eventpoll.c 60.1 KB
Newer Older
Linus Torvalds's avatar
Linus Torvalds committed
1
/*
2 3
 *  fs/eventpoll.c (Efficient event retrieval implementation)
 *  Copyright (C) 2001,...,2009	 Davide Libenzi
Linus Torvalds's avatar
Linus Torvalds committed
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
 *
 *  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 of the License, or
 *  (at your option) any later version.
 *
 *  Davide Libenzi <davidel@xmailserver.org>
 *
 */

#include <linux/init.h>
#include <linux/kernel.h>
#include <linux/sched.h>
#include <linux/fs.h>
#include <linux/file.h>
#include <linux/signal.h>
#include <linux/errno.h>
#include <linux/mm.h>
#include <linux/slab.h>
#include <linux/poll.h>
#include <linux/string.h>
#include <linux/list.h>
#include <linux/hash.h>
#include <linux/spinlock.h>
#include <linux/syscalls.h>
#include <linux/rbtree.h>
#include <linux/wait.h>
#include <linux/eventpoll.h>
#include <linux/mount.h>
#include <linux/bitops.h>
34
#include <linux/mutex.h>
35
#include <linux/anon_inodes.h>
36
#include <linux/device.h>
Linus Torvalds's avatar
Linus Torvalds committed
37 38 39
#include <asm/uaccess.h>
#include <asm/io.h>
#include <asm/mman.h>
40
#include <linux/atomic.h>
41 42
#include <linux/proc_fs.h>
#include <linux/seq_file.h>
43
#include <linux/compat.h>
44
#include <linux/rculist.h>
Linus Torvalds's avatar
Linus Torvalds committed
45 46 47 48 49

/*
 * LOCKING:
 * There are three level of locking required by epoll :
 *
50
 * 1) epmutex (mutex)
51 52
 * 2) ep->mtx (mutex)
 * 3) ep->lock (spinlock)
Linus Torvalds's avatar
Linus Torvalds committed
53 54 55 56 57 58 59 60 61
 *
 * The acquire order is the one listed above, from 1 to 3.
 * We need a spinlock (ep->lock) because we manipulate objects
 * from inside the poll callback, that might be triggered from
 * a wake_up() that in turn might be called from IRQ context.
 * So we can't sleep inside the poll callback and hence we need
 * a spinlock. During the event transfer loop (from kernel to
 * user space) we could end up sleeping due a copy_to_user(), so
 * we need a lock that will allow us to sleep. This lock is a
62 63 64 65 66
 * mutex (ep->mtx). It is acquired during the event transfer loop,
 * during epoll_ctl(EPOLL_CTL_DEL) and during eventpoll_release_file().
 * Then we also need a global mutex to serialize eventpoll_release_file()
 * and ep_free().
 * This mutex is acquired by ep_free() during the epoll file
Linus Torvalds's avatar
Linus Torvalds committed
67 68
 * cleanup path and it is also acquired by eventpoll_release_file()
 * if a file has been pushed inside an epoll set and it is then
69
 * close()d without a previous call to epoll_ctl(EPOLL_CTL_DEL).
70 71 72 73 74 75 76
 * It is also acquired when inserting an epoll fd onto another epoll
 * fd. We do this so that we walk the epoll tree and ensure that this
 * insertion does not create a cycle of epoll file descriptors, which
 * could lead to deadlock. We need a global mutex to prevent two
 * simultaneous inserts (A into B and B into A) from racing and
 * constructing a cycle without either insert observing that it is
 * going to.
77 78 79 80 81 82 83 84 85
 * It is necessary to acquire multiple "ep->mtx"es at once in the
 * case when one epoll fd is added to another. In this case, we
 * always acquire the locks in the order of nesting (i.e. after
 * epoll_ctl(e1, EPOLL_CTL_ADD, e2), e1->mtx will always be acquired
 * before e2->mtx). Since we disallow cycles of epoll file
 * descriptors, this ensures that the mutexes are well-ordered. In
 * order to communicate this nesting to lockdep, when walking a tree
 * of epoll file descriptors, we use the current recursion depth as
 * the lockdep subkey.
86 87 88
 * It is possible to drop the "ep->mtx" and to use the global
 * mutex "epmutex" (together with "ep->lock") to have it working,
 * but having "ep->mtx" will make the interface more scalable.
89
 * Events that require holding "epmutex" are very rare, while for
90 91
 * normal operations the epoll private "ep->mtx" will guarantee
 * a better scalability.
Linus Torvalds's avatar
Linus Torvalds committed
92 93 94
 */

/* Epoll private bits inside the event mask */
95
#define EP_PRIVATE_BITS (EPOLLWAKEUP | EPOLLONESHOT | EPOLLET | EPOLLEXCLUSIVE)
Linus Torvalds's avatar
Linus Torvalds committed
96

97 98 99 100 101
#define EPOLLINOUT_BITS (POLLIN | POLLOUT)

#define EPOLLEXCLUSIVE_OK_BITS (EPOLLINOUT_BITS | POLLERR | POLLHUP | \
				EPOLLWAKEUP | EPOLLET | EPOLLEXCLUSIVE)

102 103
/* Maximum number of nesting allowed inside epoll sets */
#define EP_MAX_NESTS 4
Linus Torvalds's avatar
Linus Torvalds committed
104

Davide Libenzi's avatar
Davide Libenzi committed
105 106
#define EP_MAX_EVENTS (INT_MAX / sizeof(struct epoll_event))

107 108
#define EP_UNACTIVE_PTR ((void *) -1L)

109 110
#define EP_ITEM_COST (sizeof(struct epitem) + sizeof(struct eppoll_entry))

Linus Torvalds's avatar
Linus Torvalds committed
111 112 113
struct epoll_filefd {
	struct file *file;
	int fd;
114
} __packed;
Linus Torvalds's avatar
Linus Torvalds committed
115 116

/*
117 118
 * Structure used to track possible nested calls, for too deep recursions
 * and loop cycles.
Linus Torvalds's avatar
Linus Torvalds committed
119
 */
120
struct nested_call_node {
Linus Torvalds's avatar
Linus Torvalds committed
121
	struct list_head llink;
122
	void *cookie;
123
	void *ctx;
Linus Torvalds's avatar
Linus Torvalds committed
124 125 126
};

/*
127 128
 * This structure is used as collector for nested calls, to check for
 * maximum recursion dept and loop cycles.
Linus Torvalds's avatar
Linus Torvalds committed
129
 */
130 131
struct nested_calls {
	struct list_head tasks_call_list;
Linus Torvalds's avatar
Linus Torvalds committed
132 133 134
	spinlock_t lock;
};

135 136 137
/*
 * Each file descriptor added to the eventpoll interface will
 * have an entry of this type linked to the "rbr" RB tree.
138 139
 * Avoid increasing the size of this struct, there can be many thousands
 * of these on a server and we do not want this to take another cache line.
140 141
 */
struct epitem {
142 143 144 145 146 147
	union {
		/* RB tree node links this structure to the eventpoll RB tree */
		struct rb_node rbn;
		/* Used to free the struct epitem */
		struct rcu_head rcu;
	};
148 149 150 151

	/* List header used to link this structure to the eventpoll ready list */
	struct list_head rdllink;

152 153 154 155 156 157
	/*
	 * Works together "struct eventpoll"->ovflist in keeping the
	 * single linked chain of items.
	 */
	struct epitem *next;

158 159 160 161 162 163 164 165 166 167 168 169 170 171 172
	/* The file descriptor information this item refers to */
	struct epoll_filefd ffd;

