file.c 27 KB
Newer Older
Linus Torvalds's avatar
Linus Torvalds committed
1
/*
Tejun Heo's avatar
Tejun Heo committed
2
3
4
5
6
7
8
9
10
 * fs/sysfs/file.c - sysfs regular (text) file implementation
 *
 * Copyright (c) 2001-3 Patrick Mochel
 * Copyright (c) 2007 SUSE Linux Products GmbH
 * Copyright (c) 2007 Tejun Heo <teheo@suse.de>
 *
 * This file is released under the GPLv2.
 *
 * Please see Documentation/filesystems/sysfs.txt for more information.
Linus Torvalds's avatar
Linus Torvalds committed
11
12
13
14
 */

#include <linux/module.h>
#include <linux/kobject.h>
15
#include <linux/kallsyms.h>
16
#include <linux/slab.h>
17
#include <linux/fsnotify.h>
18
#include <linux/namei.h>
19
#include <linux/poll.h>
20
#include <linux/list.h>
21
#include <linux/mutex.h>
Andrew Morton's avatar
Andrew Morton committed
22
#include <linux/limits.h>
23
#include <linux/uaccess.h>
24
#include <linux/seq_file.h>
25
#include <linux/mm.h>
Linus Torvalds's avatar
Linus Torvalds committed
26
27
28

#include "sysfs.h"

Tejun Heo's avatar
Tejun Heo committed
29
/*
30
 * There's one sysfs_open_file for each open file and one sysfs_open_dirent
Tejun Heo's avatar
Tejun Heo committed
31
 * for each sysfs_dirent with one or more open files.
Tejun Heo's avatar
Tejun Heo committed
32
 *
Tejun Heo's avatar
Tejun Heo committed
33
34
35
 * sysfs_dirent->s_attr.open points to sysfs_open_dirent.  s_attr.open is
 * protected by sysfs_open_dirent_lock.
 *
36
37
 * filp->private_data points to seq_file whose ->private points to
 * sysfs_open_file.  sysfs_open_files are chained at
38
 * sysfs_open_dirent->files, which is protected by sysfs_open_file_mutex.
Tejun Heo's avatar
Tejun Heo committed
39
 */
Jiri Slaby's avatar
Jiri Slaby committed
40
static DEFINE_SPINLOCK(sysfs_open_dirent_lock);
Tejun Heo's avatar
Tejun Heo committed
41
static DEFINE_MUTEX(sysfs_open_file_mutex);
Tejun Heo's avatar
Tejun Heo committed
42
43
44

struct sysfs_open_dirent {
	atomic_t		refcnt;
45
46
	atomic_t		event;
	wait_queue_head_t	poll;
47
	struct list_head	files; /* goes through sysfs_open_file.list */
Tejun Heo's avatar
Tejun Heo committed
48
49
};

50
struct sysfs_open_file {
51
52
	struct sysfs_dirent	*sd;
	struct file		*file;
53
	struct mutex		mutex;
54
	int			event;
Tejun Heo's avatar
Tejun Heo committed
55
	struct list_head	list;
56
57
58

	bool			mmapped;
	const struct vm_operations_struct *vm_ops;
59
};
Linus Torvalds's avatar
Linus Torvalds committed
60

61
62
63
64
65
static bool sysfs_is_bin(struct sysfs_dirent *sd)
{
	return sysfs_type(sd) == SYSFS_KOBJ_BIN_ATTR;
}

66
67
68
69
70
static struct sysfs_open_file *sysfs_of(struct file *file)
{
	return ((struct seq_file *)file->private_data)->private;
}

Tejun Heo's avatar
Tejun Heo committed
71
72
73
74
75
76
77
78
/*
 * Determine ktype->sysfs_ops for the given sysfs_dirent.  This function
 * must be called while holding an active reference.
 */
static const struct sysfs_ops *sysfs_file_ops(struct sysfs_dirent *sd)
{
	struct kobject *kobj = sd->s_parent->s_dir.kobj;

79
80
	if (!sysfs_ignore_lockdep(sd))
		lockdep_assert_held(sd);
Tejun Heo's avatar
Tejun Heo committed
81
82
83
	return kobj->ktype ? kobj->ktype->sysfs_ops : NULL;
}

84
85
86
87
/*
 * Reads on sysfs are handled through seq_file, which takes care of hairy
 * details like buffering and seeking.  The following function pipes
 * sysfs_ops->show() result through seq_file.
Linus Torvalds's avatar
Linus Torvalds committed
88
 */
89
static int sysfs_seq_show(struct seq_file *sf, void *v)
Linus Torvalds's avatar
Linus Torvalds committed
90
{
91
92
	struct sysfs_open_file *of = sf->private;
	struct kobject *kobj = of->sd->s_parent->s_dir.kobj;
Tejun Heo's avatar
Tejun Heo committed
93
	const struct sysfs_ops *ops;
94
	char *buf;
Linus Torvalds's avatar
Linus Torvalds committed
95
96
	ssize_t count;

97
98
99
100
101
102
	/* acquire buffer and ensure that it's >= PAGE_SIZE */
	count = seq_get_buf(sf, &buf);
	if (count < PAGE_SIZE) {
		seq_commit(sf, -1);
		return 0;
	}
Linus Torvalds's avatar
Linus Torvalds committed
103

104
105
106
107
108
109
110
111
	/*
	 * Need @of->sd for attr and ops, its parent for kobj.  @of->mutex
	 * nests outside active ref and is just to ensure that the ops
	 * aren't called concurrently for the same open file.
	 */
	mutex_lock(&of->mutex);
	if (!sysfs_get_active(of->sd)) {
		mutex_unlock(&of->mutex);
112
		return -ENODEV;
113
	}
114

115
	of->event = atomic_read(&of->sd->s_attr.open->event);
Tejun Heo's avatar
Tejun Heo committed
116

