1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
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
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
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
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
|
/*
* Copyright (C) 2008 Red Hat, Inc., Eric Paris <eparis@redhat.com>
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation; either version 2, or (at your option)
* any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; see the file COPYING. If not, write to
* the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA.
*/
/*
* Basic idea behind the notification queue: An fsnotify group (like inotify)
* sends the userspace notification about events asyncronously some time after
* the event happened. When inotify gets an event it will need to add that
* event to the group notify queue. Since a single event might need to be on
* multiple group's notification queues we can't add the event directly to each
* queue and instead add a small "event_holder" to each queue. This event_holder
* has a pointer back to the original event. Since the majority of events are
* going to end up on one, and only one, notification queue we embed one
* event_holder into each event. This means we have a single allocation instead
* of always needing two. If the embedded event_holder is already in use by
* another group a new event_holder (from fsnotify_event_holder_cachep) will be
* allocated and used.
*/
#include <linux/fs.h>
#include <linux/init.h>
#include <linux/kernel.h>
#include <linux/list.h>
#include <linux/module.h>
#include <linux/mount.h>
#include <linux/mutex.h>
#include <linux/namei.h>
#include <linux/path.h>
#include <linux/slab.h>
#include <linux/spinlock.h>
#include <asm/atomic.h>
#include <linux/fsnotify_backend.h>
#include "fsnotify.h"
static struct kmem_cache *fsnotify_event_cachep;
static struct kmem_cache *fsnotify_event_holder_cachep;
/*
* This is a magic event we send when the q is too full. Since it doesn't
* hold real event information we just keep one system wide and use it any time
* it is needed. It's refcnt is set 1 at kernel init time and will never
* get set to 0 so it will never get 'freed'
*/
static struct fsnotify_event *q_overflow_event;
static atomic_t fsnotify_sync_cookie = ATOMIC_INIT(0);
/**
* fsnotify_get_cookie - return a unique cookie for use in synchronizing events.
* Called from fsnotify_move, which is inlined into filesystem modules.
*/
u32 fsnotify_get_cookie(void)
{
return atomic_inc_return(&fsnotify_sync_cookie);
}
EXPORT_SYMBOL_GPL(fsnotify_get_cookie);
/* return true if the notify queue is empty, false otherwise */
bool fsnotify_notify_queue_is_empty(struct fsnotify_group *group)
{
BUG_ON(!mutex_is_locked(&group->notification_mutex));
return list_empty(&group->notification_list) ? true : false;
}
void fsnotify_get_event(struct fsnotify_event *event)
{
atomic_inc(&event->refcnt);
}
void fsnotify_put_event(struct fsnotify_event *event)
{
if (!event)
return;
if (atomic_dec_and_test(&event->refcnt)) {
pr_debug("%s: event=%p\n", __func__, event);
if (event->data_type == FSNOTIFY_EVENT_PATH)
path_put(&event->path);
BUG_ON(!list_empty(&event->private_data_list));
kfree(event->file_name);
put_pid(event->tgid);
kmem_cache_free(fsnotify_event_cachep, event);
}
}
struct fsnotify_event_holder *fsnotify_alloc_event_holder(void)
{
return kmem_cache_alloc(fsnotify_event_holder_cachep, GFP_KERNEL);
}
void fsnotify_destroy_event_holder(struct fsnotify_event_holder *holder)
{
if (holder)
kmem_cache_free(fsnotify_event_holder_cachep, holder);
}
/*
* Find the private data that the group previously attached to this event when
* the group added the event to the notification queue (fsnotify_add_notify_event)
*/
struct fsnotify_event_private_data *fsnotify_remove_priv_from_event(struct fsnotify_group *group, struct fsnotify_event *event)
{
struct fsnotify_event_private_data *lpriv;
struct fsnotify_event_private_data *priv = NULL;
assert_spin_locked(&event->lock);
list_for_each_entry(lpriv, &event->private_data_list, event_list) {
if (lpriv->group == group) {
priv = lpriv;
list_del(&priv->event_list);
break;
}
}
return priv;
}
/*
* Add an event to the group notification queue. The group can later pull this
* event off the queue to deal with. If the event is successfully added to the
* group's notification queue, a reference is taken on event.