	/* Number of active wait queue attached to poll operations */
	int nwait;

	/* List containing poll wait queues */
	struct list_head pwqlist;

	/* The "container" of this item */
	struct eventpoll *ep;

	/* List header used to link this item to the "struct file" items list */
	struct list_head fllink;

173
	/* wakeup_source used when EPOLLWAKEUP is set */
174
	struct wakeup_source __rcu *ws;
175

176 177
	/* The structure that describe the interested events and the source fd */
	struct epoll_event event;
178 179
};

Linus Torvalds's avatar
Linus Torvalds committed
180 181
/*
 * This structure is stored inside the "private_data" member of the file
182
 * structure and represents the main data structure for the eventpoll
Linus Torvalds's avatar
Linus Torvalds committed
183 184 185
 * interface.
 */
struct eventpoll {
186
	/* Protect the access to this structure */
187
	spinlock_t lock;
Linus Torvalds's avatar
Linus Torvalds committed
188 189

	/*
190 191 192 193
	 * This mutex is used to ensure that files are not removed
	 * while epoll is using them. This is held during the event
	 * collection loop, the file cleanup path, the epoll file exit
	 * code and the ctl operations.
Linus Torvalds's avatar
Linus Torvalds committed
194
	 */
195
	struct mutex mtx;
Linus Torvalds's avatar
Linus Torvalds committed
196 197 198 199 200 201 202 203 204 205

	/* Wait queue used by sys_epoll_wait() */
	wait_queue_head_t wq;

	/* Wait queue used by file->poll() */
	wait_queue_head_t poll_wait;

	/* List of ready file descriptors */
	struct list_head rdllist;

206
	/* RB tree root used to store monitored fd structs */
Linus Torvalds's avatar
Linus Torvalds committed
207
	struct rb_root rbr;
208 209 210

	/*
	 * This is a single linked list that chains all the "struct epitem" that
211
	 * happened while transferring ready events to userspace w/out
212 213 214
	 * holding ->lock.
	 */
	struct epitem *ovflist;
215

216 217 218
	/* wakeup_source used when ep_scan_ready_list is running */
	struct wakeup_source *ws;

219 220
	/* The user that created the eventpoll descriptor */
	struct user_struct *user;
Jason Baron's avatar
Jason Baron committed
221 222 223 224 225 226

	struct file *file;

	/* used to optimize loop detection check */
	int visited;
	struct list_head visited_list_link;
Linus Torvalds's avatar
Linus Torvalds committed
227 228 229 230 231 232 233 234
};

/* Wait structure used by the poll hooks */
struct eppoll_entry {
	/* List header used to link this structure to the "struct epitem" */
	struct list_head llink;

	/* The "base" pointer is set to the container "struct epitem" */
235
	struct epitem *base;
Linus Torvalds's avatar
Linus Torvalds committed
236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252

	/*
	 * Wait queue item that will be linked to the target file wait
	 * queue head.
	 */
	wait_queue_t wait;

	/* The wait queue head that linked the "wait" wait queue item */
	wait_queue_head_t *whead;
};

/* Wrapper struct used by poll queueing */
struct ep_pqueue {
	poll_table pt;
	struct epitem *epi;
};

253 254 255 256 257 258
/* Used by the ep_send_events() function as callback private data */
struct ep_send_events_data {
	int maxevents;
	struct epoll_event __user *events;
};

259 260 261 262
/*
 * Configuration options available inside /proc/sys/fs/epoll/
 */
/* Maximum number of epoll watched descriptors, per user */
263
static long max_user_watches __read_mostly;
264

Linus Torvalds's avatar
Linus Torvalds committed
265
/*
266
 * This mutex is used to serialize ep_free() and eventpoll_release_file().
Linus Torvalds's avatar
Linus Torvalds committed
267
 */
268
static DEFINE_MUTEX(epmutex);
Linus Torvalds's avatar
Linus Torvalds committed
269

270 271 272
/* Used to check for epoll file descriptor inclusion loops */
static struct nested_calls poll_loop_ncalls;

273 274 275 276 277
/* Used for safe wake up implementation */
static struct nested_calls poll_safewake_ncalls;

/* Used to call file's f_op->poll() under the nested calls boundaries */
static struct nested_calls poll_readywalk_ncalls;
Linus Torvalds's avatar
Linus Torvalds committed
278 279

/* Slab cache used to allocate "struct epitem" */
280
static struct kmem_cache *epi_cache __read_mostly;
Linus Torvalds's avatar
Linus Torvalds committed
281 282

/* Slab cache used to allocate "struct eppoll_entry" */
283
static struct kmem_cache *pwq_cache __read_mostly;
Linus Torvalds's avatar
Linus Torvalds committed
284

Jason Baron's avatar
Jason Baron committed
285 286 287 288 289 290 291 292 293
/* Visited nodes during ep_loop_check(), so we can unset them when we finish */
static LIST_HEAD(visited_list);

/*
 * List of files with newly added links, where we may need to limit the number
 * of emanating paths. Protected by the epmutex.
 */
static LIST_HEAD(tfile_check_list);

294 295 296 297
#ifdef CONFIG_SYSCTL

#include <linux/sysctl.h>

298 299
static long zero;
static long long_max = LONG_MAX;
300

301
struct ctl_table epoll_table[] = {
302 303 304
	{
		.procname	= "max_user_watches",
		.data		= &max_user_watches,
305
		.maxlen		= sizeof(max_user_watches),
306
		.mode		= 0644,
307
		.proc_handler	= proc_doulongvec_minmax,
308
		.extra1		= &zero,
309
		.extra2		= &long_max,
310
	},
311
	{ }
312 313 314
};
#endif /* CONFIG_SYSCTL */

Jason Baron's avatar
Jason Baron committed
315 316 317 318 319 320
static const struct file_operations eventpoll_fops;

static inline int is_file_epoll(struct file *f)
{
	return f->f_op == &eventpoll_fops;
}
321

322
/* Setup the structure that is used as key for the RB tree */
323 324 325 326 327 328 329
static inline void ep_set_ffd(struct epoll_filefd *ffd,
			      struct file *file, int fd)
{
	ffd->file = file;
	ffd->fd = fd;
}

330
/* Compare RB tree keys */
331 332 333 334 335 336 337 338 339 340 341 342 343
static inline int ep_cmp_ffd(struct epoll_filefd *p1,
			     struct epoll_filefd *p2)
{
	return (p1->file > p2->file ? +1:
	        (p1->file < p2->file ? -1 : p1->fd - p2->fd));
}

/* Tells us if the item is currently linked */
static inline int ep_is_linked(struct list_head *p)
{
	return !list_empty(p);
}

344 345 346 347 348
static inline struct eppoll_entry *ep_pwq_from_wait(wait_queue_t *p)
{
	return container_of(p, struct eppoll_entry, wait);
}

349
/* Get the "struct epitem" from a wait queue pointer */
350
static inline struct epitem *ep_item_from_wait(wait_queue_t *p)
351 352 353 354 355
{
	return container_of(p, struct eppoll_entry, wait)->base;
}

/* Get the "struct epitem" from an epoll queue wrapper */
356
static inline struct epitem *ep_item_from_epqueue(poll_table *p)
357 358 359 360 361
{
	return container_of(p, struct ep_pqueue, pt)->epi;
}

/* Tells if the epoll_ctl(2) operation needs an event copy from userspace */
362
static inline int ep_op_has_event(int op)
363
{
364
	return op != EPOLL_CTL_DEL;
365 366
}

