8c39275d55b3af1d0978e3bad68c9c0326239a08
[lttng-tools.git] / src / common / consumer / consumer-timer.c
1 /*
2 * Copyright (C) 2012 - Julien Desfossez <julien.desfossez@efficios.com>
3 * David Goulet <dgoulet@efficios.com>
4 *
5 * This program is free software; you can redistribute it and/or modify it
6 * under the terms of the GNU General Public License, version 2 only, as
7 * published by the Free Software Foundation.
8 *
9 * This program is distributed in the hope that it will be useful, but WITHOUT
10 * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
11 * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
12 * more details.
13 *
14 * You should have received a copy of the GNU General Public License along with
15 * this program; if not, write to the Free Software Foundation, Inc., 51
16 * Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
17 */
18
19 #define _LGPL_SOURCE
20 #include <assert.h>
21 #include <inttypes.h>
22 #include <signal.h>
23
24 #include <bin/lttng-consumerd/health-consumerd.h>
25 #include <common/common.h>
26 #include <common/compat/endian.h>
27 #include <common/kernel-ctl/kernel-ctl.h>
28 #include <common/kernel-consumer/kernel-consumer.h>
29 #include <common/consumer/consumer-stream.h>
30 #include <common/consumer/consumer-timer.h>
31 #include <common/consumer/consumer-testpoint.h>
32 #include <common/ust-consumer/ust-consumer.h>
33
34 static struct timer_signal_data timer_signal = {
35 .tid = 0,
36 .setup_done = 0,
37 .qs_done = 0,
38 .lock = PTHREAD_MUTEX_INITIALIZER,
39 };
40
41 /*
42 * Set custom signal mask to current thread.
43 */
44 static void setmask(sigset_t *mask)
45 {
46 int ret;
47
48 ret = sigemptyset(mask);
49 if (ret) {
50 PERROR("sigemptyset");
51 }
52 ret = sigaddset(mask, LTTNG_CONSUMER_SIG_SWITCH);
53 if (ret) {
54 PERROR("sigaddset switch");
55 }
56 ret = sigaddset(mask, LTTNG_CONSUMER_SIG_TEARDOWN);
57 if (ret) {
58 PERROR("sigaddset teardown");
59 }
60 ret = sigaddset(mask, LTTNG_CONSUMER_SIG_LIVE);
61 if (ret) {
62 PERROR("sigaddset live");
63 }
64 ret = sigaddset(mask, LTTNG_CONSUMER_SIG_EXIT);
65 if (ret) {
66 PERROR("sigaddset exit");
67 }
68 }
69
70 /*
71 * Execute action on a timer switch.
72 *
73 * Beware: metadata_switch_timer() should *never* take a mutex also held
74 * while consumer_timer_switch_stop() is called. It would result in
75 * deadlocks.
76 */
77 static void metadata_switch_timer(struct lttng_consumer_local_data *ctx,
78 int sig, siginfo_t *si, void *uc)
79 {
80 int ret;
81 struct lttng_consumer_channel *channel;
82
83 channel = si->si_value.sival_ptr;
84 assert(channel);
85
86 if (channel->switch_timer_error) {
87 return;
88 }
89
90 DBG("Switch timer for channel %" PRIu64, channel->key);
91 switch (ctx->type) {
92 case LTTNG_CONSUMER32_UST:
93 case LTTNG_CONSUMER64_UST:
94 /*
95 * Locks taken by lttng_ustconsumer_request_metadata():
96 * - metadata_socket_lock
97 * - Calling lttng_ustconsumer_recv_metadata():
98 * - channel->metadata_cache->lock
99 * - Calling consumer_metadata_cache_flushed():
100 * - channel->timer_lock
101 * - channel->metadata_cache->lock
102 *
103 * Ensure that neither consumer_data.lock nor
104 * channel->lock are taken within this function, since
105 * they are held while consumer_timer_switch_stop() is
106 * called.