117
118
119
120
121
122
123
124
125
	/*
	 * Lookup @ops and invoke show().  Control may reach here via seq
	 * file lseek even if @ops->show() isn't implemented.
	 */
	ops = sysfs_file_ops(of->sd);
	if (ops->show)
		count = ops->show(kobj, of->sd->s_attr.attr, buf);
	else
		count = 0;
126

127
128
129
130
131
	sysfs_put_active(of->sd);
	mutex_unlock(&of->mutex);

	if (count < 0)
		return count;
132

133
134
135
136
	/*
	 * The code works fine with PAGE_SIZE return but it's likely to
	 * indicate truncated result or overflow in normal use cases.
	 */
137
138
139
140
141
142
	if (count >= (ssize_t)PAGE_SIZE) {
		print_symbol("fill_read_buffer: %s returned bad count\n",
			(unsigned long)ops->show);
		/* Try to struggle along */
		count = PAGE_SIZE - 1;
	}
143
144
	seq_commit(sf, count);
	return 0;
Linus Torvalds's avatar
Linus Torvalds committed
145
146
}

Tejun Heo's avatar
Tejun Heo committed
147
148
149
150
151
152
153
154
155
156
/*
 * Read method for bin files.  As reading a bin file can have side-effects,
 * the exact offset and bytes specified in read(2) call should be passed to
 * the read callback making it difficult to use seq_file.  Implement
 * simplistic custom buffering for bin files.
 */
static ssize_t sysfs_bin_read(struct file *file, char __user *userbuf,
			      size_t bytes, loff_t *off)
{
	struct sysfs_open_file *of = sysfs_of(file);
157
	struct bin_attribute *battr = of->sd->s_attr.bin_attr;
Tejun Heo's avatar
Tejun Heo committed
158
	struct kobject *kobj = of->sd->s_parent->s_dir.kobj;
159
	loff_t size = file_inode(file)->i_size;
Tejun Heo's avatar
Tejun Heo committed
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
	int count = min_t(size_t, bytes, PAGE_SIZE);
	loff_t offs = *off;
	char *buf;

	if (!bytes)
		return 0;

	if (size) {
		if (offs > size)
			return 0;
		if (offs + count > size)
			count = size - offs;
	}

	buf = kmalloc(count, GFP_KERNEL);
	if (!buf)
		return -ENOMEM;

	/* need of->sd for battr, its parent for kobj */
	mutex_lock(&of->mutex);
	if (!sysfs_get_active(of->sd)) {
		count = -ENODEV;
		mutex_unlock(&of->mutex);
		goto out_free;
	}

	if (battr->read)
		count = battr->read(file, kobj, battr, buf, offs, count);
	else
		count = -EIO;

	sysfs_put_active(of->sd);
	mutex_unlock(&of->mutex);

	if (count < 0)
		goto out_free;

	if (copy_to_user(userbuf, buf, count)) {
		count = -EFAULT;
		goto out_free;
	}

	pr_debug("offs = %lld, *off = %lld, count = %d\n", offs, *off, count);

	*off = offs + count;

 out_free:
	kfree(buf);
	return count;
}

Linus Torvalds's avatar
Linus Torvalds committed
211
/**
Tejun Heo's avatar
Tejun Heo committed
212
213
214
 * flush_write_buffer - push buffer to kobject
 * @of: open file
 * @buf: data buffer for file
215
 * @off: file offset to write to
Tejun Heo's avatar
Tejun Heo committed
216
 * @count: number of bytes
Linus Torvalds's avatar
Linus Torvalds committed
217
 *
Tejun Heo's avatar
Tejun Heo committed
218
219
 * Get the correct pointers for the kobject and the attribute we're dealing
 * with, then call the store() method for it with @buf.
Linus Torvalds's avatar
Linus Torvalds committed
220
 */
221
static int flush_write_buffer(struct sysfs_open_file *of, char *buf, loff_t off,
Tejun Heo's avatar
Tejun Heo committed
222
			      size_t count)
Linus Torvalds's avatar
Linus Torvalds committed
223
{
224
	struct kobject *kobj = of->sd->s_parent->s_dir.kobj;
Tejun Heo's avatar
Tejun Heo committed
225
	int rc = 0;
226

Tejun Heo's avatar
Tejun Heo committed
227
228
229
230
231
232
233
234
	/*
	 * Need @of->sd for attr and ops, its parent for kobj.  @of->mutex
	 * nests outside active ref and is just to ensure that the ops
	 * aren't called concurrently for the same open file.
	 */
	mutex_lock(&of->mutex);
	if (!sysfs_get_active(of->sd)) {
		mutex_unlock(&of->mutex);
235
		return -ENODEV;
Tejun Heo's avatar
Tejun Heo committed
236
	}
237

238
	if (sysfs_is_bin(of->sd)) {
239
		struct bin_attribute *battr = of->sd->s_attr.bin_attr;
240
241
242
243
244
245
246
247
248
249

		rc = -EIO;
		if (battr->write)
			rc = battr->write(of->file, kobj, battr, buf, off,
					  count);
	} else {
		const struct sysfs_ops *ops = sysfs_file_ops(of->sd);

		rc = ops->store(kobj, of->sd->s_attr.attr, buf, count);
	}
250

251
	sysfs_put_active(of->sd);
Tejun Heo's avatar
Tejun Heo committed
252
	mutex_unlock(&of->mutex);
Linus Torvalds's avatar
Linus Torvalds committed
253