Linus Torvalds's avatar
Linus Torvalds committed
367
/* Initialize the poll safe wake up structure */
368
static void ep_nested_calls_init(struct nested_calls *ncalls)
Linus Torvalds's avatar
Linus Torvalds committed
369
{
370 371
	INIT_LIST_HEAD(&ncalls->tasks_call_list);
	spin_lock_init(&ncalls->lock);
Linus Torvalds's avatar
Linus Torvalds committed
372 373
}

374 375 376 377 378 379 380 381 382 383 384 385 386
/**
 * ep_events_available - Checks if ready events might be available.
 *
 * @ep: Pointer to the eventpoll context.
 *
 * Returns: Returns a value different than zero if ready events are available,
 *          or zero otherwise.
 */
static inline int ep_events_available(struct eventpoll *ep)
{
	return !list_empty(&ep->rdllist) || ep->ovflist != EP_UNACTIVE_PTR;
}

387 388 389 390 391 392 393 394 395 396 397
/**
 * ep_call_nested - Perform a bound (possibly) nested call, by checking
 *                  that the recursion limit is not exceeded, and that
 *                  the same nested call (by the meaning of same cookie) is
 *                  no re-entered.
 *
 * @ncalls: Pointer to the nested_calls structure to be used for this call.
 * @max_nests: Maximum number of allowed nesting calls.
 * @nproc: Nested call core function pointer.
 * @priv: Opaque data to be passed to the @nproc callback.
 * @cookie: Cookie to be used to identify this nested call.
398
 * @ctx: This instance context.
399 400 401
 *
 * Returns: Returns the code returned by the @nproc callback, or -1 if
 *          the maximum recursion limit has been exceeded.
Linus Torvalds's avatar
Linus Torvalds committed
402
 */
403 404
static int ep_call_nested(struct nested_calls *ncalls, int max_nests,
			  int (*nproc)(void *, void *, int), void *priv,
405
			  void *cookie, void *ctx)
Linus Torvalds's avatar
Linus Torvalds committed
406
{
407
	int error, call_nests = 0;
Linus Torvalds's avatar
Linus Torvalds committed
408
	unsigned long flags;
409 410 411
	struct list_head *lsthead = &ncalls->tasks_call_list;
	struct nested_call_node *tncur;
	struct nested_call_node tnode;
Linus Torvalds's avatar
Linus Torvalds committed
412

413
	spin_lock_irqsave(&ncalls->lock, flags);
Linus Torvalds's avatar
Linus Torvalds committed
414

415 416 417 418 419
	/*
	 * Try to see if the current task is already inside this wakeup call.
	 * We use a list here, since the population inside this set is always
	 * very much limited.
	 */
420
	list_for_each_entry(tncur, lsthead, llink) {
421
		if (tncur->ctx == ctx &&
422
		    (tncur->cookie == cookie || ++call_nests > max_nests)) {
Linus Torvalds's avatar
Linus Torvalds committed
423 424 425 426
			/*
			 * Ops ... loop detected or maximum nest level reached.
			 * We abort this wake by breaking the cycle itself.
			 */
427 428
			error = -1;
			goto out_unlock;
Linus Torvalds's avatar
Linus Torvalds committed
429 430 431
		}
	}

432
	/* Add the current task and cookie to the list */
433
	tnode.ctx = ctx;
434
	tnode.cookie = cookie;
Linus Torvalds's avatar
Linus Torvalds committed
435 436
	list_add(&tnode.llink, lsthead);

437
	spin_unlock_irqrestore(&ncalls->lock, flags);
Linus Torvalds's avatar
Linus Torvalds committed
438

439 440
	/* Call the nested function */
	error = (*nproc)(priv, cookie, call_nests);
Linus Torvalds's avatar
Linus Torvalds committed
441 442

	/* Remove the current task from the list */
443
	spin_lock_irqsave(&ncalls->lock, flags);
Linus Torvalds's avatar
Linus Torvalds committed
444
	list_del(&tnode.llink);
445
out_unlock:
446 447 448 449 450
	spin_unlock_irqrestore(&ncalls->lock, flags);

	return error;
}

451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475
/*
 * As described in commit 0ccf831cb lockdep: annotate epoll
 * the use of wait queues used by epoll is done in a very controlled
 * manner. Wake ups can nest inside each other, but are never done
 * with the same locking. For example:
 *
 *   dfd = socket(...);
 *   efd1 = epoll_create();
 *   efd2 = epoll_create();
 *   epoll_ctl(efd1, EPOLL_CTL_ADD, dfd, ...);
 *   epoll_ctl(efd2, EPOLL_CTL_ADD, efd1, ...);
 *
 * When a packet arrives to the device underneath "dfd", the net code will
 * issue a wake_up() on its poll wake list. Epoll (efd1) has installed a
 * callback wakeup entry on that queue, and the wake_up() performed by the
 * "dfd" net code will end up in ep_poll_callback(). At this point epoll
 * (efd1) notices that it may have some event ready, so it needs to wake up
 * the waiters on its poll wait list (efd2). So it calls ep_poll_safewake()
 * that ends up in another wake_up(), after having checked about the
 * recursion constraints. That are, no more than EP_MAX_POLLWAKE_NESTS, to
 * avoid stack blasting.
 *
 * When CONFIG_DEBUG_LOCK_ALLOC is enabled, make sure lockdep can handle
 * this special case of epoll.
 */
476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493
#ifdef CONFIG_DEBUG_LOCK_ALLOC
static inline void ep_wake_up_nested(wait_queue_head_t *wqueue,
				     unsigned long events, int subclass)
{
	unsigned long flags;

	spin_lock_irqsave_nested(&wqueue->lock, flags, subclass);
	wake_up_locked_poll(wqueue, events);
	spin_unlock_irqrestore(&wqueue->lock, flags);
}
#else
static inline void ep_wake_up_nested(wait_queue_head_t *wqueue,
				     unsigned long events, int subclass)
{
	wake_up_poll(wqueue, events);
}
#endif

494 495
static int ep_poll_wakeup_proc(void *priv, void *cookie, int call_nests)
{
496 497
	ep_wake_up_nested((wait_queue_head_t *) cookie, POLLIN,
			  1 + call_nests);
498 499 500 501 502 503 504 505 506 507 508 509 510 511 512
	return 0;
}

/*
 * Perform a safe wake up of the poll wait list. The problem is that
 * with the new callback'd wake up system, it is possible that the
 * poll callback is reentered from inside the call to wake_up() done
 * on the poll wait queue head. The rule is that we cannot reenter the
 * wake up code from the same task more than EP_MAX_NESTS times,
 * and we cannot reenter the same wait queue head at all. This will
 * enable to have a hierarchy of epoll file descriptor of no more than
 * EP_MAX_NESTS deep.
 */
static void ep_poll_safewake(wait_queue_head_t *wq)
{
513 514
	int this_cpu = get_cpu();

515
	ep_call_nested(&poll_safewake_ncalls, EP_MAX_NESTS,
516 517 518
		       ep_poll_wakeup_proc, NULL, wq, (void *) (long) this_cpu);

	put_cpu();
Linus Torvalds's avatar
Linus Torvalds committed
519 520
}

521 522 523 524 525 526 527 528 529 530 531 532
static void ep_remove_wait_queue(struct eppoll_entry *pwq)
{
	wait_queue_head_t *whead;

	rcu_read_lock();
	/* If it is cleared by POLLFREE, it should be rcu-safe */
	whead = rcu_dereference(pwq->whead);
	if (whead)
		remove_wait_queue(whead, &pwq->wait);
	rcu_read_unlock();
}