*/
struct fsnotify_event *fsnotify_add_notify_event(struct fsnotify_group *group, struct fsnotify_event *event,
struct fsnotify_event_private_data *priv,
struct fsnotify_event *(*merge)(struct list_head *,
struct fsnotify_event *))
{
struct fsnotify_event *return_event = NULL;
struct fsnotify_event_holder *holder = NULL;
struct list_head *list = &group->notification_list;
pr_debug("%s: group=%p event=%p priv=%p\n", __func__, group, event, priv);
/*
* There is one fsnotify_event_holder embedded inside each fsnotify_event.
* Check if we expect to be able to use that holder. If not alloc a new
* holder.
* For the overflow event it's possible that something will use the in
* event holder before we get the lock so we may need to jump back and
* alloc a new holder, this can't happen for most events...
*/
if (!list_empty(&event->holder.event_list)) {
alloc_holder:
holder = fsnotify_alloc_event_holder();
if (!holder)
return ERR_PTR(-ENOMEM);
}
mutex_lock(&group->notification_mutex);
if (group->q_len >= group->max_events) {
event = q_overflow_event;
/*
* we need to return the overflow event
* which means we need a ref
*/
fsnotify_get_event(event);
return_event = event;
/* sorry, no private data on the overflow event */
priv = NULL;
}
if (!list_empty(list) && merge) {
struct fsnotify_event *tmp;
tmp = merge(list, event);
if (tmp) {
mutex_unlock(&group->notification_mutex);
if (return_event)
fsnotify_put_event(return_event);
if (holder != &event->holder)
fsnotify_destroy_event_holder(holder);
return tmp;
}
}
spin_lock(&event->lock);
if (list_empty(&event->holder.event_list)) {
if (unlikely(holder))
fsnotify_destroy_event_holder(holder);
holder = &event->holder;
} else if (unlikely(!holder)) {
/* between the time we checked above and got the lock the in
* event holder was used, go back and get a new one */
spin_unlock(&event->lock);
mutex_unlock(&group->notification_mutex);
if (return_event) {
fsnotify_put_event(return_event);
return_event = NULL;
}
goto alloc_holder;
}
group->q_len++;
holder->event = event;
fsnotify_get_event(event);
list_add_tail(&holder->event_list, list);
if (priv)
list_add_tail(&priv->event_list, &event->private_data_list);
spin_unlock(&event->lock);
mutex_unlock(&group->notification_mutex);
wake_up(&group->notification_waitq);
return return_event;
}
/*
* Remove and return the first event from the notification list. There is a
* reference held on this event since it was on the list. It is the responsibility
* of the caller to drop this reference.
*/
struct fsnotify_event *fsnotify_remove_notify_event(struct fsnotify_group *group)
{
struct fsnotify_event *event;
struct fsnotify_event_holder *holder;
BUG_ON(!mutex_is_locked(&group->notification_mutex));
pr_debug("%s: group=%p\n", __func__, group);
holder = list_first_entry(&group->notification_list, struct fsnotify_event_holder, event_list);
event = holder->event;
spin_lock(&event->lock);
holder->event = NULL;
list_del_init(&holder->event_list);
spin_unlock(&event->lock);
/* event == holder means we are referenced through the in event holder */
if (holder != &event->holder)
fsnotify_destroy_event_holder(holder);
group->q_len--;
return event;
}
/*
* This will not remove the event, that must be done with fsnotify_remove_notify_event()
*/
struct fsnotify_event *fsnotify_peek_notify_event(struct fsnotify_group *group)
{
struct fsnotify_event *event;
struct fsnotify_event_holder *holder;
BUG_ON(!mutex_is_locked(&group->notification_mutex));
holder = list_first_entry(&group->notification_list, struct fsnotify_event_holder, event_list);
event = holder->event;
return event;
}
/*
* Called when a group is being torn down to clean up any outstanding
* event notifications.
*/
void fsnotify_flush_notify(struct fsnotify_group *group)
{
struct fsnotify_event *event;
struct fsnotify_event_private_data *priv;
mutex_lock(&group->notification_mutex);
while (!fsnotify_notify_queue_is_empty(group)) {
event = fsnotify_remove_notify_event(group);
/* if they don't implement free_event_priv they better not have attached any */
if (group->ops->free_event_priv) {
spin_lock(&event->lock);
priv = fsnotify_remove_priv_from_event(group, event);
spin_unlock(&event->lock);
if (priv)
group->ops->free_event_priv(priv);
}
fsnotify_put_event(event); /* matches fsnotify_add_notify_event */
}
mutex_unlock(&group->notification_mutex);
}
static void initialize_event(struct fsnotify_event
|