summaryrefslogtreecommitdiff
path: root/epoll.c
blob: a0df0d21bf8963957a639139617efdcf5e782be7 (plain)
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
465
466
467
468
469
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
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
/*
 * Copyright 2000-2007 Niels Provos <provos@citi.umich.edu>
 * Copyright 2007-2012 Niels Provos, Nick Mathewson
 *
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions
 * are met:
 * 1. Redistributions of source code must retain the above copyright
 *    notice, this list of conditions and the following disclaimer.
 * 2. Redistributions in binary form must reproduce the above copyright
 *    notice, this list of conditions and the following disclaimer in the
 *    documentation and/or other materials provided with the distribution.
 * 3. The name of the author may not be used to endorse or promote products
 *    derived from this software without specific prior written permission.
 *
 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
 * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
 * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
 * IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
 * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
 * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
 * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
 * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 */
#include "event2/event-config.h"
#include "evconfig-private.h"

#ifdef EVENT__HAVE_EPOLL

#include <stdint.h>
#include <sys/types.h>
#include <sys/resource.h>
#ifdef EVENT__HAVE_SYS_TIME_H
#include <sys/time.h>
#endif
#include <sys/queue.h>
#include <sys/epoll.h>
#include <signal.h>
#include <limits.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <unistd.h>
#include <errno.h>
#ifdef EVENT__HAVE_FCNTL_H
#include <fcntl.h>
#endif
#ifdef EVENT__HAVE_SYS_TIMERFD_H
#include <sys/timerfd.h>
#endif

#include "event-internal.h"
#include "evsignal-internal.h"
#include "event2/thread.h"
#include "evthread-internal.h"
#include "log-internal.h"
#include "evmap-internal.h"
#include "changelist-internal.h"
#include "time-internal.h"

/* Since Linux 2.6.17, epoll is able to report about peer half-closed connection
   using special EPOLLRDHUP flag on a read event.
*/
#if !defined(EPOLLRDHUP)
#define EPOLLRDHUP 0
#define EARLY_CLOSE_IF_HAVE_RDHUP 0
#else
#define EARLY_CLOSE_IF_HAVE_RDHUP EV_FEATURE_EARLY_CLOSE
#endif

#include "epolltable-internal.h"

#if defined(EVENT__HAVE_SYS_TIMERFD_H) &&			  \
	defined(EVENT__HAVE_TIMERFD_CREATE) &&			  \
	defined(HAVE_POSIX_MONOTONIC) && defined(TFD_NONBLOCK) && \
	defined(TFD_CLOEXEC)
/* Note that we only use timerfd if TFD_NONBLOCK and TFD_CLOEXEC are available
   and working.  This means that we can't support it on 2.6.25 (where timerfd
   was introduced) or 2.6.26, since 2.6.27 introduced those flags.
 */
#define USING_TIMERFD
#endif

struct epollop {
	struct epoll_event *events;
	int nevents;
	int epfd;
#ifdef USING_TIMERFD
	int timerfd;
#endif
};

static void *epoll_init(struct event_base *);
static int epoll_dispatch(struct event_base *, struct timeval *);
static void epoll_dealloc(struct event_base *);

static const struct eventop epollops_changelist = {
	"epoll (with changelist)",
	epoll_init,
	event_changelist_add_,
	event_changelist_del_,
	epoll_dispatch,
	epoll_dealloc,
	1, /* need reinit */
	EV_FEATURE_ET|EV_FEATURE_O1| EARLY_CLOSE_IF_HAVE_RDHUP,
	EVENT_CHANGELIST_FDINFO_SIZE
};


static int epoll_nochangelist_add(struct event_base *base, evutil_socket_t fd,
    short old, short events, void *p);
static int epoll_nochangelist_del(struct event_base *base, evutil_socket_t fd,
    short old, short events, void *p);

const struct eventop epollops = {
	"epoll",
	epoll_init,
	epoll_nochangelist_add,
	epoll_nochangelist_del,
	epoll_dispatch,
	epoll_dealloc,
	1, /* need reinit */
	EV_FEATURE_ET|EV_FEATURE_O1|EV_FEATURE_EARLY_CLOSE,
	0
};

