Loading...
1#include "builtin.h"
2#include "perf.h"
3
4#include "util/util.h"
5#include "util/evlist.h"
6#include "util/cache.h"
7#include "util/evsel.h"
8#include "util/symbol.h"
9#include "util/thread.h"
10#include "util/header.h"
11#include "util/session.h"
12#include "util/tool.h"
13
14#include "util/parse-options.h"
15#include "util/trace-event.h"
16
17#include "util/debug.h"
18
19#include <sys/prctl.h>
20#include <sys/resource.h>
21
22#include <semaphore.h>
23#include <pthread.h>
24#include <math.h>
25
26#define PR_SET_NAME 15 /* Set process name */
27#define MAX_CPUS 4096
28#define COMM_LEN 20
29#define SYM_LEN 129
30#define MAX_PID 65536
31
32struct sched_atom;
33
34struct task_desc {
35 unsigned long nr;
36 unsigned long pid;
37 char comm[COMM_LEN];
38
39 unsigned long nr_events;
40 unsigned long curr_event;
41 struct sched_atom **atoms;
42
43 pthread_t thread;
44 sem_t sleep_sem;
45
46 sem_t ready_for_work;
47 sem_t work_done_sem;
48
49 u64 cpu_usage;
50};
51
52enum sched_event_type {
53 SCHED_EVENT_RUN,
54 SCHED_EVENT_SLEEP,
55 SCHED_EVENT_WAKEUP,
56 SCHED_EVENT_MIGRATION,
57};
58
59struct sched_atom {
60 enum sched_event_type type;
61 int specific_wait;
62 u64 timestamp;
63 u64 duration;
64 unsigned long nr;
65 sem_t *wait_sem;
66 struct task_desc *wakee;
67};
68
69#define TASK_STATE_TO_CHAR_STR "RSDTtZX"
70
71enum thread_state {
72 THREAD_SLEEPING = 0,
73 THREAD_WAIT_CPU,
74 THREAD_SCHED_IN,
75 THREAD_IGNORE
76};
77
78struct work_atom {
79 struct list_head list;
80 enum thread_state state;
81 u64 sched_out_time;
82 u64 wake_up_time;
83 u64 sched_in_time;
84 u64 runtime;
85};
86
87struct work_atoms {
88 struct list_head work_list;
89 struct thread *thread;
90 struct rb_node node;
91 u64 max_lat;
92 u64 max_lat_at;
93 u64 total_lat;
94 u64 nb_atoms;
95 u64 total_runtime;
96};
97
98typedef int (*sort_fn_t)(struct work_atoms *, struct work_atoms *);
99
100struct perf_sched;
101
102struct trace_sched_handler {
103 int (*switch_event)(struct perf_sched *sched, struct perf_evsel *evsel,
104 struct perf_sample *sample, struct machine *machine);
105
106 int (*runtime_event)(struct perf_sched *sched, struct perf_evsel *evsel,
107 struct perf_sample *sample, struct machine *machine);
108
109 int (*wakeup_event)(struct perf_sched *sched, struct perf_evsel *evsel,
110 struct perf_sample *sample, struct machine *machine);
111
112 /* PERF_RECORD_FORK event, not sched_process_fork tracepoint */
113 int (*fork_event)(struct perf_sched *sched, union perf_event *event,
114 struct machine *machine);
115
116 int (*migrate_task_event)(struct perf_sched *sched,
117 struct perf_evsel *evsel,
118 struct perf_sample *sample,
119 struct machine *machine);
120};
121
122struct perf_sched {
123 struct perf_tool tool;
124 const char *sort_order;
125 unsigned long nr_tasks;
126 struct task_desc *pid_to_task[MAX_PID];
127 struct task_desc **tasks;
128 const struct trace_sched_handler *tp_handler;
129 pthread_mutex_t start_work_mutex;
130 pthread_mutex_t work_done_wait_mutex;
131 int profile_cpu;
132/*
133 * Track the current task - that way we can know whether there's any
134 * weird events, such as a task being switched away that is not current.
135 */
136 int max_cpu;
137 u32 curr_pid[MAX_CPUS];
138 struct thread *curr_thread[MAX_CPUS];
139 char next_shortname1;
140 char next_shortname2;
141 unsigned int replay_repeat;
142 unsigned long nr_run_events;
143 unsigned long nr_sleep_events;
144 unsigned long nr_wakeup_events;
145 unsigned long nr_sleep_corrections;
146 unsigned long nr_run_events_optimized;
147 unsigned long targetless_wakeups;
148 unsigned long multitarget_wakeups;
149 unsigned long nr_runs;
150 unsigned long nr_timestamps;
151 unsigned long nr_unordered_timestamps;
152 unsigned long nr_state_machine_bugs;
153 unsigned long nr_context_switch_bugs;
154 unsigned long nr_events;
155 unsigned long nr_lost_chunks;
156 unsigned long nr_lost_events;
157 u64 run_measurement_overhead;
158 u64 sleep_measurement_overhead;
159 u64 start_time;
160 u64 cpu_usage;
161 u64 runavg_cpu_usage;
162 u64 parent_cpu_usage;
163 u64 runavg_parent_cpu_usage;
164 u64 sum_runtime;
165 u64 sum_fluct;
166 u64 run_avg;
167 u64 all_runtime;
168 u64 all_count;
169 u64 cpu_last_switched[MAX_CPUS];
170 struct rb_root atom_root, sorted_atom_root;
171 struct list_head sort_list, cmp_pid;
172};
173
174static u64 get_nsecs(void)
175{
176 struct timespec ts;
177
178 clock_gettime(CLOCK_MONOTONIC, &ts);
179
180 return ts.tv_sec * 1000000000ULL + ts.tv_nsec;
181}
182
183static void burn_nsecs(struct perf_sched *sched, u64 nsecs)
184{
185 u64 T0 = get_nsecs(), T1;
186
187 do {
188 T1 = get_nsecs();
189 } while (T1 + sched->run_measurement_overhead < T0 + nsecs);
190}
191
192static void sleep_nsecs(u64 nsecs)
193{
194 struct timespec ts;
195
196 ts.tv_nsec = nsecs % 999999999;
197 ts.tv_sec = nsecs / 999999999;
198
199 nanosleep(&ts, NULL);
200}
201
202static void calibrate_run_measurement_overhead(struct perf_sched *sched)
203{
204 u64 T0, T1, delta, min_delta = 1000000000ULL;
205 int i;
206
207 for (i = 0; i < 10; i++) {
208 T0 = get_nsecs();
209 burn_nsecs(sched, 0);
210 T1 = get_nsecs();
211 delta = T1-T0;
212 min_delta = min(min_delta, delta);
213 }
214 sched->run_measurement_overhead = min_delta;
215
216 printf("run measurement overhead: %" PRIu64 " nsecs\n", min_delta);
217}
218
219static void calibrate_sleep_measurement_overhead(struct perf_sched *sched)
220{
221 u64 T0, T1, delta, min_delta = 1000000000ULL;
222 int i;
223
224 for (i = 0; i < 10; i++) {
225 T0 = get_nsecs();
226 sleep_nsecs(10000);
227 T1 = get_nsecs();
228 delta = T1-T0;
229 min_delta = min(min_delta, delta);
230 }
231 min_delta -= 10000;
232 sched->sleep_measurement_overhead = min_delta;
233
234 printf("sleep measurement overhead: %" PRIu64 " nsecs\n", min_delta);
235}
236
237static struct sched_atom *
238get_new_event(struct task_desc *task, u64 timestamp)
239{
240 struct sched_atom *event = zalloc(sizeof(*event));
241 unsigned long idx = task->nr_events;
242 size_t size;
243
244 event->timestamp = timestamp;
245 event->nr = idx;
246
247 task->nr_events++;
248 size = sizeof(struct sched_atom *) * task->nr_events;
249 task->atoms = realloc(task->atoms, size);
250 BUG_ON(!task->atoms);
251
252 task->atoms[idx] = event;
253
254 return event;
255}
256
257static struct sched_atom *last_event(struct task_desc *task)
258{
259 if (!task->nr_events)
260 return NULL;
261
262 return task->atoms[task->nr_events - 1];
263}
264
265static void add_sched_event_run(struct perf_sched *sched, struct task_desc *task,
266 u64 timestamp, u64 duration)
267{
268 struct sched_atom *event, *curr_event = last_event(task);
269
270 /*
271 * optimize an existing RUN event by merging this one
272 * to it:
273 */
274 if (curr_event && curr_event->type == SCHED_EVENT_RUN) {
275 sched->nr_run_events_optimized++;
276 curr_event->duration += duration;
277 return;
278 }
279
280 event = get_new_event(task, timestamp);
281
282 event->type = SCHED_EVENT_RUN;
283 event->duration = duration;
284
285 sched->nr_run_events++;
286}
287
288static void add_sched_event_wakeup(struct perf_sched *sched, struct task_desc *task,
289 u64 timestamp, struct task_desc *wakee)
290{
291 struct sched_atom *event, *wakee_event;
292
293 event = get_new_event(task, timestamp);
294 event->type = SCHED_EVENT_WAKEUP;
295 event->wakee = wakee;
296
297 wakee_event = last_event(wakee);
298 if (!wakee_event || wakee_event->type != SCHED_EVENT_SLEEP) {
299 sched->targetless_wakeups++;
300 return;
301 }
302 if (wakee_event->wait_sem) {
303 sched->multitarget_wakeups++;
304 return;
305 }
306
307 wakee_event->wait_sem = zalloc(sizeof(*wakee_event->wait_sem));
308 sem_init(wakee_event->wait_sem, 0, 0);
309 wakee_event->specific_wait = 1;
310 event->wait_sem = wakee_event->wait_sem;
311
312 sched->nr_wakeup_events++;
313}
314
315static void add_sched_event_sleep(struct perf_sched *sched, struct task_desc *task,
316 u64 timestamp, u64 task_state __maybe_unused)
317{
318 struct sched_atom *event = get_new_event(task, timestamp);
319
320 event->type = SCHED_EVENT_SLEEP;
321
322 sched->nr_sleep_events++;
323}
324
325static struct task_desc *register_pid(struct perf_sched *sched,
326 unsigned long pid, const char *comm)
327{
328 struct task_desc *task;
329
330 BUG_ON(pid >= MAX_PID);
331
332 task = sched->pid_to_task[pid];
333
334 if (task)
335 return task;
336
337 task = zalloc(sizeof(*task));
338 task->pid = pid;
339 task->nr = sched->nr_tasks;
340 strcpy(task->comm, comm);
341 /*
342 * every task starts in sleeping state - this gets ignored
343 * if there's no wakeup pointing to this sleep state:
344 */
345 add_sched_event_sleep(sched, task, 0, 0);
346
347 sched->pid_to_task[pid] = task;
348 sched->nr_tasks++;
349 sched->tasks = realloc(sched->tasks, sched->nr_tasks * sizeof(struct task_task *));
350 BUG_ON(!sched->tasks);
351 sched->tasks[task->nr] = task;
352
353 if (verbose)
354 printf("registered task #%ld, PID %ld (%s)\n", sched->nr_tasks, pid, comm);
355
356 return task;
357}
358
359
360static void print_task_traces(struct perf_sched *sched)
361{
362 struct task_desc *task;
363 unsigned long i;
364
365 for (i = 0; i < sched->nr_tasks; i++) {
366 task = sched->tasks[i];
367 printf("task %6ld (%20s:%10ld), nr_events: %ld\n",
368 task->nr, task->comm, task->pid, task->nr_events);
369 }
370}
371
372static void add_cross_task_wakeups(struct perf_sched *sched)
373{
374 struct task_desc *task1, *task2;
375 unsigned long i, j;