107 */
108 ret = lttng_ustconsumer_request_metadata(ctx, channel, 1, 1);
109 if (ret < 0) {
110 channel->switch_timer_error = 1;
111 }
112 break;
113 case LTTNG_CONSUMER_KERNEL:
114 case LTTNG_CONSUMER_UNKNOWN:
115 assert(0);
116 break;
117 }
118 }
119
120 static int send_empty_index(struct lttng_consumer_stream *stream, uint64_t ts,
121 uint64_t stream_id)
122 {
123 int ret;
124 struct ctf_packet_index index;
125
126 memset(&index, 0, sizeof(index));
127 index.stream_id = htobe64(stream_id);
128 index.timestamp_end = htobe64(ts);
129 ret = consumer_stream_write_index(stream, &index);
130 if (ret < 0) {
131 goto error;
132 }
133
134 error:
135 return ret;
136 }
137
138 int consumer_flush_kernel_index(struct lttng_consumer_stream *stream)
139 {
140 uint64_t ts, stream_id;
141 int ret;
142
143 ret = kernctl_get_current_timestamp(stream->wait_fd, &ts);
144 if (ret < 0) {
145 ERR("Failed to get the current timestamp");
146 goto end;
147 }
148 ret = kernctl_buffer_flush(stream->wait_fd);
149 if (ret < 0) {
150 ERR("Failed to flush kernel stream");
151 goto end;
152 }
153 ret = kernctl_snapshot(stream->wait_fd);
154 if (ret < 0) {
155 if (ret != -EAGAIN && ret != -ENODATA) {
156 PERROR("live timer kernel snapshot");
157 ret = -1;
158 goto end;
159 }
160 ret = kernctl_get_stream_id(stream->wait_fd, &stream_id);
161 if (ret < 0) {
162 PERROR("kernctl_get_stream_id");
163 goto end;
164 }
165 DBG("Stream %" PRIu64 " empty, sending beacon", stream->key);
166 ret = send_empty_index(stream, ts, stream_id);
167 if (ret < 0) {
168 goto end;
169 }
170 }
171 ret = 0;
172 end:
173 return ret;
174 }
175
176 static int check_kernel_stream(struct lttng_consumer_stream *stream)
177 {
178 int ret;
179
180 /*
181 * While holding the stream mutex, try to take a snapshot, if it
182 * succeeds, it means that data is ready to be sent, just let the data
183 * thread handle that. Otherwise, if the snapshot returns EAGAIN, it
184 * means that there is no data to read after the flush, so we can
185 * safely send the empty index.
186 *
187 * Doing a trylock and checking if waiting on metadata if
188 * trylock fails. Bail out of the stream is indeed waiting for
189 * metadata to be pushed. Busy wait on trylock otherwise.
190 */
191 for (;;) {
192 ret = pthread_mutex_trylock(&stream->lock);
193 switch (ret) {
194 case 0:
195 break; /* We have the lock. */
196 case EBUSY:
197 pthread_mutex_lock(&stream->metadata_timer_lock);
198 if (stream->waiting_on_metadata) {
199 ret = 0;
200 stream->missed_metadata_flush = true;
201 pthread_mutex_unlock(&stream->metadata_timer_lock);
202 goto end; /* Bail out. */
203 }
204 pthread_mutex_unlock(&stream->metadata_timer_lock);
205 /* Try again. */
206 caa_cpu_relax();
207 continue;
208 default:
209 ERR("Unexpected pthread_mutex_trylock error %d", ret);
210 ret = -1;
211 goto end;
212 }
213 break;
214 }
215 ret = consumer_flush_kernel_index(stream);
216 pthread_mutex_unlock(&stream->lock);
217 end:
218 return ret;
219 }
220
221 int consumer_flush_ust_index(struct lttng_consumer_stream *stream)
222 {
223 uint64_t ts, stream_id;
224 int ret;
225
226 ret = cds_lfht_is_node_deleted(&stream->node.node);
227 if (ret) {
228 goto end;
229 }
230
231 ret = lttng_ustconsumer_get_current_timestamp(stream, &ts);
232 if (ret < 0) {
233 ERR("Failed to get the current timestamp");
234 goto end;
235 }
236 lttng_ustconsumer_flush_buffer(stream, 1);
237 ret = lttng_ustconsumer_take_snapshot(stream);
238 if (ret < 0) {
239 if (ret != -EAGAIN) {
240 ERR("Taking UST snapshot");
241 ret = -1;
242 goto end;
243 }
244 ret = lttng_ustconsumer_get_stream_id(stream, &stream_id);
245 if (ret < 0) {
246 PERROR("ustctl_get_stream_id");
247 goto end;
248 }
249 DBG("Stream %" PRIu64 " empty, sending beacon", stream->key);
250 ret = send_empty_index(stream, ts, stream_id);
251 if (ret < 0) {
252 goto end;
253 }
254 }
255 ret = 0;
256 end:
257 return ret;
258 }
259
260 static int check_ust_stream(struct lttng_consumer_stream *stream)
261 {
262 int ret;
263
264 assert(stream);
265 assert(stream->ustream);
266 /*
267 * While holding the stream mutex, try to take a snapshot, if it
268 * succeeds, it means that data is ready to be sent, just let the data
269 * thread handle that. Otherwise, if the snapshot returns EAGAIN, it
270 * means that there is no data to read after the flush, so we can
271 * safely send the empty index.