#define INITIAL_NEVENT 32
#define MAX_NEVENT 4096

/* On Linux kernels at least up to 2.6.24.4, epoll can't handle timeout
 * values bigger than (LONG_MAX - 999ULL)/HZ.  HZ in the wild can be
 * as big as 1000, and LONG_MAX can be as small as (1<<31)-1, so the
 * largest number of msec we can support here is 2147482.  Let's
 * round that down by 47 seconds.
 */
#define MAX_EPOLL_TIMEOUT_MSEC (35*60*1000)

static void *
epoll_init(struct event_base *base)
{
	int epfd = -1;
	struct epollop *epollop;

#ifdef EVENT__HAVE_EPOLL_CREATE1
	/* First, try the shiny new epoll_create1 interface, if we have it. */
	epfd = epoll_create1(EPOLL_CLOEXEC);
#endif
	if (epfd == -1) {
		/* Initialize the kernel queue using the old interface.  (The
		size field is ignored   since 2.6.8.) */
		if ((epfd = epoll_create(32000)) == -1) {
			if (errno != ENOSYS)
				event_warn("epoll_create");
			return (NULL);
		}
		evutil_make_socket_closeonexec(epfd);
	}

	if (!(epollop = mm_calloc(1, sizeof(struct epollop)))) {
		close(epfd);
		return (NULL);
	}

	epollop->epfd = epfd;

	/* Initialize fields */
	epollop->events = mm_calloc(INITIAL_NEVENT, sizeof(struct epoll_event));
	if (epollop->events == NULL) {
		mm_free(epollop);
		close(epfd);
		return (NULL);
	}
	epollop->nevents = INITIAL_NEVENT;

	if ((base->flags & EVENT_BASE_FLAG_EPOLL_USE_CHANGELIST) != 0 ||
	    ((base->flags & EVENT_BASE_FLAG_IGNORE_ENV) == 0 &&
		evutil_getenv_("EVENT_EPOLL_USE_CHANGELIST") != NULL)) {

		base->evsel = &epollops_changelist;
	}

#ifdef USING_TIMERFD
	/*
	  The epoll interface ordinarily gives us one-millisecond precision,
	  so on Linux it makes perfect sense to use the CLOCK_MONOTONIC_COARSE
	  timer.  But when the user has set the new PRECISE_TIMER flag for an
	  event_base, we can try to use timerfd to give them finer granularity.
	*/
	if ((base->flags & EVENT_BASE_FLAG_PRECISE_TIMER) &&
	    base->monotonic_timer.monotonic_clock == CLOCK_MONOTONIC) {
		int fd;
		fd = epollop->timerfd = timerfd_create(CLOCK_MONOTONIC, TFD_NONBLOCK|TFD_CLOEXEC);
		if (epollop->timerfd >= 0) {
			struct epoll_event epev;
			memset(&epev, 0, sizeof(epev));
			epev.data.fd = epollop->timerfd;
			epev.events = EPOLLIN;
			if (epoll_ctl(epollop->epfd, EPOLL_CTL_ADD, fd, &epev) < 0) {
				event_warn("epoll_ctl(timerfd)");
				close(fd);
				epollop->timerfd = -1;
			}
		} else {
			if (errno != EINVAL && errno != ENOSYS) {
				/* These errors probably mean that we were
				 * compiled with timerfd/TFD_* support, but
				 * we're running on a kernel that lacks those.
				 */
				event_warn("timerfd_create");
			}
			epollop->timerfd = -1;
		}
	} else {
		epollop->timerfd = -1;
	}
#endif

	evsig_init_(base);

	return (epollop);
}

static const char *
change_to_string(int change)
{
	change &= (EV_CHANGE_ADD|EV_CHANGE_DEL);
	if (change == EV_CHANGE_ADD) {
		return "add";
	} else if (change == EV_CHANGE_DEL) {
		return "del";
	} else if (change == 0) {
		return "none";
	} else {
		return "???";
	}
}

static const char *
epoll_op_to_string(int op)
{
	return op == EPOLL_CTL_ADD?"ADD":
	    op == EPOLL_CTL_DEL?"DEL":
	    op == EPOLL_CTL_MOD?"MOD":
	    "???";
}