254
	return rc;
Linus Torvalds's avatar
Linus Torvalds committed
255
256
257
}

/**
Tejun Heo's avatar
Tejun Heo committed
258
259
260
261
262
263
264
265
 * sysfs_write_file - write an attribute
 * @file: file pointer
 * @user_buf: data to write
 * @count: number of bytes
 * @ppos: starting offset
 *
 * Copy data in from userland and pass it to the matching
 * sysfs_ops->store() by invoking flush_write_buffer().
Linus Torvalds's avatar
Linus Torvalds committed
266
 *
Tejun Heo's avatar
Tejun Heo committed
267
268
269
270
271
 * There is no easy way for us to know if userspace is only doing a partial
 * write, so we don't support them. We expect the entire buffer to come on
 * the first write.  Hint: if you're writing a value, first read the file,
 * modify only the the value you're changing, then write entire buffer
 * back.
Linus Torvalds's avatar
Linus Torvalds committed
272
 */
Tejun Heo's avatar
Tejun Heo committed
273
static ssize_t sysfs_write_file(struct file *file, const char __user *user_buf,
274
				size_t count, loff_t *ppos)
Linus Torvalds's avatar
Linus Torvalds committed
275
{
276
	struct sysfs_open_file *of = sysfs_of(file);
277
	ssize_t len = min_t(size_t, count, PAGE_SIZE);
278
	loff_t size = file_inode(file)->i_size;
Tejun Heo's avatar
Tejun Heo committed
279
	char *buf;
Linus Torvalds's avatar
Linus Torvalds committed
280

281
	if (sysfs_is_bin(of->sd) && size) {
282
283
284
285
286
		if (size <= *ppos)
			return 0;
		len = min_t(ssize_t, len, size - *ppos);
	}

Tejun Heo's avatar
Tejun Heo committed
287
288
289
290
291
292
293
294
295
296
297
298
299
	if (!len)
		return 0;

	buf = kmalloc(len + 1, GFP_KERNEL);
	if (!buf)
		return -ENOMEM;

	if (copy_from_user(buf, user_buf, len)) {
		len = -EFAULT;
		goto out_free;
	}
	buf[len] = '\0';	/* guarantee string termination */

300
	len = flush_write_buffer(of, buf, *ppos, len);
Linus Torvalds's avatar
Linus Torvalds committed
301
302
	if (len > 0)
		*ppos += len;
Tejun Heo's avatar
Tejun Heo committed
303
304
out_free:
	kfree(buf);
Linus Torvalds's avatar
Linus Torvalds committed
305
306
307
	return len;
}

308
309
310
311
312
313
314
315
316
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
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
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
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
static void sysfs_bin_vma_open(struct vm_area_struct *vma)
{
	struct file *file = vma->vm_file;
	struct sysfs_open_file *of = sysfs_of(file);

	if (!of->vm_ops)
		return;

	if (!sysfs_get_active(of->sd))
		return;

	if (of->vm_ops->open)
		of->vm_ops->open(vma);

	sysfs_put_active(of->sd);
}

static int sysfs_bin_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
{
	struct file *file = vma->vm_file;
	struct sysfs_open_file *of = sysfs_of(file);
	int ret;

	if (!of->vm_ops)
		return VM_FAULT_SIGBUS;

	if (!sysfs_get_active(of->sd))
		return VM_FAULT_SIGBUS;

	ret = VM_FAULT_SIGBUS;
	if (of->vm_ops->fault)
		ret = of->vm_ops->fault(vma, vmf);

	sysfs_put_active(of->sd);
	return ret;
}

static int sysfs_bin_page_mkwrite(struct vm_area_struct *vma,
				  struct vm_fault *vmf)
{
	struct file *file = vma->vm_file;
	struct sysfs_open_file *of = sysfs_of(file);
	int ret;

	if (!of->vm_ops)
		return VM_FAULT_SIGBUS;

	if (!sysfs_get_active(of->sd))
		return VM_FAULT_SIGBUS;

	ret = 0;
	if (of->vm_ops->page_mkwrite)
		ret = of->vm_ops->page_mkwrite(vma, vmf);
	else
		file_update_time(file);

	sysfs_put_active(of->sd);
	return ret;
}

static int sysfs_bin_access(struct vm_area_struct *vma, unsigned long addr,
			    void *buf, int len, int write)
{
	struct file *file = vma->vm_file;
	struct sysfs_open_file *of = sysfs_of(file);
	int ret;

	if (!of->vm_ops)
		return -EINVAL;

	if (!sysfs_get_active(of->sd))
		return -EINVAL;

	ret = -EINVAL;
	if (of->vm_ops->access)
		ret = of->vm_ops->access(vma, addr, buf, len, write);

	sysfs_put_active(of->sd);
	return ret;
}

#ifdef CONFIG_NUMA
static int sysfs_bin_set_policy(struct vm_area_struct *vma,
				struct mempolicy *new)
{
	struct file *file = vma->vm_file;
	struct sysfs_open_file *of = sysfs_of(file);
	int ret;

	if (!of->vm_ops)
		return 0;

	if (!sysfs_get_active(of->sd))
		return -EINVAL;

	ret = 0;
	if (of->vm_ops->set_policy)
		ret = of->vm_ops->set_policy(vma, new);

	sysfs_put_active(of->sd);
	return ret;
}

static struct mempolicy *sysfs_bin_get_policy(struct vm_area_struct *vma,
					      unsigned long addr)
{
	struct file *file = vma->vm_file;
	struct sysfs_open_file *of = sysfs_of(file);
	struct mempolicy *pol;

	if (!of->vm_ops)
		return vma->vm_policy;

	if (!sysfs_get_active(of->sd))
		return vma->vm_policy;

	pol = vma->vm_policy;
	if (of->vm_ops->get_policy)
		pol = of->vm_ops->get_policy(vma, addr);

	sysfs_put_active(of->sd);
	return pol;
}

static int sysfs_bin_migrate(struct vm_area_struct *vma, const nodemask_t *from,
			     const nodemask_t *to, unsigned long flags)
{
	struct file *file = vma->vm_file;
	struct sysfs_open_file *of = sysfs_of(file);
	int ret;

	if (!of->vm_ops)
		return 0;