272 *
273 * Doing a trylock and checking if waiting on metadata if
274 * trylock fails. Bail out of the stream is indeed waiting for
275 * metadata to be pushed. Busy wait on trylock otherwise.
276 */
277 for (;;) {
278 ret = pthread_mutex_trylock(&stream->lock);
279 switch (ret) {
280 case 0:
281 break; /* We have the lock. */
282 case EBUSY:
283 pthread_mutex_lock(&stream->metadata_timer_lock);
284 if (stream->waiting_on_metadata) {
285 ret = 0;
286 stream->missed_metadata_flush = true;
287 pthread_mutex_unlock(&stream->metadata_timer_lock);
288 goto end; /* Bail out. */
289 }
290 pthread_mutex_unlock(&stream->metadata_timer_lock);
291 /* Try again. */
292 caa_cpu_relax();
293 continue;
294 default:
295 ERR("Unexpected pthread_mutex_trylock error %d", ret);
296 ret = -1;
297 goto end;
298 }
299 break;
300 }
301 ret = consumer_flush_ust_index(stream);
302 pthread_mutex_unlock(&stream->lock);
303 end:
304 return ret;
305 }
306
307 /*
308 * Execute action on a live timer
309 */
310 static void live_timer(struct lttng_consumer_local_data *ctx,
311 int sig, siginfo_t *si, void *uc)
312 {
313 int ret;
314 struct lttng_consumer_channel *channel;
315 struct lttng_consumer_stream *stream;
316 struct lttng_ht *ht;
317 struct lttng_ht_iter iter;
318
319 channel = si->si_value.sival_ptr;
320 assert(channel);
321
322 if (channel->switch_timer_error) {
323 goto error;
324 }
325 ht = consumer_data.stream_per_chan_id_ht;
326
327 DBG("Live timer for channel %" PRIu64, channel->key);
328
329 rcu_read_lock();
330 switch (ctx->type) {
331 case LTTNG_CONSUMER32_UST:
332 case LTTNG_CONSUMER64_UST:
333 cds_lfht_for_each_entry_duplicate(ht->ht,
334 ht->hash_fct(&channel->key, lttng_ht_seed),
335 ht->match_fct, &channel->key, &iter.iter,
336 stream, node_channel_id.node) {
337 ret = check_ust_stream(stream);
338 if (ret < 0) {
339 goto error_unlock;
340 }
341 }
342 break;
343 case LTTNG_CONSUMER_KERNEL:
344 cds_lfht_for_each_entry_duplicate(ht->ht,
345 ht->hash_fct(&channel->key, lttng_ht_seed),
346 ht->match_fct, &channel->key, &iter.iter,
347 stream, node_channel_id.node) {
348 ret = check_kernel_stream(stream);
349 if (ret < 0) {
350 goto error_unlock;
351 }
352 }
353 break;
354 case LTTNG_CONSUMER_UNKNOWN:
355 assert(0);
356 break;
357 }
358
359 error_unlock:
360 rcu_read_unlock();
361
362 error:
363 return;
364 }
365
366 static
367 void consumer_timer_signal_thread_qs(unsigned int signr)
368 {
369 sigset_t pending_set;
370 int ret;
371
372 /*
373 * We need to be the only thread interacting with the thread
374 * that manages signals for teardown synchronization.