376
377 for (i = 0; i < sched->nr_tasks; i++) {
378 task1 = sched->tasks[i];
379 j = i + 1;
380 if (j == sched->nr_tasks)
381 j = 0;
382 task2 = sched->tasks[j];
383 add_sched_event_wakeup(sched, task1, 0, task2);
384 }
385}
386
387static void perf_sched__process_event(struct perf_sched *sched,
388 struct sched_atom *atom)
389{
390 int ret = 0;
391
392 switch (atom->type) {
393 case SCHED_EVENT_RUN:
394 burn_nsecs(sched, atom->duration);
395 break;
396 case SCHED_EVENT_SLEEP:
397 if (atom->wait_sem)
398 ret = sem_wait(atom->wait_sem);
399 BUG_ON(ret);
400 break;
401 case SCHED_EVENT_WAKEUP:
402 if (atom->wait_sem)
403 ret = sem_post(atom->wait_sem);
404 BUG_ON(ret);
405 break;
406 case SCHED_EVENT_MIGRATION:
407 break;
408 default:
409 BUG_ON(1);
410 }
411}
412
413static u64 get_cpu_usage_nsec_parent(void)
414{
415 struct rusage ru;
416 u64 sum;
417 int err;
418
419 err = getrusage(RUSAGE_SELF, &ru);
420 BUG_ON(err);
421
422 sum = ru.ru_utime.tv_sec*1e9 + ru.ru_utime.tv_usec*1e3;
423 sum += ru.ru_stime.tv_sec*1e9 + ru.ru_stime.tv_usec*1e3;
424
425 return sum;
426}
427
428static int self_open_counters(void)
429{
430 struct perf_event_attr attr;
431 int fd;
432
433 memset(&attr, 0, sizeof(attr));
434
435 attr.type = PERF_TYPE_SOFTWARE;
436 attr.config = PERF_COUNT_SW_TASK_CLOCK;
437
438 fd = sys_perf_event_open(&attr, 0, -1, -1, 0);
439
440 if (fd < 0)
441 pr_err("Error: sys_perf_event_open() syscall returned "
442 "with %d (%s)\n", fd, strerror(errno));
443 return fd;
444}
445
446static u64 get_cpu_usage_nsec_self(int fd)
447{
448 u64 runtime;
449 int ret;
450
451 ret = read(fd, &runtime, sizeof(runtime));
452 BUG_ON(ret != sizeof(runtime));
453
454 return runtime;
455}
456
457struct sched_thread_parms {
458 struct task_desc *task;
459 struct perf_sched *sched;
460};
461
462static void *thread_func(void *ctx)
463{
464 struct sched_thread_parms *parms = ctx;
465 struct task_desc *this_task = parms->task;
466 struct perf_sched *sched = parms->sched;
467 u64 cpu_usage_0, cpu_usage_1;
468 unsigned long i, ret;
469 char comm2[22];
470 int fd;
471
472 zfree(&parms);
473
474 sprintf(comm2, ":%s", this_task->comm);
475 prctl(PR_SET_NAME, comm2);
476 fd = self_open_counters();
477 if (fd < 0)
478 return NULL;
479again:
480 ret = sem_post(&this_task->ready_for_work);
481 BUG_ON(ret);
482 ret = pthread_mutex_lock(&sched->start_work_mutex);
483 BUG_ON(ret);
484 ret = pthread_mutex_unlock(&sched->start_work_mutex);
485 BUG_ON(ret);
486
487 cpu_usage_0 = get_cpu_usage_nsec_self(fd);
488
489 for (i = 0; i < this_task->nr_events; i++) {
490 this_task->curr_event = i;
491 perf_sched__process_event(sched, this_task->atoms[i]);
492 }
493
494 cpu_usage_1 = get_cpu_usage_nsec_self(fd);
495 this_task->cpu_usage = cpu_usage_1 - cpu_usage_0;
496 ret = sem_post(&this_task->work_done_sem);
497 BUG_ON(ret);
498
499 ret = pthread_mutex_lock(&sched->work_done_wait_mutex);
500 BUG_ON(ret);
501 ret = pthread_mutex_unlock(&sched->work_done_wait_mutex);
502 BUG_ON(ret);
503
504 goto again;
505}
506
507static void create_tasks(struct perf_sched *sched)
508{
509 struct task_desc *task;
510 pthread_attr_t attr;
511 unsigned long i;
512 int err;
513
514 err = pthread_attr_init(&attr);
515 BUG_ON(err);
516 err = pthread_attr_setstacksize(&attr,
517 (size_t) max(16 * 1024, PTHREAD_STACK_MIN));
518 BUG_ON(err);
519 err = pthread_mutex_lock(&sched->start_work_mutex);
520 BUG_ON(err);
521 err = pthread_mutex_lock(&sched->work_done_wait_mutex);
522 BUG_ON(err);
523 for (i = 0; i < sched->nr_tasks; i++) {
524 struct sched_thread_parms *parms = malloc(sizeof(*parms));
525 BUG_ON(parms == NULL);
526 parms->task = task = sched->tasks[i];
527 parms->sched = sched;
528 sem_init(&task->sleep_sem, 0, 0);
529 sem_init(&task->ready_for_work, 0, 0);
530 sem_init(&task->work_done_sem, 0, 0);
531 task->curr_event = 0;
532 err = pthread_create(&task->thread, &attr, thread_func, parms);
533 BUG_ON(err);
534 }
535}
536
537static void wait_for_tasks(struct perf_sched *sched)
538{
539 u64 cpu_usage_0, cpu_usage_1;
540 struct task_desc *task;
541 unsigned long i, ret;
542
543 sched->start_time = get_nsecs();
544 sched->cpu_usage = 0;
545 pthread_mutex_unlock(&sched->work_done_wait_mutex);
546
547 for (i = 0; i < sched->nr_tasks; i++) {
548 task = sched->tasks[i];
549 ret = sem_wait(&task->ready_for_work);
550 BUG_ON(ret);
551 sem_init(&task->ready_for_work, 0, 0);
552 }
553 ret = pthread_mutex_lock(&sched->work_done_wait_mutex);
554 BUG_ON(ret);
555
556 cpu_usage_0 = get_cpu_usage_nsec_parent();
557
558 pthread_mutex_unlock(&sched->start_work_mutex);
559
560 for (i = 0; i < sched->nr_tasks; i++) {
561 task = sched->tasks[i];
562 ret = sem_wait(&task->work_done_sem);
563 BUG_ON(ret);
564 sem_init(&task->work_done_sem, 0, 0);
565 sched->cpu_usage += task->cpu_usage;
566 task->cpu_usage = 0;
567 }
568
569 cpu_usage_1 = get_cpu_usage_nsec_parent();
570 if (!sched->runavg_cpu_usage)
571 sched->runavg_cpu_usage = sched->cpu_usage;
572 sched->runavg_cpu_usage = (sched->runavg_cpu_usage * 9 + sched->cpu_usage) / 10;
573
574 sched->parent_cpu_usage = cpu_usage_1 - cpu_usage_0;
575 if (!sched->runavg_parent_cpu_usage)
576 sched->runavg_parent_cpu_usage = sched->parent_cpu_usage;
577 sched->runavg_parent_cpu_usage = (sched->runavg_parent_cpu_usage * 9 +
578 sched->parent_cpu_usage)/10;
579
580 ret = pthread_mutex_lock(&sched->start_work_mutex);
581 BUG_ON(ret);
582
583 for (i = 0; i < sched->nr_tasks; i++) {
584 task = sched->tasks[i];
585 sem_init(&task->sleep_sem, 0, 0);
586 task->curr_event = 0;
587 }
588}
589
590static void run_one_test(struct perf_sched *sched)
591{
592 u64 T0, T1, delta, avg_delta, fluct;
593
594 T0 = get_nsecs();
595 wait_for_tasks(sched);
596 T1 = get_nsecs();
597
598 delta = T1 - T0;
599 sched->sum_runtime += delta;
600 sched->nr_runs++;
601
602 avg_delta = sched->sum_runtime / sched->nr_runs;
603 if (delta < avg_delta)
604 fluct = avg_delta - delta;
605 else
606 fluct = delta - avg_delta;
607 sched->sum_fluct += fluct;
608 if (!sched->run_avg)
609 sched->run_avg = delta;
610 sched->run_avg = (sched->run_avg * 9 + delta) / 10;
611
612 printf("#%-3ld: %0.3f, ", sched->nr_runs, (double)delta / 1000000.0);
613
614 printf("ravg: %0.2f, ", (double)sched->run_avg / 1e6);
615
616 printf("cpu: %0.2f / %0.2f",
617 (double)sched->cpu_usage / 1e6, (double)sched->runavg_cpu_usage / 1e6);
618
619#if 0
620 /*
621 * rusage statistics done by the parent, these are less
622 * accurate than the sched->sum_exec_runtime based statistics:
623 */
624 printf(" [%0.2f / %0.2f]",
625 (double)sched->parent_cpu_usage/1e6,
626 (double)sched->runavg_parent_cpu_usage/1e6);
627#endif
628
629 printf("\n");
630
631 if (sched->nr_sleep_corrections)
632 printf(" (%ld sleep corrections)\n", sched->nr_sleep_corrections);
633 sched->nr_sleep_corrections = 0;
634}
635
636static void test_calibrations(struct perf_sched *sched)
637{
638 u64 T0, T1;
639
640 T0 = get_nsecs();
641 burn_nsecs(sched, 1e6);
642 T1 = get_nsecs();
643
644 printf("the run test took %" PRIu64 " nsecs\n", T1 - T0);
645
646 T0 = get_nsecs();
647 sleep_nsecs(1e6);
648 T1 = get_nsecs();
649
650 printf("the sleep test took %" PRIu64 " nsecs\n", T1 - T0);
651}
652
653static int
654replay_wakeup_event(struct perf_sched *sched,
655 struct perf_evsel *evsel, struct perf_sample *sample,
656 struct machine *machine __maybe_unused)
657{
658 const char *comm = perf_evsel__strval(evsel, sample, "comm");
659 const u32 pid = perf_evsel__intval(evsel, sample, "pid");
660 struct task_desc *waker, *wakee;
661
662 if (verbose) {
663 printf("sched_wakeup event %p\n", evsel);
664
665 printf(" ... pid %d woke up %s/%d\n", sample->tid, comm, pid);
666 }
667
668 waker = register_pid(sched, sample->tid, "<unknown>");
669 wakee = register_pid(sched, pid, comm);
670
671 add_sched_event_wakeup(sched, waker, sample->time, wakee);
672 return 0;
673}
674
675static int replay_switch_event(struct perf_sched *sched,
676 struct perf_evsel *evsel,
677 struct perf_sample *sample,
678 struct machine *machine __maybe_unused)
679{
680 const char *prev_comm = perf_evsel__strval(evsel, sample, "prev_comm"),
681 *next_comm = perf_evsel__strval(evsel, sample, "next_comm");
682 const u32 prev_pid = perf_evsel__intval(evsel, sample, "prev_pid"),
683 next_pid = perf_evsel__intval(evsel, sample, "next_pid");
684 const u64 prev_state = perf_evsel__intval(evsel, sample, "prev_state");
685 struct task_desc *prev, __maybe_unused *next;
686 u64 timestamp0, timestamp = sample->time;
687 int cpu = sample->cpu;
688 s64 delta;
689
690 if (verbose)
691 printf("sched_switch event %p\n", evsel);
692
693 if (cpu >= MAX_CPUS || cpu < 0)
694 return 0;
695
696 timestamp0 = sched->cpu_last_switched[cpu];
697 if (timestamp0)
698 delta = timestamp - timestamp0;
699 else
700 delta = 0;
701
702 if (delta < 0) {
703 pr_err("hm, delta: %" PRIu64 " < 0 ?\n", delta);
704 return -1;
705 }
706
707 pr_debug(" ... switch from %s/%d to %s/%d [ran %" PRIu64 " nsecs]\n",
708 prev_comm, prev_pid, next_comm, next_pid, delta);
709
710 prev = register_pid(sched, prev_pid, prev_comm);
711 next = register_pid(sched, next_pid, next_comm);
712
713 sched->cpu_last_switched[cpu] = timestamp;
714
715 add_sched_event_run(sched, prev, timestamp, delta);
716 add_sched_event_sleep(sched, prev, timestamp, prev_state);
717
718 return 0;
719}
720
721static int replay_fork_event(struct perf_sched *sched,
722 union perf_event *event,
723 struct machine *machine)
724{