	if (!sysfs_get_active(of->sd))
		return 0;

	ret = 0;
	if (of->vm_ops->migrate)
		ret = of->vm_ops->migrate(vma, from, to, flags);

	sysfs_put_active(of->sd);
	return ret;
}
#endif

static const struct vm_operations_struct sysfs_bin_vm_ops = {
	.open		= sysfs_bin_vma_open,
	.fault		= sysfs_bin_fault,
	.page_mkwrite	= sysfs_bin_page_mkwrite,
	.access		= sysfs_bin_access,
#ifdef CONFIG_NUMA
	.set_policy	= sysfs_bin_set_policy,
	.get_policy	= sysfs_bin_get_policy,
	.migrate	= sysfs_bin_migrate,
#endif
};

static int sysfs_bin_mmap(struct file *file, struct vm_area_struct *vma)
{
	struct sysfs_open_file *of = sysfs_of(file);
469
	struct bin_attribute *battr = of->sd->s_attr.bin_attr;
470
471
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
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
	struct kobject *kobj = of->sd->s_parent->s_dir.kobj;
	int rc;

	mutex_lock(&of->mutex);

	/* need of->sd for battr, its parent for kobj */
	rc = -ENODEV;
	if (!sysfs_get_active(of->sd))
		goto out_unlock;

	if (!battr->mmap)
		goto out_put;

	rc = battr->mmap(file, kobj, battr, vma);
	if (rc)
		goto out_put;

	/*
	 * PowerPC's pci_mmap of legacy_mem uses shmem_zero_setup()
	 * to satisfy versions of X which crash if the mmap fails: that
	 * substitutes a new vm_file, and we don't then want bin_vm_ops.
	 */
	if (vma->vm_file != file)
		goto out_put;

	rc = -EINVAL;
	if (of->mmapped && of->vm_ops != vma->vm_ops)
		goto out_put;

	/*
	 * It is not possible to successfully wrap close.
	 * So error if someone is trying to use close.
	 */
	rc = -EINVAL;
	if (vma->vm_ops && vma->vm_ops->close)
		goto out_put;

	rc = 0;
	of->mmapped = 1;
	of->vm_ops = vma->vm_ops;
	vma->vm_ops = &sysfs_bin_vm_ops;
out_put:
	sysfs_put_active(of->sd);
out_unlock:
	mutex_unlock(&of->mutex);

	return rc;
}

Tejun Heo's avatar
Tejun Heo committed
519
520
521
/**
 *	sysfs_get_open_dirent - get or create sysfs_open_dirent
 *	@sd: target sysfs_dirent
522
 *	@of: sysfs_open_file for this instance of open
Tejun Heo's avatar
Tejun Heo committed
523
524
 *
 *	If @sd->s_attr.open exists, increment its reference count;
525
 *	otherwise, create one.  @of is chained to the files list.
Tejun Heo's avatar
Tejun Heo committed
526
527
528
529
530
531
532
533
 *
 *	LOCKING:
 *	Kernel thread context (may sleep).
 *
 *	RETURNS:
 *	0 on success, -errno on failure.
 */
static int sysfs_get_open_dirent(struct sysfs_dirent *sd,
534
				 struct sysfs_open_file *of)
Tejun Heo's avatar
Tejun Heo committed
535
536
537
538
{
	struct sysfs_open_dirent *od, *new_od = NULL;

 retry:
Tejun Heo's avatar
Tejun Heo committed
539
	mutex_lock(&sysfs_open_file_mutex);
540
	spin_lock_irq(&sysfs_open_dirent_lock);
Tejun Heo's avatar
Tejun Heo committed
541
542
543
544
545
546
547
548
549

	if (!sd->s_attr.open && new_od) {
		sd->s_attr.open = new_od;
		new_od = NULL;
	}

	od = sd->s_attr.open;
	if (od) {
		atomic_inc(&od->refcnt);
550
		list_add_tail(&of->list, &od->files);
Tejun Heo's avatar
Tejun Heo committed
551
552
	}

553
	spin_unlock_irq(&sysfs_open_dirent_lock);
Tejun Heo's avatar
Tejun Heo committed
554
	mutex_unlock(&sysfs_open_file_mutex);
Tejun Heo's avatar
Tejun Heo committed
555
556
557
558
559
560
561
562
563
564
565
566

	if (od) {
		kfree(new_od);
		return 0;
	}

	/* not there, initialize a new one and retry */
	new_od = kmalloc(sizeof(*new_od), GFP_KERNEL);
	if (!new_od)
		return -ENOMEM;

	atomic_set(&new_od->refcnt, 0);
567
568
	atomic_set(&new_od->event, 1);
	init_waitqueue_head(&new_od->poll);
569
	INIT_LIST_HEAD(&new_od->files);
Tejun Heo's avatar
Tejun Heo committed
570
571
572
573
574
575
	goto retry;
}

/**
 *	sysfs_put_open_dirent - put sysfs_open_dirent
 *	@sd: target sysfs_dirent
576
 *	@of: associated sysfs_open_file
Tejun Heo's avatar
Tejun Heo committed
577
 *
578
579
 *	Put @sd->s_attr.open and unlink @of from the files list.  If
 *	reference count reaches zero, disassociate and free it.
Tejun Heo's avatar
Tejun Heo committed
580
581
582
583
584
 *
 *	LOCKING:
 *	None.
 */
static void sysfs_put_open_dirent(struct sysfs_dirent *sd,
585
				  struct sysfs_open_file *of)
Tejun Heo's avatar
Tejun Heo committed
586
587
{
	struct sysfs_open_dirent *od = sd->s_attr.open;
588
	unsigned long flags;
Tejun Heo's avatar
Tejun Heo committed
589