Linus Torvalds's avatar
Linus Torvalds committed
533
/*
534 535 536
 * This function unregisters poll callbacks from the associated file
 * descriptor.  Must be called with "mtx" held (or "epmutex" if called from
 * ep_free).
Linus Torvalds's avatar
Linus Torvalds committed
537
 */
538
static void ep_unregister_pollwait(struct eventpoll *ep, struct epitem *epi)
Linus Torvalds's avatar
Linus Torvalds committed
539
{
540 541
	struct list_head *lsthead = &epi->pwqlist;
	struct eppoll_entry *pwq;
Linus Torvalds's avatar
Linus Torvalds committed
542

543 544
	while (!list_empty(lsthead)) {
		pwq = list_first_entry(lsthead, struct eppoll_entry, llink);
Linus Torvalds's avatar
Linus Torvalds committed
545

546
		list_del(&pwq->llink);
547
		ep_remove_wait_queue(pwq);
548
		kmem_cache_free(pwq_cache, pwq);
Linus Torvalds's avatar
Linus Torvalds committed
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
/* call only when ep->mtx is held */
static inline struct wakeup_source *ep_wakeup_source(struct epitem *epi)
{
	return rcu_dereference_check(epi->ws, lockdep_is_held(&epi->ep->mtx));
}

/* call only when ep->mtx is held */
static inline void ep_pm_stay_awake(struct epitem *epi)
{
	struct wakeup_source *ws = ep_wakeup_source(epi);

	if (ws)
		__pm_stay_awake(ws);
}

static inline bool ep_has_wakeup_source(struct epitem *epi)
{
	return rcu_access_pointer(epi->ws) ? true : false;
}

/* call when ep->mtx cannot be held (ep_poll_callback) */
static inline void ep_pm_stay_awake_rcu(struct epitem *epi)
{
	struct wakeup_source *ws;

	rcu_read_lock();
	ws = rcu_dereference(epi->ws);
	if (ws)
		__pm_stay_awake(ws);
	rcu_read_unlock();
}

584 585 586 587 588 589 590 591
/**
 * ep_scan_ready_list - Scans the ready list in a way that makes possible for
 *                      the scan code, to call f_op->poll(). Also allows for
 *                      O(NumReady) performance.
 *
 * @ep: Pointer to the epoll private data structure.
 * @sproc: Pointer to the scan callback.
 * @priv: Private opaque data passed to the @sproc callback.
592
 * @depth: The current depth of recursive f_op->poll calls.
593
 * @ep_locked: caller already holds ep->mtx
594 595 596 597 598 599
 *
 * Returns: The same integer error code returned by the @sproc callback.
 */
static int ep_scan_ready_list(struct eventpoll *ep,
			      int (*sproc)(struct eventpoll *,
					   struct list_head *, void *),
600
			      void *priv, int depth, bool ep_locked)
601 602 603 604
{
	int error, pwake = 0;
	unsigned long flags;
	struct epitem *epi, *nepi;
605
	LIST_HEAD(txlist);
606 607 608

	/*
	 * We need to lock this because we could be hit by
609
	 * eventpoll_release_file() and epoll_ctl().
610
	 */
611 612 613

	if (!ep_locked)
		mutex_lock_nested(&ep->mtx, depth);
614 615 616 617 618 619 620 621 622 623

	/*
	 * Steal the ready list, and re-init the original one to the
	 * empty list. Also, set ep->ovflist to NULL so that events
	 * happening while looping w/out locks, are not lost. We cannot
	 * have the poll callback to queue directly on ep->rdllist,
	 * because we want the "sproc" callback to be able to do it
	 * in a lockless way.
	 */
	spin_lock_irqsave(&ep->lock, flags);
624
	list_splice_init(&ep->rdllist, &txlist);
625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646
	ep->ovflist = NULL;
	spin_unlock_irqrestore(&ep->lock, flags);

	/*
	 * Now call the callback function.
	 */
	error = (*sproc)(ep, &txlist, priv);

	spin_lock_irqsave(&ep->lock, flags);
	/*
	 * During the time we spent inside the "sproc" callback, some
	 * other events might have been queued by the poll callback.
	 * We re-insert them inside the main ready-list here.
	 */
	for (nepi = ep->ovflist; (epi = nepi) != NULL;
	     nepi = epi->next, epi->next = EP_UNACTIVE_PTR) {
		/*
		 * We need to check if the item is already in the list.
		 * During the "sproc" callback execution time, items are
		 * queued into ->ovflist but the "txlist" might already
		 * contain them, and the list_splice() below takes care of them.
		 */
647
		if (!ep_is_linked(&epi->rdllink)) {
648
			list_add_tail(&epi->rdllink, &ep->rdllist);
649
			ep_pm_stay_awake(epi);
650
		}
651 652 653 654 655 656 657 658 659 660 661 662
	}
	/*
	 * We need to set back ep->ovflist to EP_UNACTIVE_PTR, so that after
	 * releasing the lock, events will be queued in the normal way inside
	 * ep->rdllist.
	 */
	ep->ovflist = EP_UNACTIVE_PTR;

	/*
	 * Quickly re-inject items left on "txlist".
	 */
	list_splice(&txlist, &ep->rdllist);
663
	__pm_relax(ep->ws);
664 665 666

	if (!list_empty(&ep->rdllist)) {
		/*
667 668
		 * Wake up (if active) both the eventpoll wait list and
		 * the ->poll() wait list (delayed after we release the lock).
669 670 671 672 673 674 675 676
		 */
		if (waitqueue_active(&ep->wq))
			wake_up_locked(&ep->wq);
		if (waitqueue_active(&ep->poll_wait))
			pwake++;
	}
	spin_unlock_irqrestore(&ep->lock, flags);

677 678
	if (!ep_locked)
		mutex_unlock(&ep->mtx);
679 680 681 682 683 684 685 686

	/* We have to call this outside the lock */
	if (pwake)
		ep_poll_safewake(&ep->poll_wait);

	return error;
}

687 688 689 690 691 692
static void epi_rcu_free(struct rcu_head *head)
{
	struct epitem *epi = container_of(head, struct epitem, rcu);
	kmem_cache_free(epi_cache, epi);
}

693 694
/*
 * Removes a "struct epitem" from the eventpoll RB tree and deallocates
695
 * all the associated resources. Must be called with "mtx" held.
696 697 698 699 700
 */
static int ep_remove(struct eventpoll *ep, struct epitem *epi)
{
	unsigned long flags;
	struct file *file = epi->ffd.file;
Linus Torvalds's avatar
Linus Torvalds committed
701 702

	/*
703 704 705 706 707 708
	 * Removes poll wait queue hooks. We _have_ to do this without holding
	 * the "ep->lock" otherwise a deadlock might occur. This because of the
	 * sequence of the lock acquisition. Here we do "ep->lock" then the wait
	 * queue head lock when unregistering the wait queue. The wakeup callback
	 * will run by holding the wait queue head lock and will call our callback
	 * that will try to get "ep->lock".
Linus Torvalds's avatar
Linus Torvalds committed
709
	 */
710
	ep_unregister_pollwait(ep, epi);
Linus Torvalds's avatar
Linus Torvalds committed
711

712
	/* Remove the current item from the list of epoll hooks */
713
	spin_lock(&file->f_lock);
714
	list_del_rcu(&epi->fllink);
715
	spin_unlock(&file->f_lock);
Linus Torvalds's avatar
Linus Torvalds committed
716

717
	rb_erase(&epi->rbn, &ep->rbr);
Linus Torvalds's avatar
Linus Torvalds committed
718