725 struct thread *child, *parent;
726
727 child = machine__findnew_thread(machine, event->fork.pid,
728 event->fork.tid);
729 parent = machine__findnew_thread(machine, event->fork.ppid,
730 event->fork.ptid);
731
732 if (child == NULL || parent == NULL) {
733 pr_debug("thread does not exist on fork event: child %p, parent %p\n",
734 child, parent);
735 return 0;
736 }
737
738 if (verbose) {
739 printf("fork event\n");
740 printf("... parent: %s/%d\n", thread__comm_str(parent), parent->tid);
741 printf("... child: %s/%d\n", thread__comm_str(child), child->tid);
742 }
743
744 register_pid(sched, parent->tid, thread__comm_str(parent));
745 register_pid(sched, child->tid, thread__comm_str(child));
746 return 0;
747}
748
749struct sort_dimension {
750 const char *name;
751 sort_fn_t cmp;
752 struct list_head list;
753};
754
755static int
756thread_lat_cmp(struct list_head *list, struct work_atoms *l, struct work_atoms *r)
757{
758 struct sort_dimension *sort;
759 int ret = 0;
760
761 BUG_ON(list_empty(list));
762
763 list_for_each_entry(sort, list, list) {
764 ret = sort->cmp(l, r);
765 if (ret)
766 return ret;
767 }
768
769 return ret;
770}
771
772static struct work_atoms *
773thread_atoms_search(struct rb_root *root, struct thread *thread,
774 struct list_head *sort_list)
775{
776 struct rb_node *node = root->rb_node;
777 struct work_atoms key = { .thread = thread };
778
779 while (node) {
780 struct work_atoms *atoms;
781 int cmp;
782
783 atoms = container_of(node, struct work_atoms, node);
784
785 cmp = thread_lat_cmp(sort_list, &key, atoms);
786 if (cmp > 0)
787 node = node->rb_left;
788 else if (cmp < 0)
789 node = node->rb_right;
790 else {
791 BUG_ON(thread != atoms->thread);
792 return atoms;
793 }
794 }
795 return NULL;
796}
797
798static void
799__thread_latency_insert(struct rb_root *root, struct work_atoms *data,
800 struct list_head *sort_list)
801{
802 struct rb_node **new = &(root->rb_node), *parent = NULL;
803
804 while (*new) {
805 struct work_atoms *this;
806 int cmp;
807
808 this = container_of(*new, struct work_atoms, node);
809 parent = *new;
810
811 cmp = thread_lat_cmp(sort_list, data, this);
812
813 if (cmp > 0)
814 new = &((*new)->rb_left);
815 else
816 new = &((*new)->rb_right);
817 }
818
819 rb_link_node(&data->node, parent, new);
820 rb_insert_color(&data->node, root);
821}
822
823static int thread_atoms_insert(struct perf_sched *sched, struct thread *thread)
824{
825 struct work_atoms *atoms = zalloc(sizeof(*atoms));
826 if (!atoms) {
827 pr_err("No memory at %s\n", __func__);
828 return -1;
829 }
830
831 atoms->thread = thread;
832 INIT_LIST_HEAD(&atoms->work_list);
833 __thread_latency_insert(&sched->atom_root, atoms, &sched->cmp_pid);
834 return 0;
835}
836
837static char sched_out_state(u64 prev_state)
838{
839 const char *str = TASK_STATE_TO_CHAR_STR;
840
841 return str[prev_state];
842}
843
844static int
845add_sched_out_event(struct work_atoms *atoms,
846 char run_state,
847 u64 timestamp)
848{
849 struct work_atom *atom = zalloc(sizeof(*atom));
850 if (!atom) {
851 pr_err("Non memory at %s", __func__);
852 return -1;
853 }
854
855 atom->sched_out_time = timestamp;
856
857 if (run_state == 'R') {
858 atom->state = THREAD_WAIT_CPU;
859 atom->wake_up_time = atom->sched_out_time;
860 }
861
862 list_add_tail(&atom->list, &atoms->work_list);
863 return 0;
864}
865
866static void
867add_runtime_event(struct work_atoms *atoms, u64 delta,
868 u64 timestamp __maybe_unused)
869{
870 struct work_atom *atom;
871
872 BUG_ON(list_empty(&atoms->work_list));
873
874 atom = list_entry(atoms->work_list.prev, struct work_atom, list);
875
876 atom->runtime += delta;
877 atoms->total_runtime += delta;
878}
879
880static void
881add_sched_in_event(struct work_atoms *atoms, u64 timestamp)
882{
883 struct work_atom *atom;
884 u64 delta;
885
886 if (list_empty(&atoms->work_list))
887 return;
888
889 atom = list_entry(atoms->work_list.prev, struct work_atom, list);
890
891 if (atom->state != THREAD_WAIT_CPU)
892 return;
893
894 if (timestamp < atom->wake_up_time) {
895 atom->state = THREAD_IGNORE;
896 return;
897 }
898
899 atom->state = THREAD_SCHED_IN;
900 atom->sched_in_time = timestamp;
901
902 delta = atom->sched_in_time - atom->wake_up_time;
903 atoms->total_lat += delta;
904 if (delta > atoms->max_lat) {
905 atoms->max_lat = delta;
906 atoms->max_lat_at = timestamp;
907 }
908 atoms->nb_atoms++;
909}
910
911static int latency_switch_event(struct perf_sched *sched,
912 struct perf_evsel *evsel,
913 struct perf_sample *sample,
914 struct machine *machine)
915{
916 const u32 prev_pid = perf_evsel__intval(evsel, sample, "prev_pid"),
917 next_pid = perf_evsel__intval(evsel, sample, "next_pid");
918 const u64 prev_state = perf_evsel__intval(evsel, sample, "prev_state");
919 struct work_atoms *out_events, *in_events;
920 struct thread *sched_out, *sched_in;
921 u64 timestamp0, timestamp = sample->time;
922 int cpu = sample->cpu;
923 s64 delta;
924
925 BUG_ON(cpu >= MAX_CPUS || cpu < 0);
926
927 timestamp0 = sched->cpu_last_switched[cpu];
928 sched->cpu_last_switched[cpu] = timestamp;
929 if (timestamp0)
930 delta = timestamp - timestamp0;
931 else
932 delta = 0;
933
934 if (delta < 0) {
935 pr_err("hm, delta: %" PRIu64 " < 0 ?\n", delta);
936 return -1;
937 }
938
939 sched_out = machine__findnew_thread(machine, 0, prev_pid);
940 sched_in = machine__findnew_thread(machine, 0, next_pid);
941
942 out_events = thread_atoms_search(&sched->atom_root, sched_out, &sched->cmp_pid);
943 if (!out_events) {
944 if (thread_atoms_insert(sched, sched_out))
945 return -1;
946 out_events = thread_atoms_search(&sched->atom_root, sched_out, &sched->cmp_pid);
947 if (!out_events) {
948 pr_err("out-event: Internal tree error");
949 return -1;
950 }
951 }
952 if (add_sched_out_event(out_events, sched_out_state(prev_state), timestamp))
953 return -1;
954
955 in_events = thread_atoms_search(&sched->atom_root, sched_in, &sched->cmp_pid);
956 if (!in_events) {
957 if (thread_atoms_insert(sched, sched_in))
958 return -1;
959 in_events = thread_atoms_search(&sched->atom_root, sched_in, &sched->cmp_pid);
960 if (!in_events) {
961 pr_err("in-event: Internal tree error");
962 return -1;
963 }
964 /*
965 * Take came in we have not heard about yet,
966 * add in an initial atom in runnable state:
967 */
968 if (add_sched_out_event(in_events, 'R', timestamp))
969 return -1;
970 }
971 add_sched_in_event(in_events, timestamp);
972
973 return 0;
974}
975
976static int latency_runtime_event(struct perf_sched *sched,
977 struct perf_evsel *evsel,
978 struct perf_sample *sample,
979 struct machine *machine)
980{
981 const u32 pid = perf_evsel__intval(evsel, sample, "pid");
982 const u64 runtime = perf_evsel__intval(evsel, sample, "runtime");
983 struct thread *thread = machine__findnew_thread(machine, 0, pid);
984 struct work_atoms *atoms = thread_atoms_search(&sched->atom_root, thread, &sched->cmp_pid);
985 u64 timestamp = sample->time;
986 int cpu = sample->cpu;
987
988 BUG_ON(cpu >= MAX_CPUS || cpu < 0);
989 if (!atoms) {
990 if (thread_atoms_insert(sched, thread))
991 return -1;
992 atoms = thread_atoms_search(&sched->atom_root, thread, &sched->cmp_pid);
993 if (!atoms) {
994 pr_err("in-event: Internal tree error");
995 return -1;
996 }
997 if (add_sched_out_event(atoms, 'R', timestamp))
998 return -1;
999 }
1000
1001 add_runtime_event(atoms, runtime, timestamp);
1002 return 0;
1003}
1004
1005static int latency_wakeup_event(struct perf_sched *sched,
1006 struct perf_evsel *evsel,
1007 struct perf_sample *sample,
1008 struct machine *machine)
1009{
1010 const u32 pid = perf_evsel__intval(evsel, sample, "pid"),
1011 success = perf_evsel__intval(evsel, sample, "success");
1012 struct work_atoms *atoms;
1013 struct work_atom *atom;
1014 struct thread *wakee;
1015 u64 timestamp = sample->time;
1016
1017 /* Note for later, it may be interesting to observe the failing cases */
1018 if (!success)
1019 return 0;
1020
1021 wakee = machine__findnew_thread(machine, 0, pid);
1022 atoms = thread_atoms_search(&sched->atom_root, wakee, &sched->cmp_pid);
1023 if (!atoms) {
1024 if (thread_atoms_insert(sched, wakee))
1025 return -1;
1026 atoms = thread_atoms_search(&sched->atom_root, wakee, &sched->cmp_pid);
1027 if (!atoms) {
1028 pr_err("wakeup-event: Internal tree error");
1029 return -1;
1030 }
1031 if (add_sched_out_event(atoms, 'S', timestamp))
1032 return -1;
1033 }
1034
1035 BUG_ON(list_empty(&atoms->work_list));
1036
1037 atom = list_entry(atoms->work_list.prev, struct work_atom, list);
1038
1039 /*
1040 * You WILL be missing events if you've recorded only
1041 * one CPU, or are only looking at only one, so don't
1042 * make useless noise.
1043 */
1044 if (sched->profile_cpu == -1 && atom->state != THREAD_SLEEPING)
1045 sched->nr_state_machine_bugs++;
1046
1047 sched->nr_timestamps++;
1048 if (atom->sched_out_time > timestamp) {
1049 sched->nr_unordered_timestamps++;
1050 return 0;
1051 }
1052
1053 atom->state = THREAD_WAIT_CPU;
1054 atom->wake_up_time = timestamp;
1055 return 0;
1056}
1057
1058static int latency_migrate_task_event(struct perf_sched *sched,
1059 struct perf_evsel *evsel,
1060 struct perf_sample *sample,
1061 struct machine *machine)
1062{
1063 const u32 pid = perf_evsel__intval(evsel, sample, "pid");
1064 u64 timestamp = sample->time;
1065 struct work_atoms *atoms;
1066 struct work_atom *atom;
1067 struct thread *migrant;
1068
1069 /*
1070 * Only need to worry about migration when profiling one CPU.