Tejun Heo's avatar
Tejun Heo committed
590
	mutex_lock(&sysfs_open_file_mutex);
591
	spin_lock_irqsave(&sysfs_open_dirent_lock, flags);
Tejun Heo's avatar
Tejun Heo committed
592

593
594
595
	if (of)
		list_del(&of->list);

Tejun Heo's avatar
Tejun Heo committed
596
597
598
599
600
	if (atomic_dec_and_test(&od->refcnt))
		sd->s_attr.open = NULL;
	else
		od = NULL;

601
	spin_unlock_irqrestore(&sysfs_open_dirent_lock, flags);
Tejun Heo's avatar
Tejun Heo committed
602
	mutex_unlock(&sysfs_open_file_mutex);
Tejun Heo's avatar
Tejun Heo committed
603
604
605
606

	kfree(od);
}

607
static int sysfs_open_file(struct inode *inode, struct file *file)
Linus Torvalds's avatar
Linus Torvalds committed
608
{
609
	struct sysfs_dirent *attr_sd = file->f_path.dentry->d_fsdata;
610
	struct kobject *kobj = attr_sd->s_parent->s_dir.kobj;
611
	struct sysfs_open_file *of;
612
	bool has_read, has_write, has_mmap;
613
	int error = -EACCES;
Linus Torvalds's avatar
Linus Torvalds committed
614

615
	/* need attr_sd for attr and ops, its parent for kobj */
616
	if (!sysfs_get_active(attr_sd))
617
		return -ENODEV;
Linus Torvalds's avatar
Linus Torvalds committed
618

619
	if (sysfs_is_bin(attr_sd)) {
620
		struct bin_attribute *battr = attr_sd->s_attr.bin_attr;
Linus Torvalds's avatar
Linus Torvalds committed
621

622
623
		has_read = battr->read || battr->mmap;
		has_write = battr->write || battr->mmap;
624
		has_mmap = battr->mmap;
625
626
	} else {
		const struct sysfs_ops *ops = sysfs_file_ops(attr_sd);
Linus Torvalds's avatar
Linus Torvalds committed
627

628
629
630
631
		/* every kobject with an attribute needs a ktype assigned */
		if (WARN(!ops, KERN_ERR
			 "missing sysfs attribute operations for kobject: %s\n",
			 kobject_name(kobj)))
632
			goto err_out;
633
634
635

		has_read = ops->show;
		has_write = ops->store;
636
		has_mmap = false;
Linus Torvalds's avatar
Linus Torvalds committed
637
638
	}

639
640
641
642
643
644
645
646
647
	/* check perms and supported operations */
	if ((file->f_mode & FMODE_WRITE) &&
	    (!(inode->i_mode & S_IWUGO) || !has_write))
		goto err_out;

	if ((file->f_mode & FMODE_READ) &&
	    (!(inode->i_mode & S_IRUGO) || !has_read))
		goto err_out;

648
	/* allocate a sysfs_open_file for the file */
649
	error = -ENOMEM;
650
651
	of = kzalloc(sizeof(struct sysfs_open_file), GFP_KERNEL);
	if (!of)
652
		goto err_out;
Linus Torvalds's avatar
Linus Torvalds committed
653

654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
	/*
	 * The following is done to give a different lockdep key to
	 * @of->mutex for files which implement mmap.  This is a rather
	 * crude way to avoid false positive lockdep warning around
	 * mm->mmap_sem - mmap nests @of->mutex under mm->mmap_sem and
	 * reading /sys/block/sda/trace/act_mask grabs sr_mutex, under
	 * which mm->mmap_sem nests, while holding @of->mutex.  As each
	 * open file has a separate mutex, it's okay as long as those don't
	 * happen on the same file.  At this point, we can't easily give
	 * each file a separate locking class.  Let's differentiate on
	 * whether the file has mmap or not for now.
	 */
	if (has_mmap)
		mutex_init(&of->mutex);
	else
		mutex_init(&of->mutex);

671
672
	of->sd = attr_sd;
	of->file = file;
673
674

	/*
675
676
677
	 * Always instantiate seq_file even if read access doesn't use
	 * seq_file or is not requested.  This unifies private data access
	 * and readable regular files are the vast majority anyway.
678
	 */
679
680
681
682
	if (sysfs_is_bin(attr_sd))
		error = single_open(file, NULL, of);
	else
		error = single_open(file, sysfs_seq_show, of);
683
684
685
686
687
688
	if (error)
		goto err_free;

	/* seq_file clears PWRITE unconditionally, restore it if WRITE */
	if (file->f_mode & FMODE_WRITE)
		file->f_mode |= FMODE_PWRITE;
689

Tejun Heo's avatar
Tejun Heo committed
690
	/* make sure we have open dirent struct */
691
	error = sysfs_get_open_dirent(attr_sd, of);
Tejun Heo's avatar
Tejun Heo committed
692
	if (error)
693
		goto err_close;
Tejun Heo's avatar
Tejun Heo committed
694

695
	/* open succeeded, put active references */
696
	sysfs_put_active(attr_sd);
697
698
	return 0;

699
700
701
err_close:
	single_release(inode, file);
err_free:
702
	kfree(of);
703
err_out:
704
	sysfs_put_active(attr_sd);
Linus Torvalds's avatar
Linus Torvalds committed
705
706
707
	return error;
}

Tejun Heo's avatar
Tejun Heo committed
708
static int sysfs_release(struct inode *inode, struct file *filp)
Linus Torvalds's avatar
Linus Torvalds committed
709
{
Tejun Heo's avatar
Tejun Heo committed
710
	struct sysfs_dirent *sd = filp->f_path.dentry->d_fsdata;
711
	struct sysfs_open_file *of = sysfs_of(filp);
Linus Torvalds's avatar
Linus Torvalds committed
712

713
	sysfs_put_open_dirent(sd, of);
714
	single_release(inode, filp);
715
	kfree(of);
716