719 720 721 722
	spin_lock_irqsave(&ep->lock, flags);
	if (ep_is_linked(&epi->rdllink))
		list_del_init(&epi->rdllink);
	spin_unlock_irqrestore(&ep->lock, flags);
Linus Torvalds's avatar
Linus Torvalds committed
723

724
	wakeup_source_unregister(ep_wakeup_source(epi));
725 726 727 728 729 730 731 732
	/*
	 * At this point it is safe to free the eventpoll item. Use the union
	 * field epi->rcu, since we are trying to minimize the size of
	 * 'struct epitem'. The 'rbn' field is no longer in use. Protected by
	 * ep->mtx. The rcu read side, reverse_path_check_proc(), does not make
	 * use of the rbn field.
	 */
	call_rcu(&epi->rcu, epi_rcu_free);
Linus Torvalds's avatar
Linus Torvalds committed
733

734
	atomic_long_dec(&ep->user->epoll_watches);
735

736
	return 0;
Linus Torvalds's avatar
Linus Torvalds committed
737 738
}

739
static void ep_free(struct eventpoll *ep)
Linus Torvalds's avatar
Linus Torvalds committed
740
{
741 742
	struct rb_node *rbp;
	struct epitem *epi;
Linus Torvalds's avatar
Linus Torvalds committed
743

744 745
	/* We need to release all tasks waiting for these file */
	if (waitqueue_active(&ep->poll_wait))
746
		ep_poll_safewake(&ep->poll_wait);
Linus Torvalds's avatar
Linus Torvalds committed
747

748 749 750
	/*
	 * We need to lock this because we could be hit by
	 * eventpoll_release_file() while we're freeing the "struct eventpoll".
751
	 * We do not need to hold "ep->mtx" here because the epoll file
752 753
	 * is on the way to be removed and no one has references to it
	 * anymore. The only hit might come from eventpoll_release_file() but
754
	 * holding "epmutex" is sufficient here.
755 756
	 */
	mutex_lock(&epmutex);
Linus Torvalds's avatar
Linus Torvalds committed
757 758

	/*
759
	 * Walks through the whole tree by unregistering poll callbacks.
Linus Torvalds's avatar
Linus Torvalds committed
760
	 */
761 762 763 764
	for (rbp = rb_first(&ep->rbr); rbp; rbp = rb_next(rbp)) {
		epi = rb_entry(rbp, struct epitem, rbn);

		ep_unregister_pollwait(ep, epi);
765
		cond_resched();
766
	}
Linus Torvalds's avatar
Linus Torvalds committed
767 768

	/*
769 770
	 * Walks through the whole tree by freeing each "struct epitem". At this
	 * point we are sure no poll callbacks will be lingering around, and also by
771
	 * holding "epmutex" we can be sure that no file cleanup code will hit
772
	 * us during this operation. So we can avoid the lock on "ep->lock".
773 774
	 * We do not need to lock ep->mtx, either, we only do it to prevent
	 * a lockdep warning.
Linus Torvalds's avatar
Linus Torvalds committed
775
	 */
776
	mutex_lock(&ep->mtx);
777
	while ((rbp = rb_first(&ep->rbr)) != NULL) {
778 779
		epi = rb_entry(rbp, struct epitem, rbn);
		ep_remove(ep, epi);
780
		cond_resched();
781
	}
782
	mutex_unlock(&ep->mtx);
Linus Torvalds's avatar
Linus Torvalds committed
783

784
	mutex_unlock(&epmutex);
785
	mutex_destroy(&ep->mtx);
786
	free_uid(ep->user);
787
	wakeup_source_unregister(ep->ws);
788
	kfree(ep);
789
}
Linus Torvalds's avatar
Linus Torvalds committed
790

791 792 793
static int ep_eventpoll_release(struct inode *inode, struct file *file)
{
	struct eventpoll *ep = file->private_data;
Linus Torvalds's avatar
Linus Torvalds committed
794

795
	if (ep)
796 797 798
		ep_free(ep);

	return 0;
Linus Torvalds's avatar
Linus Torvalds committed
799 800
}

801 802 803 804 805 806 807
static inline unsigned int ep_item_poll(struct epitem *epi, poll_table *pt)
{
	pt->_key = epi->event.events;

	return epi->ffd.file->f_op->poll(epi->ffd.file, pt) & epi->event.events;
}

808 809
static int ep_read_events_proc(struct eventpoll *ep, struct list_head *head,
			       void *priv)
810 811
{
	struct epitem *epi, *tmp;
812
	poll_table pt;
813

814
	init_poll_funcptr(&pt, NULL);
815

816
	list_for_each_entry_safe(epi, tmp, head, rdllink) {
817
		if (ep_item_poll(epi, &pt))
818
			return POLLIN | POLLRDNORM;
819
		else {
820 821 822 823 824
			/*
			 * Item has been dropped into the ready list by the poll
			 * callback, but it's not actually ready, as far as
			 * caller requested events goes. We can remove it here.
			 */
825
			__pm_relax(ep_wakeup_source(epi));
826
			list_del_init(&epi->rdllink);
827
		}
828 829 830 831 832
	}

	return 0;
}

833 834 835 836 837 838 839 840
static void ep_ptable_queue_proc(struct file *file, wait_queue_head_t *whead,
				 poll_table *pt);

struct readyevents_arg {
	struct eventpoll *ep;
	bool locked;
};

841 842
static int ep_poll_readyevents_proc(void *priv, void *cookie, int call_nests)
{
843 844 845 846
	struct readyevents_arg *arg = priv;

	return ep_scan_ready_list(arg->ep, ep_read_events_proc, NULL,
				  call_nests + 1, arg->locked);
847 848
}

849 850
static unsigned int ep_eventpoll_poll(struct file *file, poll_table *wait)
{
851
	int pollflags;
852
	struct eventpoll *ep = file->private_data;
853 854 855 856 857 858 859 860
	struct readyevents_arg arg;

	/*
	 * During ep_insert() we already hold the ep->mtx for the tfile.
	 * Prevent re-aquisition.
	 */
	arg.locked = wait && (wait->_qproc == ep_ptable_queue_proc);
	arg.ep = ep;
Linus Torvalds's avatar
Linus Torvalds committed
861

862 863 864
	/* Insert inside our poll wait queue */
	poll_wait(file, &ep->poll_wait, wait);

865 866 867 868 869 870 871
	/*
	 * Proceed to find out if wanted events are really available inside
	 * the ready list. This need to be done under ep_call_nested()
	 * supervision, since the call to f_op->poll() done on listed files
	 * could re-enter here.
	 */
	pollflags = ep_call_nested(&poll_readywalk_ncalls, EP_MAX_NESTS,
872
				   ep_poll_readyevents_proc, &arg, ep, current);
873

874
	return pollflags != -1 ? pollflags : 0;
875 876
}

877
#ifdef CONFIG_PROC_FS
878
static void ep_show_fdinfo(struct seq_file *m, struct file *f)
879 880 881 882 883 884 885 886
{
	struct eventpoll *ep = f->private_data;
	struct rb_node *rbp;

	mutex_lock(&ep->mtx);
	for (rbp = rb_first(&ep->rbr); rbp; rbp = rb_next(rbp)) {
		struct epitem *epi = rb_entry(rbp, struct epitem, rbn);

887 888 889 890
		seq_printf(m, "tfd: %8d events: %8x data: %16llx\n",
			   epi->ffd.fd, epi->event.events,
			   (long long)epi->event.data);
		if (seq_has_overflowed(m))
891 892 893 894 895 896
			break;
	}
	mutex_unlock(&ep->mtx);
}
#endif