1071 */
1072 if (sched->profile_cpu == -1)
1073 return 0;
1074
1075 migrant = machine__findnew_thread(machine, 0, pid);
1076 atoms = thread_atoms_search(&sched->atom_root, migrant, &sched->cmp_pid);
1077 if (!atoms) {
1078 if (thread_atoms_insert(sched, migrant))
1079 return -1;
1080 register_pid(sched, migrant->tid, thread__comm_str(migrant));
1081 atoms = thread_atoms_search(&sched->atom_root, migrant, &sched->cmp_pid);
1082 if (!atoms) {
1083 pr_err("migration-event: Internal tree error");
1084 return -1;
1085 }
1086 if (add_sched_out_event(atoms, 'R', timestamp))
1087 return -1;
1088 }
1089
1090 BUG_ON(list_empty(&atoms->work_list));
1091
1092 atom = list_entry(atoms->work_list.prev, struct work_atom, list);
1093 atom->sched_in_time = atom->sched_out_time = atom->wake_up_time = timestamp;
1094
1095 sched->nr_timestamps++;
1096
1097 if (atom->sched_out_time > timestamp)
1098 sched->nr_unordered_timestamps++;
1099
1100 return 0;
1101}
1102
1103static void output_lat_thread(struct perf_sched *sched, struct work_atoms *work_list)
1104{
1105 int i;
1106 int ret;
1107 u64 avg;
1108
1109 if (!work_list->nb_atoms)
1110 return;
1111 /*
1112 * Ignore idle threads:
1113 */
1114 if (!strcmp(thread__comm_str(work_list->thread), "swapper"))
1115 return;
1116
1117 sched->all_runtime += work_list->total_runtime;
1118 sched->all_count += work_list->nb_atoms;
1119
1120 ret = printf(" %s:%d ", thread__comm_str(work_list->thread), work_list->thread->tid);
1121
1122 for (i = 0; i < 24 - ret; i++)
1123 printf(" ");
1124
1125 avg = work_list->total_lat / work_list->nb_atoms;
1126
1127 printf("|%11.3f ms |%9" PRIu64 " | avg:%9.3f ms | max:%9.3f ms | max at: %13.6f s\n",
1128 (double)work_list->total_runtime / 1e6,
1129 work_list->nb_atoms, (double)avg / 1e6,
1130 (double)work_list->max_lat / 1e6,
1131 (double)work_list->max_lat_at / 1e9);
1132}
1133
1134static int pid_cmp(struct work_atoms *l, struct work_atoms *r)
1135{
1136 if (l->thread->tid < r->thread->tid)
1137 return -1;
1138 if (l->thread->tid > r->thread->tid)
1139 return 1;
1140
1141 return 0;
1142}
1143
1144static int avg_cmp(struct work_atoms *l, struct work_atoms *r)
1145{
1146 u64 avgl, avgr;
1147
1148 if (!l->nb_atoms)
1149 return -1;
1150
1151 if (!r->nb_atoms)
1152 return 1;
1153
1154 avgl = l->total_lat / l->nb_atoms;
1155 avgr = r->total_lat / r->nb_atoms;
1156
1157 if (avgl < avgr)
1158 return -1;
1159 if (avgl > avgr)
1160 return 1;
1161
1162 return 0;
1163}
1164
1165static int max_cmp(struct work_atoms *l, struct work_atoms *r)
1166{
1167 if (l->max_lat < r->max_lat)
1168 return -1;
1169 if (l->max_lat > r->max_lat)
1170 return 1;
1171
1172 return 0;
1173}
1174
1175static int switch_cmp(struct work_atoms *l, struct work_atoms *r)
1176{
1177 if (l->nb_atoms < r->nb_atoms)
1178 return -1;
1179 if (l->nb_atoms > r->nb_atoms)
1180 return 1;
1181
1182 return 0;
1183}
1184
1185static int runtime_cmp(struct work_atoms *l, struct work_atoms *r)
1186{
1187 if (l->total_runtime < r->total_runtime)
1188 return -1;
1189 if (l->total_runtime > r->total_runtime)
1190 return 1;
1191
1192 return 0;
1193}
1194
1195static int sort_dimension__add(const char *tok, struct list_head *list)
1196{
1197 size_t i;
1198 static struct sort_dimension avg_sort_dimension = {
1199 .name = "avg",
1200 .cmp = avg_cmp,
1201 };
1202 static struct sort_dimension max_sort_dimension = {
1203 .name = "max",
1204 .cmp = max_cmp,
1205 };
1206 static struct sort_dimension pid_sort_dimension = {
1207 .name = "pid",
1208 .cmp = pid_cmp,
1209 };
1210 static struct sort_dimension runtime_sort_dimension = {
1211 .name = "runtime",
1212 .cmp = runtime_cmp,
1213 };
1214 static struct sort_dimension switch_sort_dimension = {
1215 .name = "switch",
1216 .cmp = switch_cmp,
1217 };
1218 struct sort_dimension *available_sorts[] = {
1219 &pid_sort_dimension,
1220 &avg_sort_dimension,
1221 &max_sort_dimension,
1222 &switch_sort_dimension,
1223 &runtime_sort_dimension,
1224 };
1225
1226 for (i = 0; i < ARRAY_SIZE(available_sorts); i++) {
1227 if (!strcmp(available_sorts[i]->name, tok)) {
1228 list_add_tail(&available_sorts[i]->list, list);
1229
1230 return 0;
1231 }
1232 }
1233
1234 return -1;
1235}
1236
1237static void perf_sched__sort_lat(struct perf_sched *sched)
1238{
1239 struct rb_node *node;
1240
1241 for (;;) {
1242 struct work_atoms *data;
1243 node = rb_first(&sched->atom_root);
1244 if (!node)
1245 break;
1246
1247 rb_erase(node, &sched->atom_root);
1248 data = rb_entry(node, struct work_atoms, node);
1249 __thread_latency_insert(&sched->sorted_atom_root, data, &sched->sort_list);
1250 }
1251}
1252
1253static int process_sched_wakeup_event(struct perf_tool *tool,
1254 struct perf_evsel *evsel,
1255 struct perf_sample *sample,
1256 struct machine *machine)
1257{
1258 struct perf_sched *sched = container_of(tool, struct perf_sched, tool);
1259
1260 if (sched->tp_handler->wakeup_event)
1261 return sched->tp_handler->wakeup_event(sched, evsel, sample, machine);
1262
1263 return 0;
1264}
1265
1266static int map_switch_event(struct perf_sched *sched, struct perf_evsel *evsel,
1267 struct perf_sample *sample, struct machine *machine)
1268{
1269 const u32 prev_pid = perf_evsel__intval(evsel, sample, "prev_pid"),
1270 next_pid = perf_evsel__intval(evsel, sample, "next_pid");
1271 struct thread *sched_out __maybe_unused, *sched_in;
1272 int new_shortname;
1273 u64 timestamp0, timestamp = sample->time;
1274 s64 delta;
1275 int cpu, this_cpu = sample->cpu;
1276
1277 BUG_ON(this_cpu >= MAX_CPUS || this_cpu < 0);
1278
1279 if (this_cpu > sched->max_cpu)
1280 sched->max_cpu = this_cpu;
1281
1282 timestamp0 = sched->cpu_last_switched[this_cpu];
1283 sched->cpu_last_switched[this_cpu] = timestamp;
1284 if (timestamp0)
1285 delta = timestamp - timestamp0;
1286 else
1287 delta = 0;
1288
1289 if (delta < 0) {
1290 pr_err("hm, delta: %" PRIu64 " < 0 ?\n", delta);
1291 return -1;
1292 }
1293
1294 sched_out = machine__findnew_thread(machine, 0, prev_pid);
1295 sched_in = machine__findnew_thread(machine, 0, next_pid);
1296
1297 sched->curr_thread[this_cpu] = sched_in;
1298
1299 printf(" ");
1300
1301 new_shortname = 0;
1302 if (!sched_in->shortname[0]) {
1303 sched_in->shortname[0] = sched->next_shortname1;
1304 sched_in->shortname[1] = sched->next_shortname2;
1305
1306 if (sched->next_shortname1 < 'Z') {
1307 sched->next_shortname1++;
1308 } else {
1309 sched->next_shortname1='A';
1310 if (sched->next_shortname2 < '9') {
1311 sched->next_shortname2++;
1312 } else {
1313 sched->next_shortname2='0';
1314 }
1315 }
1316 new_shortname = 1;
1317 }
1318
1319 for (cpu = 0; cpu <= sched->max_cpu; cpu++) {
1320 if (cpu != this_cpu)
1321 printf(" ");
1322 else
1323 printf("*");
1324
1325 if (sched->curr_thread[cpu]) {
1326 if (sched->curr_thread[cpu]->tid)
1327 printf("%2s ", sched->curr_thread[cpu]->shortname);
1328 else
1329 printf(". ");
1330 } else
1331 printf(" ");
1332 }
1333
1334 printf(" %12.6f secs ", (double)timestamp/1e9);
1335 if (new_shortname) {
1336 printf("%s => %s:%d\n",
1337 sched_in->shortname, thread__comm_str(sched_in), sched_in->tid);
1338 } else {
1339 printf("\n");
1340 }
1341
1342 return 0;
1343}
1344
1345static int process_sched_switch_event(struct perf_tool *tool,
1346 struct perf_evsel *evsel,
1347 struct perf_sample *sample,
1348 struct machine *machine)
1349{
1350 struct perf_sched *sched = container_of(tool, struct perf_sched, tool);
1351 int this_cpu = sample->cpu, err = 0;
1352 u32 prev_pid = perf_evsel__intval(evsel, sample, "prev_pid"),
1353 next_pid = perf_evsel__intval(evsel, sample, "next_pid");
1354
1355 if (sched->curr_pid[this_cpu] != (u32)-1) {
1356 /*
1357 * Are we trying to switch away a PID that is
1358 * not current?