Linus Torvalds's avatar
Linus Torvalds committed
717
718
719
	return 0;
}

720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
void sysfs_unmap_bin_file(struct sysfs_dirent *sd)
{
	struct sysfs_open_dirent *od;
	struct sysfs_open_file *of;

	if (!sysfs_is_bin(sd))
		return;

	spin_lock_irq(&sysfs_open_dirent_lock);
	od = sd->s_attr.open;
	if (od)
		atomic_inc(&od->refcnt);
	spin_unlock_irq(&sysfs_open_dirent_lock);
	if (!od)
		return;

	mutex_lock(&sysfs_open_file_mutex);
	list_for_each_entry(of, &od->files, list) {
		struct inode *inode = file_inode(of->file);
		unmap_mapping_range(inode->i_mapping, 0, 0, 1);
	}
	mutex_unlock(&sysfs_open_file_mutex);

	sysfs_put_open_dirent(sd, NULL);
}

746
747
748
749
750
751
752
/* Sysfs attribute files are pollable.  The idea is that you read
 * the content and then you use 'poll' or 'select' to wait for
 * the content to change.  When the content changes (assuming the
 * manager for the kobject supports notification), poll will
 * return POLLERR|POLLPRI, and select will return the fd whether
 * it is waiting for read, write, or exceptions.
 * Once poll/select indicates that the value has changed, you
753
 * need to close and re-open the file, or seek to 0 and read again.
754
755
 * Reminder: this only works for attributes which actively support
 * it, and it is not possible to test an attribute from userspace
756
 * to see if it supports poll (Neither 'poll' nor 'select' return
757
758
759
760
 * an appropriate error code).  When in doubt, set a suitable timeout value.
 */
static unsigned int sysfs_poll(struct file *filp, poll_table *wait)
{
761
	struct sysfs_open_file *of = sysfs_of(filp);
762
	struct sysfs_dirent *attr_sd = filp->f_path.dentry->d_fsdata;
763
	struct sysfs_open_dirent *od = attr_sd->s_attr.open;
764
765

	/* need parent for the kobj, grab both */
766
	if (!sysfs_get_active(attr_sd))
767
		goto trigger;
768

769
	poll_wait(filp, &od->poll, wait);
770

771
	sysfs_put_active(attr_sd);
772

773
	if (of->event != atomic_read(&od->event))
774
		goto trigger;
775

776
	return DEFAULT_POLLMASK;
777
778

 trigger:
779
	return DEFAULT_POLLMASK|POLLERR|POLLPRI;
780
781
}

782
783
784
void sysfs_notify_dirent(struct sysfs_dirent *sd)
{
	struct sysfs_open_dirent *od;
785
	unsigned long flags;
786

787
	spin_lock_irqsave(&sysfs_open_dirent_lock, flags);
788

789
790
791
792
793
794
	if (!WARN_ON(sysfs_type(sd) != SYSFS_KOBJ_ATTR)) {
		od = sd->s_attr.open;
		if (od) {
			atomic_inc(&od->event);
			wake_up_interruptible(&od->poll);
		}
795
796
	}

797
	spin_unlock_irqrestore(&sysfs_open_dirent_lock, flags);
798
799
800
}
EXPORT_SYMBOL_GPL(sysfs_notify_dirent);

801
void sysfs_notify(struct kobject *k, const char *dir, const char *attr)
802
{
803
	struct sysfs_dirent *sd = k->sd;
804

805
806
807
	mutex_lock(&sysfs_mutex);

	if (sd && dir)
Tejun Heo's avatar
Tejun Heo committed
808
		sd = sysfs_find_dirent(sd, dir, NULL);
809
	if (sd && attr)
Tejun Heo's avatar
Tejun Heo committed
810
		sd = sysfs_find_dirent(sd, attr, NULL);
811
812
	if (sd)
		sysfs_notify_dirent(sd);
813
814

	mutex_unlock(&sysfs_mutex);
815
816
817
}
EXPORT_SYMBOL_GPL(sysfs_notify);

818
const struct file_operations sysfs_file_operations = {
819
	.read		= seq_read,
Linus Torvalds's avatar
Linus Torvalds committed
820
	.write		= sysfs_write_file,
821
	.llseek		= generic_file_llseek,
Linus Torvalds's avatar
Linus Torvalds committed
822
823
	.open		= sysfs_open_file,
	.release	= sysfs_release,
824
	.poll		= sysfs_poll,
Linus Torvalds's avatar
Linus Torvalds committed
825
826
};

827
const struct file_operations sysfs_bin_operations = {
Tejun Heo's avatar
Tejun Heo committed
828
	.read		= sysfs_bin_read,
829
830
	.write		= sysfs_write_file,
	.llseek		= generic_file_llseek,
831
	.mmap		= sysfs_bin_mmap,
832
833
834
	.open		= sysfs_open_file,
	.release	= sysfs_release,
	.poll		= sysfs_poll,
835
836
};

837
838
839
int sysfs_add_file_mode_ns(struct sysfs_dirent *dir_sd,
			   const struct attribute *attr, int type,
			   umode_t amode, const void *ns)
Linus Torvalds's avatar
Linus Torvalds committed
840
{
841
	umode_t mode = (amode & S_IALLUGO) | S_IFREG;
842
	struct sysfs_addrm_cxt acxt;
843
	struct sysfs_dirent *sd;
844
	int rc;
Linus Torvalds's avatar
Linus Torvalds committed
845

846
847
848
	sd = sysfs_new_dirent(attr->name, mode, type);
	if (!sd)
		return -ENOMEM;
849
850

	sd->s_ns = ns;
851
	sd->s_attr.attr = (void *)attr;
852
	sysfs_dirent_init_lockdep(sd);
Linus Torvalds's avatar
Linus Torvalds committed
853