375 */
376 pthread_mutex_lock(&timer_signal.lock);
377
378 /* Ensure we don't have any signal queued for this channel. */
379 for (;;) {
380 ret = sigemptyset(&pending_set);
381 if (ret == -1) {
382 PERROR("sigemptyset");
383 }
384 ret = sigpending(&pending_set);
385 if (ret == -1) {
386 PERROR("sigpending");
387 }
388 if (!sigismember(&pending_set, signr)) {
389 break;
390 }
391 caa_cpu_relax();
392 }
393
394 /*
395 * From this point, no new signal handler will be fired that would try to
396 * access "chan". However, we still need to wait for any currently
397 * executing handler to complete.
398 */
399 cmm_smp_mb();
400 CMM_STORE_SHARED(timer_signal.qs_done, 0);
401 cmm_smp_mb();
402
403 /*
404 * Kill with LTTNG_CONSUMER_SIG_TEARDOWN, so signal management thread wakes
405 * up.
406 */
407 kill(getpid(), LTTNG_CONSUMER_SIG_TEARDOWN);
408
409 while (!CMM_LOAD_SHARED(timer_signal.qs_done)) {
410 caa_cpu_relax();
411 }
412 cmm_smp_mb();
413
414 pthread_mutex_unlock(&timer_signal.lock);
415 }
416
417 /*
418 * Set the timer for periodical metadata flush.
419 */
420 void consumer_timer_switch_start(struct lttng_consumer_channel *channel,
421 unsigned int switch_timer_interval)
422 {
423 int ret;
424 struct sigevent sev;
425 struct itimerspec its;
426
427 assert(channel);
428 assert(channel->key);
429
430 if (switch_timer_interval == 0) {
431 return;
432 }
433
434 sev.sigev_notify = SIGEV_SIGNAL;
435 sev.sigev_signo = LTTNG_CONSUMER_SIG_SWITCH;
436 sev.sigev_value.sival_ptr = channel;
437 ret = timer_create(CLOCKID, &sev, &channel->switch_timer);
438 if (ret == -1) {
439 PERROR("timer_create");
440 }
441 channel->switch_timer_enabled = 1;
442
443 its.it_value.tv_sec = switch_timer_interval / 1000000;
444 its.it_value.tv_nsec = (switch_timer_interval % 1000000) * 1000;
445 its.it_interval.tv_sec = its.it_value.tv_sec;
446 its.it_interval.tv_nsec = its.it_value.tv_nsec;
447
448 ret = timer_settime(channel->switch_timer, 0, &its, NULL);
449 if (ret == -1) {
450 PERROR("timer_settime");
451 }
452 }
453
454 /*
455 * Stop and delete timer.
456 */
457 void consumer_timer_switch_stop(struct lttng_consumer_channel *channel)
458 {
459 int ret;
460
461 assert(channel);
462
463 ret = timer_delete(channel->switch_timer);
464 if (ret == -1) {
465 PERROR("timer_delete");
466 }
467
468 consumer_timer_signal_thread_qs(LTTNG_CONSUMER_SIG_SWITCH);
469
470 channel->switch_timer = 0;
471 channel->switch_timer_enabled = 0;
472 }
473
474 /*
475 * Set the timer for the live mode.