1359 */
1360 if (sched->curr_pid[this_cpu] != prev_pid)
1361 sched->nr_context_switch_bugs++;
1362 }
1363
1364 if (sched->tp_handler->switch_event)
1365 err = sched->tp_handler->switch_event(sched, evsel, sample, machine);
1366
1367 sched->curr_pid[this_cpu] = next_pid;
1368 return err;
1369}
1370
1371static int process_sched_runtime_event(struct perf_tool *tool,
1372 struct perf_evsel *evsel,
1373 struct perf_sample *sample,
1374 struct machine *machine)
1375{
1376 struct perf_sched *sched = container_of(tool, struct perf_sched, tool);
1377
1378 if (sched->tp_handler->runtime_event)
1379 return sched->tp_handler->runtime_event(sched, evsel, sample, machine);
1380
1381 return 0;
1382}
1383
1384static int perf_sched__process_fork_event(struct perf_tool *tool,
1385 union perf_event *event,
1386 struct perf_sample *sample,
1387 struct machine *machine)
1388{
1389 struct perf_sched *sched = container_of(tool, struct perf_sched, tool);
1390
1391 /* run the fork event through the perf machineruy */
1392 perf_event__process_fork(tool, event, sample, machine);
1393
1394 /* and then run additional processing needed for this command */
1395 if (sched->tp_handler->fork_event)
1396 return sched->tp_handler->fork_event(sched, event, machine);
1397
1398 return 0;
1399}
1400
1401static int process_sched_migrate_task_event(struct perf_tool *tool,
1402 struct perf_evsel *evsel,
1403 struct perf_sample *sample,
1404 struct machine *machine)
1405{
1406 struct perf_sched *sched = container_of(tool, struct perf_sched, tool);
1407
1408 if (sched->tp_handler->migrate_task_event)
1409 return sched->tp_handler->migrate_task_event(sched, evsel, sample, machine);
1410
1411 return 0;
1412}
1413
1414typedef int (*tracepoint_handler)(struct perf_tool *tool,
1415 struct perf_evsel *evsel,
1416 struct perf_sample *sample,
1417 struct machine *machine);
1418
1419static int perf_sched__process_tracepoint_sample(struct perf_tool *tool __maybe_unused,
1420 union perf_event *event __maybe_unused,
1421 struct perf_sample *sample,
1422 struct perf_evsel *evsel,
1423 struct machine *machine)
1424{
1425 int err = 0;
1426
1427 evsel->hists.stats.total_period += sample->period;
1428 hists__inc_nr_events(&evsel->hists, PERF_RECORD_SAMPLE);
1429
1430 if (evsel->handler != NULL) {
1431 tracepoint_handler f = evsel->handler;
1432 err = f(tool, evsel, sample, machine);
1433 }
1434
1435 return err;
1436}
1437
1438static int perf_sched__read_events(struct perf_sched *sched,
1439 struct perf_session **psession)
1440{
1441 const struct perf_evsel_str_handler handlers[] = {
1442 { "sched:sched_switch", process_sched_switch_event, },
1443 { "sched:sched_stat_runtime", process_sched_runtime_event, },
1444 { "sched:sched_wakeup", process_sched_wakeup_event, },
1445 { "sched:sched_wakeup_new", process_sched_wakeup_event, },
1446 { "sched:sched_migrate_task", process_sched_migrate_task_event, },
1447 };
1448 struct perf_session *session;
1449 struct perf_data_file file = {
1450 .path = input_name,
1451 .mode = PERF_DATA_MODE_READ,
1452 };
1453
1454 session = perf_session__new(&file, false, &sched->tool);
1455 if (session == NULL) {
1456 pr_debug("No Memory for session\n");
1457 return -1;
1458 }
1459
1460 if (perf_session__set_tracepoints_handlers(session, handlers))
1461 goto out_delete;
1462
1463 if (perf_session__has_traces(session, "record -R")) {
1464 int err = perf_session__process_events(session, &sched->tool);
1465 if (err) {
1466 pr_err("Failed to process events, error %d", err);
1467 goto out_delete;
1468 }
1469
1470 sched->nr_events = session->stats.nr_events[0];
1471 sched->nr_lost_events = session->stats.total_lost;
1472 sched->nr_lost_chunks = session->stats.nr_events[PERF_RECORD_LOST];
1473 }
1474
1475 if (psession)
1476 *psession = session;
1477 else
1478 perf_session__delete(session);
1479
1480 return 0;
1481
1482out_delete:
1483 perf_session__delete(session);
1484 return -1;
1485}
1486
1487static void print_bad_events(struct perf_sched *sched)
1488{
1489 if (sched->nr_unordered_timestamps && sched->nr_timestamps) {
1490 printf(" INFO: %.3f%% unordered timestamps (%ld out of %ld)\n",
1491 (double)sched->nr_unordered_timestamps/(double)sched->nr_timestamps*100.0,
1492 sched->nr_unordered_timestamps, sched->nr_timestamps);
1493 }
1494 if (sched->nr_lost_events && sched->nr_events) {
1495 printf(" INFO: %.3f%% lost events (%ld out of %ld, in %ld chunks)\n",
1496 (double)sched->nr_lost_events/(double)sched->nr_events * 100.0,
1497 sched->nr_lost_events, sched->nr_events, sched->nr_lost_chunks);
1498 }
1499 if (sched->nr_state_machine_bugs && sched->nr_timestamps) {
1500 printf(" INFO: %.3f%% state machine bugs (%ld out of %ld)",
1501 (double)sched->nr_state_machine_bugs/(double)sched->nr_timestamps*100.0,
1502 sched->nr_state_machine_bugs, sched->nr_timestamps);
1503 if (sched->nr_lost_events)
1504 printf(" (due to lost events?)");
1505 printf("\n");
1506 }
1507 if (sched->nr_context_switch_bugs && sched->nr_timestamps) {
1508 printf(" INFO: %.3f%% context switch bugs (%ld out of %ld)",
1509 (double)sched->nr_context_switch_bugs/(double)sched->nr_timestamps*100.0,
1510 sched->nr_context_switch_bugs, sched->nr_timestamps);
1511 if (sched->nr_lost_events)
1512 printf(" (due to lost events?)");
1513 printf("\n");
1514 }
1515}
1516
1517static int perf_sched__lat(struct perf_sched *sched)
1518{
1519 struct rb_node *next;
1520 struct perf_session *session;
1521
1522 setup_pager();
1523
1524 /* save session -- references to threads are held in work_list */
1525 if (perf_sched__read_events(sched, &session))
1526 return -1;
1527
1528 perf_sched__sort_lat(sched);
1529
1530 printf("\n -----------------------------------------------------------------------------------------------------------------\n");
1531 printf(" Task | Runtime ms | Switches | Average delay ms | Maximum delay ms | Maximum delay at |\n");
1532 printf(" -----------------------------------------------------------------------------------------------------------------\n");
1533
1534 next = rb_first(&sched->sorted_atom_root);
1535
1536 while (next) {
1537 struct work_atoms *work_list;
1538
1539 work_list = rb_entry(next, struct work_atoms, node);
1540 output_lat_thread(sched, work_list);
1541 next = rb_next(next);
1542 }
1543
1544 printf(" -----------------------------------------------------------------------------------------------------------------\n");
1545 printf(" TOTAL: |%11.3f ms |%9" PRIu64 " |\n",
1546 (double)sched->all_runtime / 1e6, sched->all_count);
1547
1548 printf(" ---------------------------------------------------\n");
1549
1550 print_bad_events(sched);
1551 printf("\n");
1552
1553 perf_session__delete(session);
1554 return 0;
1555}
1556
1557static int perf_sched__map(struct perf_sched *sched)
1558{
1559 sched->max_cpu = sysconf(_SC_NPROCESSORS_CONF);
1560
1561 setup_pager();
1562 if (perf_sched__read_events(sched, NULL))
1563 return -1;
1564 print_bad_events(sched);
1565 return 0;
1566}
1567
1568static int perf_sched__replay(struct perf_sched *sched)
1569{
1570 unsigned long i;
1571
1572 calibrate_run_measurement_overhead(sched);
1573 calibrate_sleep_measurement_overhead(sched);
1574
1575 test_calibrations(sched);
1576
1577 if (perf_sched__read_events(sched, NULL))
1578 return -1;
1579
1580 printf("nr_run_events: %ld\n", sched->nr_run_events);
1581 printf("nr_sleep_events: %ld\n", sched->nr_sleep_events);
1582 printf("nr_wakeup_events: %ld\n", sched->nr_wakeup_events);
1583
1584 if (sched->targetless_wakeups)
1585 printf("target-less wakeups: %ld\n", sched->targetless_wakeups);
1586 if (sched->multitarget_wakeups)
1587 printf("multi-target wakeups: %ld\n", sched->multitarget_wakeups);
1588 if (sched->nr_run_events_optimized)
1589 printf("run atoms optimized: %ld\n",
1590 sched->nr_run_events_optimized);
1591
1592 print_task_traces(sched);
1593 add_cross_task_wakeups(sched);
1594
1595 create_tasks(sched);
1596 printf("------------------------------------------------------------\n");
1597 for (i = 0; i < sched->replay_repeat; i++)
1598 run_one_test(sched);
1599
1600 return 0;
1601}
1602
1603static void setup_sorting(struct perf_sched *sched, const struct option *options,
1604 const char * const usage_msg[])
1605{
1606 char *tmp, *tok, *str = strdup(sched->sort_order);
1607
1608 for (tok = strtok_r(str, ", ", &tmp);
1609 tok; tok = strtok_r(NULL, ", ", &tmp)) {
1610 if (sort_dimension__add(tok, &sched->sort_list) < 0) {
1611 error("Unknown --sort key: `%s'", tok);
1612 usage_with_options(usage_msg, options);
1613 }
1614 }
1615
1616 free(str);
1617
1618 sort_dimension__add("pid", &sched->cmp_pid);
1619}
1620
1621static int __cmd_record(int argc, const char **argv)
1622{
1623 unsigned int rec_argc, i, j;
1624 const char **rec_argv;
1625 const char * const record_args[] = {
1626 "record",
1627 "-a",
1628 "-R",
1629 "-m", "1024",
1630 "-c", "1",
1631 "-e", "sched:sched_switch",
1632 "-e", "sched:sched_stat_wait",
1633 "-e", "sched:sched_stat_sleep",
1634 "-e", "sched:sched_stat_iowait",
1635 "-e", "sched:sched_stat_runtime",
1636 "-e", "sched:sched_process_fork",
1637 "-e", "sched:sched_wakeup",
1638 "-e", "sched:sched_migrate_task",
1639 };
1640
1641 rec_argc = ARRAY_SIZE(record_args) + argc - 1;
1642 rec_argv = calloc(rec_argc + 1, sizeof(char *));
1643
1644 if (rec_argv == NULL)
1645 return -ENOMEM;
1646
1647 for (i = 0; i < ARRAY_SIZE(record_args); i++)
1648 rec_argv[i] = strdup(record_args[i]);
1649
1650 for (j = 1; j < (unsigned int)argc; j++, i++)
1651 rec_argv[i] = argv[j];
1652
1653 BUG_ON(i != rec_argc);
1654
1655 return cmd_record(i, rec_argv, NULL);
1656}
1657
1658int cmd_sched(int argc, const char **argv, const char *prefix __maybe_unused)
1659{
1660 const char default_sort_order[] = "avg, max, switch, runtime";
1661 struct perf_sched sched = {
1662 .tool = {
1663 .sample = perf_sched__process_tracepoint_sample,
1664 .comm = perf_event__process_comm,
1665 .lost = perf_event__process_lost,
1666 .fork = perf_sched__process_fork_event,
1667 .ordered_samples = true,
1668 },
1669 .cmp_pid = LIST_HEAD_INIT(sched.cmp_pid),
1670 .sort_list = LIST_HEAD_INIT(sched.sort_list),
1671 .start_work_mutex = PTHREAD_MUTEX_INITIALIZER,
1672 .work_done_wait_mutex = PTHREAD_MUTEX_INITIALIZER,
1673 .sort_order = default_sort_order,
1674 .replay_repeat = 10,
1675 .profile_cpu = -1,
1676 .next_shortname1 = 'A',
1677 .next_shortname2 = '0',
1678 };