854
855
	sysfs_addrm_start(&acxt);
	rc = sysfs_add_one(&acxt, sd, dir_sd);
856
	sysfs_addrm_finish(&acxt);
857

858
	if (rc)
859
		sysfs_put(sd);
860

861
	return rc;
Linus Torvalds's avatar
Linus Torvalds committed
862
863
864
}


865
866
867
int sysfs_add_file(struct sysfs_dirent *dir_sd, const struct attribute *attr,
		   int type)
{
868
	return sysfs_add_file_mode_ns(dir_sd, attr, type, attr->mode, NULL);
869
870
}

Linus Torvalds's avatar
Linus Torvalds committed
871
/**
872
873
874
875
 * sysfs_create_file_ns - create an attribute file for an object with custom ns
 * @kobj: object we're creating for
 * @attr: attribute descriptor
 * @ns: namespace the new file should belong to
Linus Torvalds's avatar
Linus Torvalds committed
876
 */
877
878
int sysfs_create_file_ns(struct kobject *kobj, const struct attribute *attr,
			 const void *ns)
Linus Torvalds's avatar
Linus Torvalds committed
879
{
880
	BUG_ON(!kobj || !kobj->sd || !attr);
Linus Torvalds's avatar
Linus Torvalds committed
881

882
883
	return sysfs_add_file_mode_ns(kobj->sd, attr, SYSFS_KOBJ_ATTR,
				      attr->mode, ns);
Linus Torvalds's avatar
Linus Torvalds committed
884
885

}
886
EXPORT_SYMBOL_GPL(sysfs_create_file_ns);
Linus Torvalds's avatar
Linus Torvalds committed
887

888
889
890
891
892
893
894
895
896
897
898
899
int sysfs_create_files(struct kobject *kobj, const struct attribute **ptr)
{
	int err = 0;
	int i;

	for (i = 0; ptr[i] && !err; i++)
		err = sysfs_create_file(kobj, ptr[i]);
	if (err)
		while (--i >= 0)
			sysfs_remove_file(kobj, ptr[i]);
	return err;
}
900
EXPORT_SYMBOL_GPL(sysfs_create_files);
Linus Torvalds's avatar
Linus Torvalds committed
901

902
903
904
905
906
907
908
909
910
/**
 * sysfs_add_file_to_group - add an attribute file to a pre-existing group.
 * @kobj: object we're acting for.
 * @attr: attribute descriptor.
 * @group: group name.
 */
int sysfs_add_file_to_group(struct kobject *kobj,
		const struct attribute *attr, const char *group)
{
911
	struct sysfs_dirent *dir_sd;
912
913
	int error;

914
	if (group)
Tejun Heo's avatar
Tejun Heo committed
915
		dir_sd = sysfs_get_dirent(kobj->sd, group);
916
917
918
	else
		dir_sd = sysfs_get(kobj->sd);

919
920
921
922
923
924
	if (!dir_sd)
		return -ENOENT;

	error = sysfs_add_file(dir_sd, attr, SYSFS_KOBJ_ATTR);
	sysfs_put(dir_sd);

925
926
927
928
	return error;
}
EXPORT_SYMBOL_GPL(sysfs_add_file_to_group);

929
930
931
932
933
934
935
/**
 * sysfs_chmod_file - update the modified mode value on an object attribute.
 * @kobj: object we're acting for.
 * @attr: attribute descriptor.
 * @mode: file permissions.
 *
 */
936
int sysfs_chmod_file(struct kobject *kobj, const struct attribute *attr,
Al Viro's avatar
Al Viro committed
937
		     umode_t mode)
938
{
939
	struct sysfs_dirent *sd;
940
	struct iattr newattrs;
941
942
	int rc;

943
	mutex_lock(&sysfs_mutex);
944

945
	rc = -ENOENT;
Tejun Heo's avatar
Tejun Heo committed
946
	sd = sysfs_find_dirent(kobj->sd, attr->name, NULL);
947
	if (!sd)
948
		goto out;
949

950
	newattrs.ia_mode = (mode & S_IALLUGO) | (sd->s_mode & ~S_IALLUGO);
951
	newattrs.ia_valid = ATTR_MODE;
952
	rc = sysfs_sd_setattr(sd, &newattrs);
953

954
 out:
955
	mutex_unlock(&sysfs_mutex);
956
	return rc;
957
958
959
}
EXPORT_SYMBOL_GPL(sysfs_chmod_file);

Linus Torvalds's avatar
Linus Torvalds committed
960
/**
961
962
963
964
 * sysfs_remove_file_ns - remove an object attribute with a custom ns tag
 * @kobj: object we're acting for
 * @attr: attribute descriptor
 * @ns: namespace tag of the file to remove
Linus Torvalds's avatar
Linus Torvalds committed
965
 *
966
 * Hash the attribute name and namespace tag and kill the victim.
Linus Torvalds's avatar
Linus Torvalds committed
967
 */
968
969
void sysfs_remove_file_ns(struct kobject *kobj, const struct attribute *attr,
			  const void *ns)
Linus Torvalds's avatar
Linus Torvalds committed
970
{
971
	struct sysfs_dirent *dir_sd = kobj->sd;
972

Tejun Heo's avatar
Tejun Heo committed
973
	sysfs_hash_and_remove(dir_sd, attr->name, ns);
Linus Torvalds's avatar
Linus Torvalds committed
974
}
975
EXPORT_SYMBOL_GPL(sysfs_remove_file_ns);
Linus Torvalds's avatar
Linus Torvalds committed
976

977
void sysfs_remove_files(struct kobject *kobj, const struct attribute **ptr)
978
979
980
981
982
{
	int i;
	for (i = 0; ptr[i]; i++)
		sysfs_remove_file(kobj, ptr[i]);
}
983
EXPORT_SYMBOL_GPL(sysfs_remove_files);
Linus Torvalds's avatar
Linus Torvalds committed
984