897 898
/* File callbacks that implement the eventpoll file behaviour */
static const struct file_operations eventpoll_fops = {
899 900 901
#ifdef CONFIG_PROC_FS
	.show_fdinfo	= ep_show_fdinfo,
#endif
902
	.release	= ep_eventpoll_release,
903 904
	.poll		= ep_eventpoll_poll,
	.llseek		= noop_llseek,
905 906
};

Davide Libenzi's avatar
Davide Libenzi committed
907
/*
908 909 910
 * This is called from eventpoll_release() to unlink files from the eventpoll
 * interface. We need to have this facility to cleanup correctly files that are
 * closed without being removed from the eventpoll interface.
Davide Libenzi's avatar
Davide Libenzi committed
911
 */
912
void eventpoll_release_file(struct file *file)
Davide Libenzi's avatar
Davide Libenzi committed
913
{
914
	struct eventpoll *ep;
915
	struct epitem *epi, *next;
Davide Libenzi's avatar
Davide Libenzi committed
916 917

	/*
918
	 * We don't want to get "file->f_lock" because it is not
919
	 * necessary. It is not necessary because we're in the "struct file"
920
	 * cleanup path, and this means that no one is using this file anymore.
921
	 * So, for example, epoll_ctl() cannot hit here since if we reach this
922
	 * point, the file counter already went to zero and fget() would fail.
923
	 * The only hit might come from ep_free() but by holding the mutex
924
	 * will correctly serialize the operation. We do need to acquire
925
	 * "ep->mtx" after "epmutex" because ep_remove() requires it when called
926
	 * from anywhere but ep_free().
927 928
	 *
	 * Besides, ep_remove() acquires the lock, so we can't hold it here.
Davide Libenzi's avatar
Davide Libenzi committed
929
	 */
930
	mutex_lock(&epmutex);
931
	list_for_each_entry_safe(epi, next, &file->f_ep_links, fllink) {
932
		ep = epi->ep;
933
		mutex_lock_nested(&ep->mtx, 0);
934
		ep_remove(ep, epi);
935
		mutex_unlock(&ep->mtx);
Davide Libenzi's avatar
Davide Libenzi committed
936
	}
937
	mutex_unlock(&epmutex);
Davide Libenzi's avatar
Davide Libenzi committed
938 939
}

940
static int ep_alloc(struct eventpoll **pep)
Linus Torvalds's avatar
Linus Torvalds committed
941
{
942 943 944
	int error;
	struct user_struct *user;
	struct eventpoll *ep;
Linus Torvalds's avatar
Linus Torvalds committed
945

946 947 948 949 950
	user = get_current_user();
	error = -ENOMEM;
	ep = kzalloc(sizeof(*ep), GFP_KERNEL);
	if (unlikely(!ep))
		goto free_uid;
Linus Torvalds's avatar
Linus Torvalds committed
951

952
	spin_lock_init(&ep->lock);
953
	mutex_init(&ep->mtx);
Linus Torvalds's avatar
Linus Torvalds committed
954 955 956 957
	init_waitqueue_head(&ep->wq);
	init_waitqueue_head(&ep->poll_wait);
	INIT_LIST_HEAD(&ep->rdllist);
	ep->rbr = RB_ROOT;
958
	ep->ovflist = EP_UNACTIVE_PTR;
959
	ep->user = user;
Linus Torvalds's avatar
Linus Torvalds committed
960

961
	*pep = ep;
Linus Torvalds's avatar
Linus Torvalds committed
962 963

	return 0;
964 965 966 967

free_uid:
	free_uid(user);
	return error;
Linus Torvalds's avatar
Linus Torvalds committed
968 969 970
}

/*
971 972 973
 * Search the file inside the eventpoll tree. The RB tree operations
 * are protected by the "mtx" mutex, and ep_find() must be called with
 * "mtx" held.
Linus Torvalds's avatar
Linus Torvalds committed
974 975 976 977 978 979 980 981
 */
static struct epitem *ep_find(struct eventpoll *ep, struct file *file, int fd)
{
	int kcmp;
	struct rb_node *rbp;
	struct epitem *epi, *epir = NULL;
	struct epoll_filefd ffd;

982
	ep_set_ffd(&ffd, file, fd);
Linus Torvalds's avatar
Linus Torvalds committed
983 984
	for (rbp = ep->rbr.rb_node; rbp; ) {
		epi = rb_entry(rbp, struct epitem, rbn);
985
		kcmp = ep_cmp_ffd(&ffd, &epi->ffd);
Linus Torvalds's avatar
Linus Torvalds committed
986 987 988 989 990 991 992 993 994 995 996 997 998 999
		if (kcmp > 0)
			rbp = rbp->rb_right;
		else if (kcmp < 0)
			rbp = rbp->rb_left;
		else {
			epir = epi;
			break;
		}
	}

	return epir;
}

/*
1000
 * This is the callback that is passed to the wait queue wakeup
1001
 * mechanism. It is called by the stored file descriptors when they
1002
 * have events to report.
Linus Torvalds's avatar
Linus Torvalds committed
1003
 */
1004
static int ep_poll_callback(wait_queue_t *wait, unsigned mode, int sync, void *key)
Linus Torvalds's avatar
Linus Torvalds committed
1005
{
1006 1007 1008 1009
	int pwake = 0;
	unsigned long flags;
	struct epitem *epi = ep_item_from_wait(wait);
	struct eventpoll *ep = epi->ep;
1010
	int ewake = 0;
Linus Torvalds's avatar
Linus Torvalds committed
1011

1012 1013 1014 1015 1016 1017 1018 1019
	if ((unsigned long)key & POLLFREE) {
		ep_pwq_from_wait(wait)->whead = NULL;
		/*
		 * whead = NULL above can race with ep_remove_wait_queue()
		 * which can do another remove_wait_queue() after us, so we
		 * can't use __remove_wait_queue(). whead->lock is held by
		 * the caller.
		 */
1020
		list_del_init(&wait->task_list);
1021
	}
1022

1023
	spin_lock_irqsave(&ep->lock, flags);
Linus Torvalds's avatar
Linus Torvalds committed
1024

1025 1026 1027 1028 1029 1030 1031
	/*
	 * If the event mask does not contain any poll(2) event, we consider the
	 * descriptor to be disabled. This condition is likely the effect of the
	 * EPOLLONESHOT bit that disables the descriptor when an event is received,
	 * until the next EPOLL_CTL_MOD will be issued.
	 */
	if (!(epi->event.events & ~EP_PRIVATE_BITS))
1032 1033
		goto out_unlock;

1034 1035 1036 1037 1038 1039 1040 1041 1042
	/*
	 * Check the events coming with the callback. At this stage, not
	 * every device reports the events in the "key" parameter of the
	 * callback. We need to be able to handle both cases here, hence the
	 * test for "key" != NULL before the event match test.
	 */
	if (key && !((unsigned long) key & epi->event.events))
		goto out_unlock;

1043
	/*
1044
	 * If we are transferring events to userspace, we can hold no locks
1045
	 * (because we're accessing user memory, and because of linux f_op->poll()
1046
	 * semantics). All the events that happen during that period of time are
1047 1048 1049 1050 1051 1052
	 * chained in ep->ovflist and requeued later on.
	 */
	if (unlikely(ep->ovflist != EP_UNACTIVE_PTR)) {
		if (epi->next == EP_UNACTIVE_PTR) {
			epi->next = ep->ovflist;
			ep->ovflist = epi;
1053 1054 1055 1056 1057 1058 1059 1060
			if (epi->ws) {
				/*
				 * Activate ep->ws since epi->ws may get
				 * deactivated at any time.
				 */
				__pm_stay_awake(ep->ws);
			}