1679 const struct option latency_options[] = {
1680 OPT_STRING('s', "sort", &sched.sort_order, "key[,key2...]",
1681 "sort by key(s): runtime, switch, avg, max"),
1682 OPT_INCR('v', "verbose", &verbose,
1683 "be more verbose (show symbol address, etc)"),
1684 OPT_INTEGER('C', "CPU", &sched.profile_cpu,
1685 "CPU to profile on"),
1686 OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace,
1687 "dump raw trace in ASCII"),
1688 OPT_END()
1689 };
1690 const struct option replay_options[] = {
1691 OPT_UINTEGER('r', "repeat", &sched.replay_repeat,
1692 "repeat the workload replay N times (-1: infinite)"),
1693 OPT_INCR('v', "verbose", &verbose,
1694 "be more verbose (show symbol address, etc)"),
1695 OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace,
1696 "dump raw trace in ASCII"),
1697 OPT_END()
1698 };
1699 const struct option sched_options[] = {
1700 OPT_STRING('i', "input", &input_name, "file",
1701 "input file name"),
1702 OPT_INCR('v', "verbose", &verbose,
1703 "be more verbose (show symbol address, etc)"),
1704 OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace,
1705 "dump raw trace in ASCII"),
1706 OPT_END()
1707 };
1708 const char * const latency_usage[] = {
1709 "perf sched latency [<options>]",
1710 NULL
1711 };
1712 const char * const replay_usage[] = {
1713 "perf sched replay [<options>]",
1714 NULL
1715 };
1716 const char * const sched_usage[] = {
1717 "perf sched [<options>] {record|latency|map|replay|script}",
1718 NULL
1719 };
1720 struct trace_sched_handler lat_ops = {
1721 .wakeup_event = latency_wakeup_event,
1722 .switch_event = latency_switch_event,
1723 .runtime_event = latency_runtime_event,
1724 .migrate_task_event = latency_migrate_task_event,
1725 };
1726 struct trace_sched_handler map_ops = {
1727 .switch_event = map_switch_event,
1728 };
1729 struct trace_sched_handler replay_ops = {
1730 .wakeup_event = replay_wakeup_event,
1731 .switch_event = replay_switch_event,
1732 .fork_event = replay_fork_event,
1733 };
1734 unsigned int i;
1735
1736 for (i = 0; i < ARRAY_SIZE(sched.curr_pid); i++)
1737 sched.curr_pid[i] = -1;
1738
1739 argc = parse_options(argc, argv, sched_options, sched_usage,
1740 PARSE_OPT_STOP_AT_NON_OPTION);
1741 if (!argc)
1742 usage_with_options(sched_usage, sched_options);
1743
1744 /*
1745 * Aliased to 'perf script' for now:
1746 */
1747 if (!strcmp(argv[0], "script"))
1748 return cmd_script(argc, argv, prefix);
1749
1750 symbol__init();
1751 if (!strncmp(argv[0], "rec", 3)) {
1752 return __cmd_record(argc, argv);
1753 } else if (!strncmp(argv[0], "lat", 3)) {
1754 sched.tp_handler = &lat_ops;
1755 if (argc > 1) {
1756 argc = parse_options(argc, argv, latency_options, latency_usage, 0);
1757 if (argc)
1758 usage_with_options(latency_usage, latency_options);
1759 }
1760 setup_sorting(&sched, latency_options, latency_usage);
1761 return perf_sched__lat(&sched);
1762 } else if (!strcmp(argv[0], "map")) {
1763 sched.tp_handler = &map_ops;
1764 setup_sorting(&sched, latency_options, latency_usage);
1765 return perf_sched__map(&sched);
1766 } else if (!strncmp(argv[0], "rep", 3)) {
1767 sched.tp_handler = &replay_ops;
1768 if (argc) {
1769 argc = parse_options(argc, argv, replay_options, replay_usage, 0);
1770 if (argc)
1771 usage_with_options(replay_usage, replay_options);
1772 }
1773 return perf_sched__replay(&sched);
1774 } else {
1775 usage_with_options(sched_usage, sched_options);
1776 }
1777
1778 return 0;
1779}
1#include "builtin.h"
2#include "perf.h"
3
4#include "util/util.h"
5#include "util/evlist.h"
6#include "util/cache.h"
7#include "util/evsel.h"
8#include "util/symbol.h"
9#include "util/thread.h"
10#include "util/header.h"
11#include "util/session.h"
12#include "util/tool.h"
13
14#include "util/parse-options.h"
15#include "util/trace-event.h"
16
17#include "util/debug.h"
18
19#include <sys/prctl.h>
20#include <sys/resource.h>
21
22#include <semaphore.h>
23#include <pthread.h>
24#include <math.h>
25
26static const char *input_name;
27
28static char default_sort_order[] = "avg, max, switch, runtime";
29static const char *sort_order = default_sort_order;
30
31static int profile_cpu = -1;
32
33#define PR_SET_NAME 15 /* Set process name */
34#define MAX_CPUS 4096
35
36static u64 run_measurement_overhead;
37static u64 sleep_measurement_overhead;
38
39#define COMM_LEN 20
40#define SYM_LEN 129
41
42#define MAX_PID 65536
43
44static unsigned long nr_tasks;
45
46struct sched_atom;
47
48struct task_desc {
49 unsigned long nr;
50 unsigned long pid;
51 char comm[COMM_LEN];
52
53 unsigned long nr_events;
54 unsigned long curr_event;
55 struct sched_atom **atoms;
56
57 pthread_t thread;
58 sem_t sleep_sem;
59
60 sem_t ready_for_work;
61 sem_t work_done_sem;
62
63 u64 cpu_usage;
64};
65
66enum sched_event_type {
67 SCHED_EVENT_RUN,
68 SCHED_EVENT_SLEEP,
69 SCHED_EVENT_WAKEUP,
70 SCHED_EVENT_MIGRATION,
71};
72
73struct sched_atom {
74 enum sched_event_type type;
75 int specific_wait;
76 u64 timestamp;
77 u64 duration;
78 unsigned long nr;
79 sem_t *wait_sem;
80 struct task_desc *wakee;
81};
82
83static struct task_desc *pid_to_task[MAX_PID];
84
85static struct task_desc **tasks;
86
87static pthread_mutex_t start_work_mutex = PTHREAD_MUTEX_INITIALIZER;
88static u64 start_time;
89
90static pthread_mutex_t work_done_wait_mutex = PTHREAD_MUTEX_INITIALIZER;
91
92static unsigned long nr_run_events;
93static unsigned long nr_sleep_events;
94static unsigned long nr_wakeup_events;
95
96static unsigned long nr_sleep_corrections;
97static unsigned long nr_run_events_optimized;
98
99static unsigned long targetless_wakeups;
100static unsigned long multitarget_wakeups;
101
102static u64 cpu_usage;
103static u64 runavg_cpu_usage;
104static u64 parent_cpu_usage;
105static u64 runavg_parent_cpu_usage;
106
107static unsigned long nr_runs;
108static u64 sum_runtime;
109static u64 sum_fluct;
110static u64 run_avg;
111
112static unsigned int replay_repeat = 10;
113static unsigned long nr_timestamps;
114static unsigned long nr_unordered_timestamps;
115static unsigned long nr_state_machine_bugs;
116static unsigned long nr_context_switch_bugs;
117static unsigned long nr_events;
118static unsigned long nr_lost_chunks;
119static unsigned long nr_lost_events;
120
121#define TASK_STATE_TO_CHAR_STR "RSDTtZX"
122
123enum thread_state {
124 THREAD_SLEEPING = 0,
125 THREAD_WAIT_CPU,
126 THREAD_SCHED_IN,
127 THREAD_IGNORE
128};
129
130struct work_atom {
131 struct list_head list;
132 enum thread_state state;
133 u64 sched_out_time;
134 u64 wake_up_time;
135 u64 sched_in_time;
136 u64 runtime;
137};
138
139struct work_atoms {
140 struct list_head work_list;
141 struct thread *thread;
142 struct rb_node node;
143 u64 max_lat;
144 u64 max_lat_at;
145 u64 total_lat;
146 u64 nb_atoms;
147 u64 total_runtime;
148};
149
150typedef int (*sort_fn_t)(struct work_atoms *, struct work_atoms *);
151
152static struct rb_root atom_root, sorted_atom_root;
153
154static u64 all_runtime;
155static u64 all_count;
156
157
158static u64 get_nsecs(void)
159{
160 struct timespec ts;
161
162 clock_gettime(CLOCK_MONOTONIC, &ts);
163
164 return ts.tv_sec * 1000000000ULL + ts.tv_nsec;
165}
166
167static void burn_nsecs(u64 nsecs)
168{
169 u64 T0 = get_nsecs(), T1;
170
171 do {
172 T1 = get_nsecs();
173 } while (T1 + run_measurement_overhead < T0 + nsecs);
174}
175
176static void sleep_nsecs(u64 nsecs)
177{
178 struct timespec ts;
179
180 ts.tv_nsec = nsecs % 999999999;
181 ts.tv_sec = nsecs / 999999999;
182
183 nanosleep(&ts, NULL);
184}
185
186static void calibrate_run_measurement_overhead(void)
187{
188 u64 T0, T1, delta, min_delta = 1000000000ULL;
189 int i;
190
191 for (i = 0; i < 10; i++) {
192 T0 = get_nsecs();
193 burn_nsecs(0);
194 T1 = get_nsecs();
195 delta = T1-T0;
196 min_delta = min(min_delta, delta);
197 }
198 run_measurement_overhead = min_delta;
199
200 printf("run measurement overhead: %" PRIu64 " nsecs\n", min_delta);
201}
202
203static void calibrate_sleep_measurement_overhead(void)
204{
205 u64 T0, T1, delta, min_delta = 1000000000ULL;
206 int i;
207
208 for (i = 0; i < 10; i++) {
209 T0 = get_nsecs();
210 sleep_nsecs(10000);
211 T1 = get_nsecs();
212 delta = T1-T0;
213 min_delta = min(min_delta, delta);
214 }
215 min_delta -= 10000;
216 sleep_measurement_overhead = min_delta;
217
218 printf("sleep measurement overhead: %" PRIu64 " nsecs\n", min_delta);
219}
220
221static struct sched_atom *
222get_new_event(struct task_desc *task, u64 timestamp)
223{
224 struct sched_atom *event = zalloc(sizeof(*event));
225 unsigned long idx = task->nr_events;
226 size_t size;
227
228 event->timestamp = timestamp;
229 event->nr = idx;
230
231 task->nr_events++;
232 size = sizeof(struct sched_atom *) * task->nr_events;
233 task->atoms = realloc(task->atoms, size);
234 BUG_ON(!task->atoms);
235
236 task->atoms[idx] = event;
237
238 return event;
239}
240
241static struct sched_atom *last_event(struct task_desc *task)
242{
243 if (!task->nr_events)
244 return NULL;
245
246 return task->atoms[task->nr_events - 1];
247}
248
249static void
250add_sched_event_run(struct task_desc *task, u64 timestamp, u64 duration)
251{
252 struct sched_atom *event, *curr_event = last_event(task);
253
254 /*
255 * optimize an existing RUN event by merging this one
256 * to it:
257 */
258 if (curr_event && curr_event->type == SCHED_EVENT_RUN) {
259 nr_run_events_optimized++;
260 curr_event->duration += duration;
261 return;
262 }
263
264 event = get_new_event(task, timestamp);
265
266 event->type = SCHED_EVENT_RUN;
267 event->duration = duration;
268
269 nr_run_events++;
270}
271
272static void
273add_sched_event_wakeup(struct task_desc *task, u64 timestamp,
274 struct task_desc *wakee)
275{
276 struct sched_atom *event, *wakee_event;
277
278 event = get_new_event(task, timestamp);
279 event->type = SCHED_EVENT_WAKEUP;
280 event->wakee = wakee;
281
282 wakee_event = last_event(wakee);
283 if (!wakee_event || wakee_event->type != SCHED_EVENT_SLEEP) {
284 targetless_wakeups++;
285 return;
286 }
287 if (wakee_event->wait_sem) {
288 multitarget_wakeups++;
289 return;
290 }
291
292 wakee_event->wait_sem = zalloc(sizeof(*wakee_event->wait_sem));
293 sem_init(wakee_event->wait_sem, 0, 0);
294 wakee_event->specific_wait = 1;
295 event->wait_sem = wakee_event->wait_sem;
296
297 nr_wakeup_events++;
298}
299
300static void
301add_sched_event_sleep(struct task_desc *task, u64 timestamp,
302 u64 task_state __used)