985
986
987
988
989
990
991
992
993
/**
 * sysfs_remove_file_from_group - remove an attribute file from a group.
 * @kobj: object we're acting for.
 * @attr: attribute descriptor.
 * @group: group name.
 */
void sysfs_remove_file_from_group(struct kobject *kobj,
		const struct attribute *attr, const char *group)
{
994
	struct sysfs_dirent *dir_sd;
995

996
	if (group)
Tejun Heo's avatar
Tejun Heo committed
997
		dir_sd = sysfs_get_dirent(kobj->sd, group);
998
999
	else
		dir_sd = sysfs_get(kobj->sd);
1000
	if (dir_sd) {
Tejun Heo's avatar
Tejun Heo committed
1001
		sysfs_hash_and_remove(dir_sd, attr->name, NULL);
1002
		sysfs_put(dir_sd);
1003
1004
1005
1006
	}
}
EXPORT_SYMBOL_GPL(sysfs_remove_file_from_group);

1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
/**
 *	sysfs_create_bin_file - create binary file for object.
 *	@kobj:	object.
 *	@attr:	attribute descriptor.
 */
int sysfs_create_bin_file(struct kobject *kobj,
			  const struct bin_attribute *attr)
{
	BUG_ON(!kobj || !kobj->sd || !attr);

	return sysfs_add_file(kobj->sd, &attr->attr, SYSFS_KOBJ_BIN_ATTR);
}
EXPORT_SYMBOL_GPL(sysfs_create_bin_file);

/**
 *	sysfs_remove_bin_file - remove binary file for object.
 *	@kobj:	object.
 *	@attr:	attribute descriptor.
 */
void sysfs_remove_bin_file(struct kobject *kobj,
			   const struct bin_attribute *attr)
{
	sysfs_hash_and_remove(kobj->sd, attr->attr.name, NULL);
}
EXPORT_SYMBOL_GPL(sysfs_remove_bin_file);

1033
struct sysfs_schedule_callback_struct {
1034
1035
	struct list_head	workq_list;
	struct kobject		*kobj;
1036
1037
	void			(*func)(void *);
	void			*data;
1038
	struct module		*owner;
1039
1040
1041
	struct work_struct	work;
};

1042
static struct workqueue_struct *sysfs_workqueue;
1043
1044
static DEFINE_MUTEX(sysfs_workq_mutex);
static LIST_HEAD(sysfs_workq);
1045
1046
1047
1048
1049
1050
1051
static void sysfs_schedule_callback_work(struct work_struct *work)
{
	struct sysfs_schedule_callback_struct *ss = container_of(work,
			struct sysfs_schedule_callback_struct, work);

	(ss->func)(ss->data);
	kobject_put(ss->kobj);
1052
	module_put(ss->owner);
1053
1054
1055
	mutex_lock(&sysfs_workq_mutex);
	list_del(&ss->workq_list);
	mutex_unlock(&sysfs_workq_mutex);
1056
1057
1058
1059
1060
1061
1062
1063
	kfree(ss);
}

/**
 * sysfs_schedule_callback - helper to schedule a callback for a kobject
 * @kobj: object we're acting for.
 * @func: callback function to invoke later.
 * @data: argument to pass to @func.
1064
 * @owner: module owning the callback code
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
 *
 * sysfs attribute methods must not unregister themselves or their parent
 * kobject (which would amount to the same thing).  Attempts to do so will
 * deadlock, since unregistration is mutually exclusive with driver
 * callbacks.
 *
 * Instead methods can call this routine, which will attempt to allocate
 * and schedule a workqueue request to call back @func with @data as its
 * argument in the workqueue's process context.  @kobj will be pinned
 * until @func returns.
 *
 * Returns 0 if the request was submitted, -ENOMEM if storage could not
1077
1078
 * be allocated, -ENODEV if a reference to @owner isn't available,
 * -EAGAIN if a callback has already been scheduled for @kobj.
1079
1080
 */
int sysfs_schedule_callback(struct kobject *kobj, void (*func)(void *),
1081
		void *data, struct module *owner)
1082
{
1083
	struct sysfs_schedule_callback_struct *ss, *tmp;
1084

1085
1086
	if (!try_module_get(owner))
		return -ENODEV;
1087
1088
1089
1090

	mutex_lock(&sysfs_workq_mutex);
	list_for_each_entry_safe(ss, tmp, &sysfs_workq, workq_list)
		if (ss->kobj == kobj) {
1091
			module_put(owner);
1092
1093
1094
1095
1096
			mutex_unlock(&sysfs_workq_mutex);
			return -EAGAIN;
		}
	mutex_unlock(&sysfs_workq_mutex);

1097
	if (sysfs_workqueue == NULL) {
1098
		sysfs_workqueue = create_singlethread_workqueue("sysfsd");
1099
1100
1101
1102
1103
1104
		if (sysfs_workqueue == NULL) {
			module_put(owner);
			return -ENOMEM;
		}
	}

1105
	ss = kmalloc(sizeof(*ss), GFP_KERNEL);
1106
1107
	if (!ss) {
		module_put(owner);
1108
		return -ENOMEM;
1109
	}
1110
1111
1112
1113
	kobject_get(kobj);
	ss->kobj = kobj;
	ss->func = func;
	ss->data = data;
1114
	ss->owner = owner;
1115
	INIT_WORK(&ss->work, sysfs_schedule_callback_work);
1116
1117
1118
1119
	INIT_LIST_HEAD(&ss->workq_list);
	mutex_lock(&sysfs_workq_mutex);
	list_add_tail(&ss->workq_list, &sysfs_workq);
	mutex_unlock(&sysfs_workq_mutex);
1120
	queue_work(sysfs_workqueue, &ss->work);
1121
1122
1123
	return 0;
}
EXPORT_SYMBOL_GPL(sysfs_schedule_callback);