#define PRINT_CHANGES(op, events, ch, status)  \
	"Epoll %s(%d) on fd %d " status ". "       \
	"Old events were %d; "                     \
	"read change was %d (%s); "                \
	"write change was %d (%s); "               \
	"close change was %d (%s)",                \
	epoll_op_to_string(op),                    \
	events,                                    \
	ch->fd,                                    \
	ch->old_events,                            \
	ch->read_change,                           \
	change_to_string(ch->read_change),         \
	ch->write_change,                          \
	change_to_string(ch->write_change),        \
	ch->close_change,                          \
	change_to_string(ch->close_change)

static int
epoll_apply_one_change(struct event_base *base,
    struct epollop *epollop,
    const struct event_change *ch)
{
	struct epoll_event epev;
	int op, events = 0;
	int idx;

	idx = EPOLL_OP_TABLE_INDEX(ch);
	op = epoll_op_table[idx].op;
	events = epoll_op_table[idx].events;

	if (!events) {
		EVUTIL_ASSERT(op == 0);
		return 0;
	}

	if ((ch->read_change|ch->write_change) & EV_CHANGE_ET)
		events |= EPOLLET;

	memset(&epev, 0, sizeof(epev));
	epev.data.fd = ch->fd;
	epev.events = events;
	if (epoll_ctl(epollop->epfd, op, ch->fd, &epev) == 0) {
		event_debug((PRINT_CHANGES(op, epev.events, ch, "okay")));
		return 0;
	}

	switch (op) {
	case EPOLL_CTL_MOD:
		if (errno == ENOENT) {
			/* If a MOD operation fails with ENOENT, the
			 * fd was probably closed and re-opened.  We
			 * should retry the operation as an ADD.
			 */
			if (epoll_ctl(epollop->epfd, EPOLL_CTL_ADD, ch->fd, &epev) == -1) {
				event_warn("Epoll MOD(%d) on %d retried as ADD; that failed too",
				    (int)epev.events, ch->fd);
				return -1;
			} else {
				event_debug(("Epoll MOD(%d) on %d retried as ADD; succeeded.",
					(int)epev.events,
					ch->fd));
				return 0;
			}
		}
		break;
	case EPOLL_CTL_ADD:
		if (errno == EEXIST) {
			/* If an ADD operation fails with EEXIST,
			 * either the operation was redundant (as with a
			 * precautionary add), or we ran into a fun
			 * kernel bug where using dup*() to duplicate the
			 * same file into the same fd gives you the same epitem
			 * rather than a fresh one.  For the second case,
			 * we must retry with MOD. */
			if (epoll_ctl(epollop->epfd, EPOLL_CTL_MOD, ch->fd, &epev) == -1) {
				event_warn("Epoll ADD(%d) on %d retried as MOD; that failed too",
				    (int)epev.events, ch->fd);
				return -1;
			} else {
				event_debug(("Epoll ADD(%d) on %d retried as MOD; succeeded.",
					(int)epev.events,
					ch->fd));
				return 0;
			}
		}
		break;
	case EPOLL_CTL_DEL:
		if (errno == ENOENT || errno == EBADF || errno == EPERM) {
			/* If a delete fails with one of these errors,
			 * that's fine too: we closed the fd before we
			 * got around to calling epoll_dispatch. */
			event_debug(("Epoll DEL(%d) on fd %d gave %s: DEL was unnecessary.",
				(int)epev.events,
				ch->fd,
				strerror(errno)));
			return 0;
		}
		break;
	default:
		break;
	}

	event_warn(PRINT_CHANGES(op, epev.events, ch, "failed"));
	return -1;
}

static int
epoll_apply_changes(struct event_base *base)
{
	struct event_changelist *changelist = &base->changelist;
	struct epollop *epollop = base->evbase;
	struct event_change *ch;

	int r = 0;
	int i;

	for (i = 0; i < changelist->n_changes; ++i) {
		ch = &changelist->changes[i];
		if (epoll_apply_one_change(base, epollop, ch) < 0)
			r = -1;
	}

	return (r);
}

static int
epoll_nochangelist_add(struct event_base *base, evutil_socket_t fd,
    short old, short events, void *p)
{
	struct event_change ch;
	ch.fd = fd;
	ch.old_events = old;
	ch.read_change = ch.write_change = ch.close_change = 0;
	if (events & EV_WRITE)
		ch.write_change = EV_CHANGE_ADD |
		    (events & EV_ET);
	if (events & EV_READ)
		ch.read_change = EV_CHANGE_ADD |
		    (events & EV_ET);
	if (events & EV_CLOSED)
		ch.close_change = EV_CHANGE_ADD |
		    (events & EV_ET);