303{
304 struct sched_atom *event = get_new_event(task, timestamp);
305
306 event->type = SCHED_EVENT_SLEEP;
307
308 nr_sleep_events++;
309}
310
311static struct task_desc *register_pid(unsigned long pid, const char *comm)
312{
313 struct task_desc *task;
314
315 BUG_ON(pid >= MAX_PID);
316
317 task = pid_to_task[pid];
318
319 if (task)
320 return task;
321
322 task = zalloc(sizeof(*task));
323 task->pid = pid;
324 task->nr = nr_tasks;
325 strcpy(task->comm, comm);
326 /*
327 * every task starts in sleeping state - this gets ignored
328 * if there's no wakeup pointing to this sleep state:
329 */
330 add_sched_event_sleep(task, 0, 0);
331
332 pid_to_task[pid] = task;
333 nr_tasks++;
334 tasks = realloc(tasks, nr_tasks*sizeof(struct task_task *));
335 BUG_ON(!tasks);
336 tasks[task->nr] = task;
337
338 if (verbose)
339 printf("registered task #%ld, PID %ld (%s)\n", nr_tasks, pid, comm);
340
341 return task;
342}
343
344
345static void print_task_traces(void)
346{
347 struct task_desc *task;
348 unsigned long i;
349
350 for (i = 0; i < nr_tasks; i++) {
351 task = tasks[i];
352 printf("task %6ld (%20s:%10ld), nr_events: %ld\n",
353 task->nr, task->comm, task->pid, task->nr_events);
354 }
355}
356
357static void add_cross_task_wakeups(void)
358{
359 struct task_desc *task1, *task2;
360 unsigned long i, j;
361
362 for (i = 0; i < nr_tasks; i++) {
363 task1 = tasks[i];
364 j = i + 1;
365 if (j == nr_tasks)
366 j = 0;
367 task2 = tasks[j];
368 add_sched_event_wakeup(task1, 0, task2);
369 }
370}
371
372static void
373process_sched_event(struct task_desc *this_task __used, struct sched_atom *atom)
374{
375 int ret = 0;
376
377 switch (atom->type) {
378 case SCHED_EVENT_RUN:
379 burn_nsecs(atom->duration);
380 break;
381 case SCHED_EVENT_SLEEP:
382 if (atom->wait_sem)
383 ret = sem_wait(atom->wait_sem);
384 BUG_ON(ret);
385 break;
386 case SCHED_EVENT_WAKEUP:
387 if (atom->wait_sem)
388 ret = sem_post(atom->wait_sem);
389 BUG_ON(ret);
390 break;
391 case SCHED_EVENT_MIGRATION:
392 break;
393 default:
394 BUG_ON(1);
395 }
396}
397
398static u64 get_cpu_usage_nsec_parent(void)
399{
400 struct rusage ru;
401 u64 sum;
402 int err;
403
404 err = getrusage(RUSAGE_SELF, &ru);
405 BUG_ON(err);
406
407 sum = ru.ru_utime.tv_sec*1e9 + ru.ru_utime.tv_usec*1e3;
408 sum += ru.ru_stime.tv_sec*1e9 + ru.ru_stime.tv_usec*1e3;
409
410 return sum;
411}
412
413static int self_open_counters(void)
414{
415 struct perf_event_attr attr;
416 int fd;
417
418 memset(&attr, 0, sizeof(attr));
419
420 attr.type = PERF_TYPE_SOFTWARE;
421 attr.config = PERF_COUNT_SW_TASK_CLOCK;
422
423 fd = sys_perf_event_open(&attr, 0, -1, -1, 0);
424
425 if (fd < 0)
426 die("Error: sys_perf_event_open() syscall returned"
427 "with %d (%s)\n", fd, strerror(errno));
428 return fd;
429}
430
431static u64 get_cpu_usage_nsec_self(int fd)
432{
433 u64 runtime;
434 int ret;
435
436 ret = read(fd, &runtime, sizeof(runtime));
437 BUG_ON(ret != sizeof(runtime));
438
439 return runtime;
440}
441
442static void *thread_func(void *ctx)
443{
444 struct task_desc *this_task = ctx;
445 u64 cpu_usage_0, cpu_usage_1;
446 unsigned long i, ret;
447 char comm2[22];
448 int fd;
449
450 sprintf(comm2, ":%s", this_task->comm);
451 prctl(PR_SET_NAME, comm2);
452 fd = self_open_counters();
453
454again:
455 ret = sem_post(&this_task->ready_for_work);
456 BUG_ON(ret);
457 ret = pthread_mutex_lock(&start_work_mutex);
458 BUG_ON(ret);
459 ret = pthread_mutex_unlock(&start_work_mutex);
460 BUG_ON(ret);
461
462 cpu_usage_0 = get_cpu_usage_nsec_self(fd);
463
464 for (i = 0; i < this_task->nr_events; i++) {
465 this_task->curr_event = i;
466 process_sched_event(this_task, this_task->atoms[i]);
467 }
468
469 cpu_usage_1 = get_cpu_usage_nsec_self(fd);
470 this_task->cpu_usage = cpu_usage_1 - cpu_usage_0;
471 ret = sem_post(&this_task->work_done_sem);
472 BUG_ON(ret);
473
474 ret = pthread_mutex_lock(&work_done_wait_mutex);
475 BUG_ON(ret);
476 ret = pthread_mutex_unlock(&work_done_wait_mutex);
477 BUG_ON(ret);
478
479 goto again;
480}
481
482static void create_tasks(void)
483{
484 struct task_desc *task;
485 pthread_attr_t attr;
486 unsigned long i;
487 int err;
488
489 err = pthread_attr_init(&attr);
490 BUG_ON(err);
491 err = pthread_attr_setstacksize(&attr,
492 (size_t) max(16 * 1024, PTHREAD_STACK_MIN));
493 BUG_ON(err);
494 err = pthread_mutex_lock(&start_work_mutex);
495 BUG_ON(err);
496 err = pthread_mutex_lock(&work_done_wait_mutex);
497 BUG_ON(err);
498 for (i = 0; i < nr_tasks; i++) {
499 task = tasks[i];
500 sem_init(&task->sleep_sem, 0, 0);
501 sem_init(&task->ready_for_work, 0, 0);
502 sem_init(&task->work_done_sem, 0, 0);
503 task->curr_event = 0;
504 err = pthread_create(&task->thread, &attr, thread_func, task);
505 BUG_ON(err);
506 }
507}
508
509static void wait_for_tasks(void)
510{
511 u64 cpu_usage_0, cpu_usage_1;
512 struct task_desc *task;
513 unsigned long i, ret;
514
515 start_time = get_nsecs();
516 cpu_usage = 0;
517 pthread_mutex_unlock(&work_done_wait_mutex);
518
519 for (i = 0; i < nr_tasks; i++) {
520 task = tasks[i];
521 ret = sem_wait(&task->ready_for_work);
522 BUG_ON(ret);
523 sem_init(&task->ready_for_work, 0, 0);
524 }
525 ret = pthread_mutex_lock(&work_done_wait_mutex);
526 BUG_ON(ret);
527
528 cpu_usage_0 = get_cpu_usage_nsec_parent();
529
530 pthread_mutex_unlock(&start_work_mutex);
531
532 for (i = 0; i < nr_tasks; i++) {
533 task = tasks[i];
534 ret = sem_wait(&task->work_done_sem);
535 BUG_ON(ret);
536 sem_init(&task->work_done_sem, 0, 0);
537 cpu_usage += task->cpu_usage;
538 task->cpu_usage = 0;
539 }
540
541 cpu_usage_1 = get_cpu_usage_nsec_parent();
542 if (!runavg_cpu_usage)
543 runavg_cpu_usage = cpu_usage;
544 runavg_cpu_usage = (runavg_cpu_usage*9 + cpu_usage)/10;
545
546 parent_cpu_usage = cpu_usage_1 - cpu_usage_0;
547 if (!runavg_parent_cpu_usage)
548 runavg_parent_cpu_usage = parent_cpu_usage;
549 runavg_parent_cpu_usage = (runavg_parent_cpu_usage*9 +
550 parent_cpu_usage)/10;
551
552 ret = pthread_mutex_lock(&start_work_mutex);
553 BUG_ON(ret);
554
555 for (i = 0; i < nr_tasks; i++) {
556 task = tasks[i];
557 sem_init(&task->sleep_sem, 0, 0);
558 task->curr_event = 0;
559 }
560}
561
562static void run_one_test(void)
563{
564 u64 T0, T1, delta, avg_delta, fluct;
565
566 T0 = get_nsecs();
567 wait_for_tasks();
568 T1 = get_nsecs();
569
570 delta = T1 - T0;
571 sum_runtime += delta;
572 nr_runs++;
573
574 avg_delta = sum_runtime / nr_runs;
575 if (delta < avg_delta)
576 fluct = avg_delta - delta;
577 else
578 fluct = delta - avg_delta;
579 sum_fluct += fluct;
580 if (!run_avg)
581 run_avg = delta;
582 run_avg = (run_avg*9 + delta)/10;
583
584 printf("#%-3ld: %0.3f, ",
585 nr_runs, (double)delta/1000000.0);
586
587 printf("ravg: %0.2f, ",
588 (double)run_avg/1e6);
589
590 printf("cpu: %0.2f / %0.2f",
591 (double)cpu_usage/1e6, (double)runavg_cpu_usage/1e6);
592
593#if 0
594 /*
595 * rusage statistics done by the parent, these are less
596 * accurate than the sum_exec_runtime based statistics:
597 */
598 printf(" [%0.2f / %0.2f]",
599 (double)parent_cpu_usage/1e6,
600 (double)runavg_parent_cpu_usage/1e6);
601#endif
602
603 printf("\n");
604
605 if (nr_sleep_corrections)
606 printf(" (%ld sleep corrections)\n", nr_sleep_corrections);
607 nr_sleep_corrections = 0;
608}
609
610static void test_calibrations(void)
611{
612 u64 T0, T1;
613
614 T0 = get_nsecs();
615 burn_nsecs(1e6);
616 T1 = get_nsecs();
617
618 printf("the run test took %" PRIu64 " nsecs\n", T1 - T0);
619
620 T0 = get_nsecs();
621 sleep_nsecs(1e6);
622 T1 = get_nsecs();
623
624 printf("the sleep test took %" PRIu64 " nsecs\n", T1 - T0);
625}
626
627#define FILL_FIELD(ptr, field, event, data) \
628 ptr.field = (typeof(ptr.field)) raw_field_value(event, #field, data)
629
630#define FILL_ARRAY(ptr, array, event, data) \
631do { \
632 void *__array = raw_field_ptr(event, #array, data); \
633 memcpy(ptr.array, __array, sizeof(ptr.array)); \
634} while(0)
635
636#define FILL_COMMON_FIELDS(ptr, event, data) \
637do { \
638 FILL_FIELD(ptr, common_type, event, data); \
639 FILL_FIELD(ptr, common_flags, event, data); \
640 FILL_FIELD(ptr, common_preempt_count, event, data); \
641 FILL_FIELD(ptr, common_pid, event, data); \
642 FILL_FIELD(ptr, common_tgid, event, data); \
643} while (0)
644
645
646
647struct trace_switch_event {
648 u32 size;
649
650 u16 common_type;
651 u8 common_flags;
652 u8 common_preempt_count;
653 u32 common_pid;
654 u32 common_tgid;
655
656 char prev_comm[16];
657 u32 prev_pid;
658 u32 prev_prio;
659 u64 prev_state;
660 char next_comm[16];
661 u32 next_pid;
662 u32 next_prio;
663};
664
665struct trace_runtime_event {
666 u32 size;
667
668 u16 common_type;
669 u8 common_flags;
670 u8 common_preempt_count;
671 u32 common_pid;
672 u32 common_tgid;
673
674 char comm[16];
675 u32 pid;
676 u64 runtime;
677 u64 vruntime;
678};
679
680struct trace_wakeup_event {
681 u32 size;
682
683 u16 common_type;
684 u8 common_flags;
685 u8 common_preempt_count;
686 u32 common_pid;
687 u32 common_tgid;
688
689 char comm[16];
690 u32 pid;
691
692 u32 prio;
693 u32 success;
694 u32 cpu;
695};
696
697struct trace_fork_event {
698 u32 size;
699
700 u16 common_type;
701 u8 common_flags;
702 u8 common_preempt_count;
703 u32 common_pid;
704 u32 common_tgid;
705
706 char parent_comm[16];
707 u32 parent_pid;
708 char child_comm[16];
709 u32 child_pid;
710};
711
712struct trace_migrate_task_event {
713 u32 size;
714
715 u16 common_type;
716 u8 common_flags;
717 u8 common_preempt_count;
718 u32 common_pid;
719 u32 common_tgid;
720
721 char comm[16];
722 u32 pid;
723
724 u32 prio;
725 u32 cpu;
726};
727
728struct trace_sched_handler {
729 void (*switch_event)(struct trace_switch_event *,
730 struct machine *,
731 struct event_format *,
732 int cpu,
733 u64 timestamp,
734 struct thread *thread);
735
736 void (*runtime_event)(struct trace_runtime_event *,
737 struct machine *,
738 struct event_format *,
739 int cpu,
740 u64 timestamp,
741 struct thread *thread);
742
743 void (*wakeup_event)(struct trace_wakeup_event *,
744 struct machine *,
745 struct event_format *,
746 int cpu,
747 u64 timestamp,
748 struct thread *thread);
749
750 void (*fork_event)(struct trace_fork_event *,
751 struct event_format *,
752 int cpu,
753 u64 timestamp,
754 struct thread *thread);
755
756 void (*migrate_task_event)(struct trace_migrate_task_event *,
757 struct machine *machine,
758 struct event_format *,
