generic_check_addressable);

974 975 976
/*
 * No-op implementation of ->fsync for in-memory filesystems.
 */
977
int noop_fsync(struct file *file, loff_t start, loff_t end, int datasync)
978 979 980
{
	return 0;
}
A
Al Viro 已提交
981

L
Linus Torvalds 已提交
982 983 984 985 986
EXPORT_SYMBOL(dcache_dir_close);
EXPORT_SYMBOL(dcache_dir_lseek);
EXPORT_SYMBOL(dcache_dir_open);
EXPORT_SYMBOL(dcache_readdir);
EXPORT_SYMBOL(generic_read_dir);
A
Al Viro 已提交
987
EXPORT_SYMBOL(mount_pseudo);
988 989
EXPORT_SYMBOL(simple_write_begin);
EXPORT_SYMBOL(simple_write_end);
L
Linus Torvalds 已提交
990 991 992 993 994
EXPORT_SYMBOL(simple_dir_inode_operations);
EXPORT_SYMBOL(simple_dir_operations);
EXPORT_SYMBOL(simple_empty);
EXPORT_SYMBOL(simple_fill_super);
EXPORT_SYMBOL(simple_getattr);
S
Stephen Boyd 已提交
995
EXPORT_SYMBOL(simple_open);
L
Linus Torvalds 已提交
996 997 998 999 1000 1001 1002 1003
EXPORT_SYMBOL(simple_link);
EXPORT_SYMBOL(simple_lookup);
EXPORT_SYMBOL(simple_pin_fs);
EXPORT_SYMBOL(simple_readpage);
EXPORT_SYMBOL(simple_release_fs);
EXPORT_SYMBOL(simple_rename);
EXPORT_SYMBOL(simple_rmdir);
EXPORT_SYMBOL(simple_statfs);
1004
EXPORT_SYMBOL(noop_fsync);
L
Linus Torvalds 已提交
1005 1006
EXPORT_SYMBOL(simple_unlink);
EXPORT_SYMBOL(simple_read_from_buffer);
1007
EXPORT_SYMBOL(simple_write_to_buffer);
1008
EXPORT_SYMBOL(memory_read_from_buffer);
1009
EXPORT_SYMBOL(simple_transaction_set);
L
Linus Torvalds 已提交
1010 1011 1012
EXPORT_SYMBOL(simple_transaction_get);
EXPORT_SYMBOL(simple_transaction_read);
EXPORT_SYMBOL(simple_transaction_release);
1013
EXPORT_SYMBOL_GPL(simple_attr_open);
1014
EXPORT_SYMBOL_GPL(simple_attr_release);
1015 1016
EXPORT_SYMBOL_GPL(simple_attr_read);
EXPORT_SYMBOL_GPL(simple_attr_write);