	return epoll_apply_one_change(base, base->evbase, &ch);
}

static int
epoll_nochangelist_del(struct event_base *base, evutil_socket_t fd,
    short old, short events, void *p)
{
	struct event_change ch;
	ch.fd = fd;
	ch.old_events = old;
	ch.read_change = ch.write_change = ch.close_change = 0;
	if (events & EV_WRITE)
		ch.write_change = EV_CHANGE_DEL |
		    (events & EV_ET);
	if (events & EV_READ)
		ch.read_change = EV_CHANGE_DEL |
		    (events & EV_ET);
	if (events & EV_CLOSED)
		ch.close_change = EV_CHANGE_DEL |
		    (events & EV_ET);

	return epoll_apply_one_change(base, base->evbase, &ch);
}

static int
epoll_dispatch(struct event_base *base, struct timeval *tv)
{
	struct epollop *epollop = base->evbase;
	struct epoll_event *events = epollop->events;
	int i, res;
	long timeout = -1;

#ifdef USING_TIMERFD
	if (epollop->timerfd >= 0) {
		struct itimerspec is;
		is.it_interval.tv_sec = 0;
		is.it_interval.tv_nsec = 0;
		if (tv == NULL) {
			/* No timeout; disarm the timer. */
			is.it_value.tv_sec = 0;
			is.it_value.tv_nsec = 0;
		} else {
			if (tv->tv_sec == 0 && tv->tv_usec == 0) {
				/* we need to exit immediately; timerfd can't
				 * do that. */
				timeout = 0;
			}
			is.it_value.tv_sec = tv->tv_sec;
			is.it_value.tv_nsec = tv->tv_usec * 1000;
		}
		/* TODO: we could avoid unnecessary syscalls here by only
		   calling timerfd_settime when the top timeout changes, or
		   when we're called with a different timeval.
		*/
		if (timerfd_settime(epollop->timerfd, 0, &is, NULL) < 0) {
			event_warn("timerfd_settime");
		}
	} else
#endif
	if (tv != NULL) {
		timeout = evutil_tv_to_msec_(tv);
		if (timeout < 0 || timeout > MAX_EPOLL_TIMEOUT_MSEC) {
			/* Linux kernels can wait forever if the timeout is
			 * too big; see comment on MAX_EPOLL_TIMEOUT_MSEC. */
			timeout = MAX_EPOLL_TIMEOUT_MSEC;
		}
	}

	epoll_apply_changes(base);
	event_changelist_remove_all_(&base->changelist, base);

	EVBASE_RELEASE_LOCK(base, th_base_lock);

	res = epoll_wait(epollop->epfd, events, epollop->nevents, timeout);

	EVBASE_ACQUIRE_LOCK(base, th_base_lock);

	if (res == -1) {
		if (errno != EINTR) {
			event_warn("epoll_wait");
			return (-1);
		}

		return (0);
	}

	event_debug(("%s: epoll_wait reports %d", __func__, res));
	EVUTIL_ASSERT(res <= epollop->nevents);

	for (i = 0; i < res; i++) {
		int what = events[i].events;
		short ev = 0;
#ifdef USING_TIMERFD
		if (events[i].data.fd == epollop->timerfd)
			continue;
#endif

		if (what & (EPOLLHUP|EPOLLERR)) {
			ev = EV_READ | EV_WRITE;
		} else {
			if (what & EPOLLIN)
				ev |= EV_READ;
			if (what & EPOLLOUT)
				ev |= EV_WRITE;
			if (what & EPOLLRDHUP)
				ev |= EV_CLOSED;
		}

		if (!ev)
			continue;

		evmap_io_active_(base, events[i].data.fd, ev | EV_ET);
	}

	if (res == epollop->nevents && epollop->nevents < MAX_NEVENT) {
		/* We used all of the event space this time.  We should
		   be ready for more events next time. */
		int new_nevents = epollop->nevents * 2;
		struct epoll_event *new_events;

		new_events = mm_realloc(epollop->events,
		    new_nevents * sizeof(struct epoll_event));
		if (new_events) {
			epollop->events = new_events;
			epollop->nevents = new_nevents;
		}
	}

	return (0);
}


static void
epoll_dealloc(struct event_base *base)
{
	struct epollop *epollop = base->evbase;

	evsig_dealloc_(base);
	if (epollop->events)
		mm_free(epollop->events);
	if (epollop->epfd >= 0)
		close(epollop->epfd);
#ifdef USING_TIMERFD
	if (epollop->timerfd >= 0)
		close(epollop->timerfd);
#endif

	memset(epollop, 0, sizeof(struct epollop));
	mm_free(epollop);
}

#endif /* EVENT__HAVE_EPOLL */