1061 1062 1063
		}
		goto out_unlock;
	}
Linus Torvalds's avatar
Linus Torvalds committed
1064

1065
	/* If this file is already in the ready list we exit soon */
1066
	if (!ep_is_linked(&epi->rdllink)) {
1067
		list_add_tail(&epi->rdllink, &ep->rdllist);
1068
		ep_pm_stay_awake_rcu(epi);
1069
	}
1070 1071 1072 1073 1074

	/*
	 * Wake up ( if active ) both the eventpoll wait list and the ->poll()
	 * wait list.
	 */
1075
	if (waitqueue_active(&ep->wq)) {
1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091
		if ((epi->event.events & EPOLLEXCLUSIVE) &&
					!((unsigned long)key & POLLFREE)) {
			switch ((unsigned long)key & EPOLLINOUT_BITS) {
			case POLLIN:
				if (epi->event.events & POLLIN)
					ewake = 1;
				break;
			case POLLOUT:
				if (epi->event.events & POLLOUT)
					ewake = 1;
				break;
			case 0:
				ewake = 1;
				break;
			}
		}
1092
		wake_up_locked(&ep->wq);
1093
	}
1094 1095 1096
	if (waitqueue_active(&ep->poll_wait))
		pwake++;

1097
out_unlock:
1098
	spin_unlock_irqrestore(&ep->lock, flags);
Linus Torvalds's avatar
Linus Torvalds committed
1099

1100 1101
	/* We have to call this outside the lock */
	if (pwake)
1102
		ep_poll_safewake(&ep->poll_wait);
1103

1104 1105 1106
	if (epi->event.events & EPOLLEXCLUSIVE)
		return ewake;

1107 1108
	return 1;
}
Linus Torvalds's avatar
Linus Torvalds committed
1109 1110 1111 1112 1113 1114 1115 1116

/*
 * This is the callback that is used to add our wait queue to the
 * target file wakeup lists.
 */
static void ep_ptable_queue_proc(struct file *file, wait_queue_head_t *whead,
				 poll_table *pt)
{
1117
	struct epitem *epi = ep_item_from_epqueue(pt);
Linus Torvalds's avatar
Linus Torvalds committed
1118 1119
	struct eppoll_entry *pwq;

1120
	if (epi->nwait >= 0 && (pwq = kmem_cache_alloc(pwq_cache, GFP_KERNEL))) {
Linus Torvalds's avatar
Linus Torvalds committed
1121 1122 1123
		init_waitqueue_func_entry(&pwq->wait, ep_poll_callback);
		pwq->whead = whead;
		pwq->base = epi;
1124 1125 1126 1127
		if (epi->event.events & EPOLLEXCLUSIVE)
			add_wait_queue_exclusive(whead, &pwq->wait);
		else
			add_wait_queue(whead, &pwq->wait);
Linus Torvalds's avatar
Linus Torvalds committed
1128 1129
		list_add_tail(&pwq->llink, &epi->pwqlist);
		epi->nwait++;
1130
	} else {
Linus Torvalds's avatar
Linus Torvalds committed
1131 1132
		/* We have to signal that an error occurred */
		epi->nwait = -1;
1133
	}
Linus Torvalds's avatar
Linus Torvalds committed
1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144
}

static void ep_rbtree_insert(struct eventpoll *ep, struct epitem *epi)
{
	int kcmp;
	struct rb_node **p = &ep->rbr.rb_node, *parent = NULL;
	struct epitem *epic;

	while (*p) {
		parent = *p;
		epic = rb_entry(parent, struct epitem, rbn);
1145
		kcmp = ep_cmp_ffd(&epi->ffd, &epic->ffd);
Linus Torvalds's avatar
Linus Torvalds committed
1146 1147 1148 1149 1150 1151 1152 1153 1154
		if (kcmp > 0)
			p = &parent->rb_right;
		else
			p = &parent->rb_left;
	}
	rb_link_node(&epi->rbn, parent, p);
	rb_insert_color(&epi->rbn, &ep->rbr);
}

1155 1156


Jason Baron's avatar
Jason Baron committed
1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173
#define PATH_ARR_SIZE 5
/*
 * These are the number paths of length 1 to 5, that we are allowing to emanate
 * from a single file of interest. For example, we allow 1000 paths of length
 * 1, to emanate from each file of interest. This essentially represents the
 * potential wakeup paths, which need to be limited in order to avoid massive
 * uncontrolled wakeup storms. The common use case should be a single ep which
 * is connected to n file sources. In this case each file source has 1 path
 * of length 1. Thus, the numbers below should be more than sufficient. These
 * path limits are enforced during an EPOLL_CTL_ADD operation, since a modify
 * and delete can't add additional paths. Protected by the epmutex.
 */
static const int path_limits[PATH_ARR_SIZE] = { 1000, 500, 100, 50, 10 };
static int path_count[PATH_ARR_SIZE];

static int path_count_inc(int nests)
{
1174 1175 1176 1177
	/* Allow an arbitrary number of depth 1 paths */
	if (nests == 0)
		return 0;

Jason Baron's avatar
Jason Baron committed
1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197
	if (++path_count[nests] > path_limits[nests])
		return -1;
	return 0;
}

static void path_count_init(void)
{
	int i;

	for (i = 0; i < PATH_ARR_SIZE; i++)
		path_count[i] = 0;
}

static int reverse_path_check_proc(void *priv, void *cookie, int call_nests)
{
	int error = 0;
	struct file *file = priv;
	struct file *child_file;
	struct epitem *epi;

1198 1199 1200
	/* CTL_DEL can remove links here, but that can't increase our count */
	rcu_read_lock();
	list_for_each_entry_rcu(epi, &file->f_ep_links, fllink) {
Jason Baron's avatar
Jason Baron committed
1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221
		child_file = epi->ep->file;
		if (is_file_epoll(child_file)) {
			if (list_empty(&child_file->f_ep_links)) {
				if (path_count_inc(call_nests)) {
					error = -1;
					break;
				}
			} else {
				error = ep_call_nested(&poll_loop_ncalls,
							EP_MAX_NESTS,
							reverse_path_check_proc,
							child_file, child_file,
							current);
			}
			if (error != 0)
				break;
		} else {
			printk(KERN_ERR "reverse_path_check_proc: "
				"file is not an ep!\n");
		}
	}
1222
	rcu_read_unlock();
Jason Baron's avatar
Jason Baron committed
1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252
	return error;
}

/**
 * reverse_path_check - The tfile_check_list is list of file *, which have
 *                      links that are proposed to be newly added. We need to
 *                      make sure that those added links don't add too many
 *                      paths such that we will spend all our time waking up
 *                      eventpoll objects.
 *
 * Returns: Returns zero if the proposed links don't create too many paths,
 *	    -1 otherwise.
 */
static int reverse_path_check(void)
{
	int error = 0;
	struct file *current_file;

	/* let's call this for all tfiles */
	list_for_each_entry(current_file, &tfile_check_list, f_tfile_llink) {
		path_count_init();
		error = ep_call_nested(&poll_loop_ncalls, EP_MAX_NESTS,
					reverse_path_check_proc, current_file,
					current_file, current);
		if (error)
			break;
	}
	return error;
}