476 */
477 void consumer_timer_live_start(struct lttng_consumer_channel *channel,
478 int live_timer_interval)
479 {
480 int ret;
481 struct sigevent sev;
482 struct itimerspec its;
483
484 assert(channel);
485 assert(channel->key);
486
487 if (live_timer_interval <= 0) {
488 return;
489 }
490
491 sev.sigev_notify = SIGEV_SIGNAL;
492 sev.sigev_signo = LTTNG_CONSUMER_SIG_LIVE;
493 sev.sigev_value.sival_ptr = channel;
494 ret = timer_create(CLOCKID, &sev, &channel->live_timer);
495 if (ret == -1) {
496 PERROR("timer_create");
497 }
498 channel->live_timer_enabled = 1;
499
500 its.it_value.tv_sec = live_timer_interval / 1000000;
501 its.it_value.tv_nsec = (live_timer_interval % 1000000) * 1000;
502 its.it_interval.tv_sec = its.it_value.tv_sec;
503 its.it_interval.tv_nsec = its.it_value.tv_nsec;
504
505 ret = timer_settime(channel->live_timer, 0, &its, NULL);
506 if (ret == -1) {
507 PERROR("timer_settime");
508 }
509 }
510
511 /*
512 * Stop and delete timer.
513 */
514 void consumer_timer_live_stop(struct lttng_consumer_channel *channel)
515 {
516 int ret;
517
518 assert(channel);
519
520 ret = timer_delete(channel->live_timer);
521 if (ret == -1) {
522 PERROR("timer_delete");
523 }
524
525 consumer_timer_signal_thread_qs(LTTNG_CONSUMER_SIG_LIVE);
526
527 channel->live_timer = 0;
528 channel->live_timer_enabled = 0;
529 }
530
531 /*
532 * Block the RT signals for the entire process. It must be called from the
533 * consumer main before creating the threads
534 */
535 int consumer_signal_init(void)
536 {
537 int ret;
538 sigset_t mask;
539
540 /* Block signal for entire process, so only our thread processes it. */
541 setmask(&mask);
542 ret = pthread_sigmask(SIG_BLOCK, &mask, NULL);
543 if (ret) {
544 errno = ret;
545 PERROR("pthread_sigmask");
546 return -1;
547 }
548 return 0;
549 }
550
551 /*
552 * This thread is the sighandler for signals LTTNG_CONSUMER_SIG_SWITCH,
553 * LTTNG_CONSUMER_SIG_TEARDOWN, LTTNG_CONSUMER_SIG_LIVE, and
554 * LTTNG_CONSUMER_SIG_EXIT.
555 */
556 void *consumer_timer_thread(void *data)
557 {
558 int signr;
559 sigset_t mask;
560 siginfo_t info;
561 struct lttng_consumer_local_data *ctx = data;
562
563 rcu_register_thread();
564
565 health_register(health_consumerd, HEALTH_CONSUMERD_TYPE_METADATA_TIMER);
566
567 if (testpoint(consumerd_thread_metadata_timer)) {
568 goto error_testpoint;
569 }
570
571 health_code_update();
572
573 /* Only self thread will receive signal mask. */
574 setmask(&mask);
575 CMM_STORE_SHARED(timer_signal.tid, pthread_self());
576
577 while (1) {
578 health_code_update();
579
580 health_poll_entry();
581 signr = sigwaitinfo(&mask, &info);
582 health_poll_exit();
583 if (signr == -1) {
584 if (errno != EINTR) {
585 PERROR("sigwaitinfo");
586 }
587 continue;
588 } else if (signr == LTTNG_CONSUMER_SIG_SWITCH) {
589 metadata_switch_timer(ctx, info.si_signo, &info, NULL);
590 } else if (signr == LTTNG_CONSUMER_SIG_TEARDOWN) {
591 cmm_smp_mb();
592 CMM_STORE_SHARED(timer_signal.qs_done, 1);
593 cmm_smp_mb();
594 DBG("Signal timer metadata thread teardown");
595 } else if (signr == LTTNG_CONSUMER_SIG_LIVE) {
596 live_timer(ctx, info.si_signo, &info, NULL);
597 } else if (signr == LTTNG_CONSUMER_SIG_EXIT) {
598 assert(consumer_quit);
599 goto end;
600 } else {
601 ERR("Unexpected signal %d\n", info.si_signo);
602 }
603 }
604
605 error_testpoint:
606 /* Only reached in testpoint error */
607 health_error();
608 end:
609 health_unregister(health_consumerd);
610 rcu_unregister_thread();
611 return NULL;
612 }
This page took 0.043101 seconds and 4 git commands to generate.