1253 1254 1255
static int ep_create_wakeup_source(struct epitem *epi)
{
	const char *name;
1256
	struct wakeup_source *ws;
1257 1258 1259 1260 1261 1262 1263 1264

	if (!epi->ep->ws) {
		epi->ep->ws = wakeup_source_register("eventpoll");
		if (!epi->ep->ws)
			return -ENOMEM;
	}

	name = epi->ffd.file->f_path.dentry->d_name.name;
1265 1266 1267
	ws = wakeup_source_register(name);

	if (!ws)
1268
		return -ENOMEM;
1269
	rcu_assign_pointer(epi->ws, ws);
1270 1271 1272 1273

	return 0;
}

1274 1275
/* rare code path, only used when EPOLL_CTL_MOD removes a wakeup source */
static noinline void ep_destroy_wakeup_source(struct epitem *epi)
1276
{
1277 1278
	struct wakeup_source *ws = ep_wakeup_source(epi);

1279
	RCU_INIT_POINTER(epi->ws, NULL);
1280 1281 1282 1283 1284 1285 1286 1287

	/*
	 * wait for ep_pm_stay_awake_rcu to finish, synchronize_rcu is
	 * used internally by wakeup_source_remove, too (called by
	 * wakeup_source_unregister), so we cannot use call_rcu
	 */
	synchronize_rcu();
	wakeup_source_unregister(ws);
1288 1289
}

1290 1291 1292
/*
 * Must be called with "mtx" held.
 */
Linus Torvalds's avatar
Linus Torvalds committed
1293
static int ep_insert(struct eventpoll *ep, struct epoll_event *event,
1294
		     struct file *tfile, int fd, int full_check)
Linus Torvalds's avatar
Linus Torvalds committed
1295 1296 1297
{
	int error, revents, pwake = 0;
	unsigned long flags;
1298
	long user_watches;
Linus Torvalds's avatar
Linus Torvalds committed
1299 1300 1301
	struct epitem *epi;
	struct ep_pqueue epq;

1302 1303
	user_watches = atomic_long_read(&ep->user->epoll_watches);
	if (unlikely(user_watches >= max_user_watches))
1304
		return -ENOSPC;
1305
	if (!(epi = kmem_cache_alloc(epi_cache, GFP_KERNEL)))
1306
		return -ENOMEM;
Linus Torvalds's avatar
Linus Torvalds committed
1307 1308 1309 1310 1311 1312

	/* Item initialization follow here ... */
	INIT_LIST_HEAD(&epi->rdllink);
	INIT_LIST_HEAD(&epi->fllink);
	INIT_LIST_HEAD(&epi->pwqlist);
	epi->ep = ep;
1313
	ep_set_ffd(&epi->ffd, tfile, fd);
Linus Torvalds's avatar
Linus Torvalds committed
1314 1315
	epi->event = *event;
	epi->nwait = 0;
1316
	epi->next = EP_UNACTIVE_PTR;
1317 1318 1319 1320 1321
	if (epi->event.events & EPOLLWAKEUP) {
		error = ep_create_wakeup_source(epi);
		if (error)
			goto error_create_wakeup_source;
	} else {
1322
		RCU_INIT_POINTER(epi->ws, NULL);
1323
	}
Linus Torvalds's avatar
Linus Torvalds committed
1324 1325 1326 1327 1328 1329 1330 1331

	/* Initialize the poll table using the queue callback */
	epq.epi = epi;
	init_poll_funcptr(&epq.pt, ep_ptable_queue_proc);

	/*
	 * Attach the item to the poll hooks and get current event bits.
	 * We can safely use the file* here because its usage count has
1332 1333 1334
	 * been increased by the caller of this function. Note that after
	 * this operation completes, the poll callback can start hitting
	 * the new item.
Linus Torvalds's avatar
Linus Torvalds committed
1335
	 */
1336
	revents = ep_item_poll(epi, &epq.pt);
Linus Torvalds's avatar
Linus Torvalds committed
1337 1338 1339 1340 1341 1342

	/*
	 * We have to check if something went wrong during the poll wait queue
	 * install process. Namely an allocation for a wait queue failed due
	 * high memory pressure.
	 */
1343
	error = -ENOMEM;
Linus Torvalds's avatar
Linus Torvalds committed
1344
	if (epi->nwait < 0)
1345
		goto error_unregister;
Linus Torvalds's avatar
Linus Torvalds committed
1346 1347

	/* Add the current item to the list of active epoll hook for this file */
1348
	spin_lock(&tfile->f_lock);
1349
	list_add_tail_rcu(&epi->fllink, &tfile->f_ep_links);
1350
	spin_unlock(&tfile->f_lock);
Linus Torvalds's avatar
Linus Torvalds committed
1351

1352 1353 1354 1355
	/*
	 * Add the current item to the RB tree. All RB tree operations are
	 * protected by "mtx", and ep_insert() is called with "mtx" held.
	 */
Linus Torvalds's avatar
Linus Torvalds committed
1356 1357
	ep_rbtree_insert(ep, epi);

Jason Baron's avatar
Jason Baron committed
1358 1359
	/* now check if we've created too many backpaths */
	error = -EINVAL;
1360
	if (full_check && reverse_path_check())
Jason Baron's avatar
Jason Baron committed
1361 1362
		goto error_remove_epi;

1363 1364 1365
	/* We have to drop the new item inside our item list to keep track of it */
	spin_lock_irqsave(&ep->lock, flags);

Linus Torvalds's avatar
Linus Torvalds committed
1366
	/* If the file is already "ready" we drop it inside the ready list */
1367
	if ((revents & event->events) && !ep_is_linked(&epi->rdllink)) {
Linus Torvalds's avatar
Linus Torvalds committed
1368
		list_add_tail(&epi->rdllink, &ep->rdllist);
1369
		ep_pm_stay_awake(epi);
Linus Torvalds's avatar
Linus Torvalds committed
1370 1371 1372

		/* Notify waiting tasks that events are available */
		if (waitqueue_active(&ep->wq))
1373
			wake_up_locked(&ep->wq);
Linus Torvalds's avatar
Linus Torvalds committed
1374 1375 1376 1377
		if (waitqueue_active(&ep->poll_wait))
			pwake++;
	}

1378
	spin_unlock_irqrestore(&ep->lock, flags);
Linus Torvalds's avatar
Linus Torvalds committed
1379

1380
	atomic_long_inc(&ep->user->epoll_watches);
1381

Linus Torvalds's avatar
Linus Torvalds committed
1382 1383
	/* We have to call this outside the lock */
	if (pwake)
1384
		ep_poll_safewake(&ep->poll_wait);
Linus Torvalds's avatar
Linus Torvalds committed
1385 1386 1387

	return 0;

Jason Baron's avatar
Jason Baron committed
1388 1389
error_remove_epi:
	spin_lock(&tfile->f_lock);
1390
	list_del_rcu(&epi->fllink);
Jason Baron's avatar
Jason Baron committed
1391 1392 1393 1394
	spin_unlock(&tfile->f_lock);

	rb_erase(&epi->rbn, &ep->rbr);

1395
error_unregister:
Linus Torvalds's avatar
Linus Torvalds committed
1396 1397 1398 1399
	ep_unregister_pollwait(ep, epi);

	/*
	 * We need to do this because an event could have been arrived on some
1400 1401 1402
	 * allocated wait queue. Note that we don't care about the ep->ovflist
	 * list, since that is used/cleaned only inside a section bound by "mtx".
	 * And ep_insert() is called with "mtx" held.
Linus Torvalds's avatar
Linus Torvalds committed
1403
	 */
1404
	spin_lock_irqsave(&ep->lock, flags);
1405
	if (ep_is_linked(&epi->rdllink))
1406
		list_del_init(&epi->rdllink);
1407
	spin_unlock_irqrestore(&ep->lock, flags);