759 int cpu,
760 u64 timestamp,
761 struct thread *thread);
762};
763
764
765static void
766replay_wakeup_event(struct trace_wakeup_event *wakeup_event,
767 struct machine *machine __used,
768 struct event_format *event,
769 int cpu __used,
770 u64 timestamp __used,
771 struct thread *thread __used)
772{
773 struct task_desc *waker, *wakee;
774
775 if (verbose) {
776 printf("sched_wakeup event %p\n", event);
777
778 printf(" ... pid %d woke up %s/%d\n",
779 wakeup_event->common_pid,
780 wakeup_event->comm,
781 wakeup_event->pid);
782 }
783
784 waker = register_pid(wakeup_event->common_pid, "<unknown>");
785 wakee = register_pid(wakeup_event->pid, wakeup_event->comm);
786
787 add_sched_event_wakeup(waker, timestamp, wakee);
788}
789
790static u64 cpu_last_switched[MAX_CPUS];
791
792static void
793replay_switch_event(struct trace_switch_event *switch_event,
794 struct machine *machine __used,
795 struct event_format *event,
796 int cpu,
797 u64 timestamp,
798 struct thread *thread __used)
799{
800 struct task_desc *prev, __used *next;
801 u64 timestamp0;
802 s64 delta;
803
804 if (verbose)
805 printf("sched_switch event %p\n", event);
806
807 if (cpu >= MAX_CPUS || cpu < 0)
808 return;
809
810 timestamp0 = cpu_last_switched[cpu];
811 if (timestamp0)
812 delta = timestamp - timestamp0;
813 else
814 delta = 0;
815
816 if (delta < 0)
817 die("hm, delta: %" PRIu64 " < 0 ?\n", delta);
818
819 if (verbose) {
820 printf(" ... switch from %s/%d to %s/%d [ran %" PRIu64 " nsecs]\n",
821 switch_event->prev_comm, switch_event->prev_pid,
822 switch_event->next_comm, switch_event->next_pid,
823 delta);
824 }
825
826 prev = register_pid(switch_event->prev_pid, switch_event->prev_comm);
827 next = register_pid(switch_event->next_pid, switch_event->next_comm);
828
829 cpu_last_switched[cpu] = timestamp;
830
831 add_sched_event_run(prev, timestamp, delta);
832 add_sched_event_sleep(prev, timestamp, switch_event->prev_state);
833}
834
835
836static void
837replay_fork_event(struct trace_fork_event *fork_event,
838 struct event_format *event,
839 int cpu __used,
840 u64 timestamp __used,
841 struct thread *thread __used)
842{
843 if (verbose) {
844 printf("sched_fork event %p\n", event);
845 printf("... parent: %s/%d\n", fork_event->parent_comm, fork_event->parent_pid);
846 printf("... child: %s/%d\n", fork_event->child_comm, fork_event->child_pid);
847 }
848 register_pid(fork_event->parent_pid, fork_event->parent_comm);
849 register_pid(fork_event->child_pid, fork_event->child_comm);
850}
851
852static struct trace_sched_handler replay_ops = {
853 .wakeup_event = replay_wakeup_event,
854 .switch_event = replay_switch_event,
855 .fork_event = replay_fork_event,
856};
857
858struct sort_dimension {
859 const char *name;
860 sort_fn_t cmp;
861 struct list_head list;
862};
863
864static LIST_HEAD(cmp_pid);
865
866static int
867thread_lat_cmp(struct list_head *list, struct work_atoms *l, struct work_atoms *r)
868{
869 struct sort_dimension *sort;
870 int ret = 0;
871
872 BUG_ON(list_empty(list));
873
874 list_for_each_entry(sort, list, list) {
875 ret = sort->cmp(l, r);
876 if (ret)
877 return ret;
878 }
879
880 return ret;
881}
882
883static struct work_atoms *
884thread_atoms_search(struct rb_root *root, struct thread *thread,
885 struct list_head *sort_list)
886{
887 struct rb_node *node = root->rb_node;
888 struct work_atoms key = { .thread = thread };
889
890 while (node) {
891 struct work_atoms *atoms;
892 int cmp;
893
894 atoms = container_of(node, struct work_atoms, node);
895
896 cmp = thread_lat_cmp(sort_list, &key, atoms);
897 if (cmp > 0)
898 node = node->rb_left;
899 else if (cmp < 0)
900 node = node->rb_right;
901 else {
902 BUG_ON(thread != atoms->thread);
903 return atoms;
904 }
905 }
906 return NULL;
907}
908
909static void
910__thread_latency_insert(struct rb_root *root, struct work_atoms *data,
911 struct list_head *sort_list)
912{
913 struct rb_node **new = &(root->rb_node), *parent = NULL;
914
915 while (*new) {
916 struct work_atoms *this;
917 int cmp;
918
919 this = container_of(*new, struct work_atoms, node);
920 parent = *new;
921
922 cmp = thread_lat_cmp(sort_list, data, this);
923
924 if (cmp > 0)
925 new = &((*new)->rb_left);
926 else
927 new = &((*new)->rb_right);
928 }
929
930 rb_link_node(&data->node, parent, new);
931 rb_insert_color(&data->node, root);
932}
933
934static void thread_atoms_insert(struct thread *thread)
935{
936 struct work_atoms *atoms = zalloc(sizeof(*atoms));
937 if (!atoms)
938 die("No memory");
939
940 atoms->thread = thread;
941 INIT_LIST_HEAD(&atoms->work_list);
942 __thread_latency_insert(&atom_root, atoms, &cmp_pid);
943}
944
945static void
946latency_fork_event(struct trace_fork_event *fork_event __used,
947 struct event_format *event __used,
948 int cpu __used,
949 u64 timestamp __used,
950 struct thread *thread __used)
951{
952 /* should insert the newcomer */
953}
954
955__used
956static char sched_out_state(struct trace_switch_event *switch_event)
957{
958 const char *str = TASK_STATE_TO_CHAR_STR;
959
960 return str[switch_event->prev_state];
961}
962
963static void
964add_sched_out_event(struct work_atoms *atoms,
965 char run_state,
966 u64 timestamp)
967{
968 struct work_atom *atom = zalloc(sizeof(*atom));
969 if (!atom)
970 die("Non memory");
971
972 atom->sched_out_time = timestamp;
973
974 if (run_state == 'R') {
975 atom->state = THREAD_WAIT_CPU;
976 atom->wake_up_time = atom->sched_out_time;
977 }
978
979 list_add_tail(&atom->list, &atoms->work_list);
980}
981
982static void
983add_runtime_event(struct work_atoms *atoms, u64 delta, u64 timestamp __used)
984{
985 struct work_atom *atom;
986
987 BUG_ON(list_empty(&atoms->work_list));
988
989 atom = list_entry(atoms->work_list.prev, struct work_atom, list);
990
991 atom->runtime += delta;
992 atoms->total_runtime += delta;
993}
994
995static void
996add_sched_in_event(struct work_atoms *atoms, u64 timestamp)
997{
998 struct work_atom *atom;
999 u64 delta;
1000
1001 if (list_empty(&atoms->work_list))
1002 return;
1003
1004 atom = list_entry(atoms->work_list.prev, struct work_atom, list);
1005
1006 if (atom->state != THREAD_WAIT_CPU)
1007 return;
1008
1009 if (timestamp < atom->wake_up_time) {
1010 atom->state = THREAD_IGNORE;
1011 return;
1012 }
1013
1014 atom->state = THREAD_SCHED_IN;
1015 atom->sched_in_time = timestamp;
1016
1017 delta = atom->sched_in_time - atom->wake_up_time;
1018 atoms->total_lat += delta;
1019 if (delta > atoms->max_lat) {
1020 atoms->max_lat = delta;
1021 atoms->max_lat_at = timestamp;
1022 }
1023 atoms->nb_atoms++;
1024}
1025
1026static void
1027latency_switch_event(struct trace_switch_event *switch_event,
1028 struct machine *machine,
1029 struct event_format *event __used,
1030 int cpu,
1031 u64 timestamp,
1032 struct thread *thread __used)
1033{
1034 struct work_atoms *out_events, *in_events;
1035 struct thread *sched_out, *sched_in;
1036 u64 timestamp0;
1037 s64 delta;
1038
1039 BUG_ON(cpu >= MAX_CPUS || cpu < 0);
1040
1041 timestamp0 = cpu_last_switched[cpu];
1042 cpu_last_switched[cpu] = timestamp;
1043 if (timestamp0)
1044 delta = timestamp - timestamp0;
1045 else
1046 delta = 0;
1047
1048 if (delta < 0)
1049 die("hm, delta: %" PRIu64 " < 0 ?\n", delta);
1050
1051
1052 sched_out = machine__findnew_thread(machine, switch_event->prev_pid);
1053 sched_in = machine__findnew_thread(machine, switch_event->next_pid);
1054
1055 out_events = thread_atoms_search(&atom_root, sched_out, &cmp_pid);
1056 if (!out_events) {
1057 thread_atoms_insert(sched_out);
1058 out_events = thread_atoms_search(&atom_root, sched_out, &cmp_pid);
1059 if (!out_events)
1060 die("out-event: Internal tree error");
1061 }
1062 add_sched_out_event(out_events, sched_out_state(switch_event), timestamp);
1063
1064 in_events = thread_atoms_search(&atom_root, sched_in, &cmp_pid);
1065 if (!in_events) {
1066 thread_atoms_insert(sched_in);
1067 in_events = thread_atoms_search(&atom_root, sched_in, &cmp_pid);
1068 if (!in_events)
1069 die("in-event: Internal tree error");
1070 /*
1071 * Take came in we have not heard about yet,
1072 * add in an initial atom in runnable state:
1073 */
1074 add_sched_out_event(in_events, 'R', timestamp);
1075 }
1076 add_sched_in_event(in_events, timestamp);
1077}
1078
1079static void
1080latency_runtime_event(struct trace_runtime_event *runtime_event,
1081 struct machine *machine,
1082 struct event_format *event __used,
1083 int cpu,
1084 u64 timestamp,
1085 struct thread *this_thread __used)
1086{
1087 struct thread *thread = machine__findnew_thread(machine, runtime_event->pid);
1088 struct work_atoms *atoms = thread_atoms_search(&atom_root, thread, &cmp_pid);
1089
1090 BUG_ON(cpu >= MAX_CPUS || cpu < 0);
1091 if (!atoms) {
1092 thread_atoms_insert(thread);
1093 atoms = thread_atoms_search(&atom_root, thread, &cmp_pid);
1094 if (!atoms)
1095 die("in-event: Internal tree error");
1096 add_sched_out_event(atoms, 'R', timestamp);
1097 }
1098
1099 add_runtime_event(atoms, runtime_event->runtime, timestamp);
1100}
1101
1102static void
1103latency_wakeup_event(struct trace_wakeup_event *wakeup_event,
1104 struct machine *machine,
1105 struct event_format *__event __used,
1106 int cpu __used,
1107 u64 timestamp,
1108 struct thread *thread __used)
1109{
1110 struct work_atoms *atoms;
1111 struct work_atom *atom;
1112 struct thread *wakee;
1113
1114 /* Note for later, it may be interesting to observe the failing cases */
1115 if (!wakeup_event->success)
1116 return;
1117
1118 wakee = machine__findnew_thread(machine, wakeup_event->pid);
1119 atoms = thread_atoms_search(&atom_root, wakee, &cmp_pid);
1120 if (!atoms) {
1121 thread_atoms_insert(wakee);
1122 atoms = thread_atoms_search(&atom_root, wakee, &cmp_pid);
1123 if (!atoms)
1124 die("wakeup-event: Internal tree error");
1125 add_sched_out_event(atoms, 'S', timestamp);
1126 }
1127
1128 BUG_ON(list_empty(&atoms->work_list));
1129
1130 atom = list_entry(atoms->work_list.prev, struct work_atom, list);
1131
1132 /*
1133 * You WILL be missing events if you've recorded only
1134 * one CPU, or are only looking at only one, so don't
1135 * make useless noise.
1136 */
1137 if (profile_cpu == -1 && atom->state != THREAD_SLEEPING)
1138 nr_state_machine_bugs++;
1139
1140 nr_timestamps++;
1141 if (atom->sched_out_time > timestamp) {
1142 nr_unordered_timestamps++;
1143 return;
1144 }
1145
1146 atom->state = THREAD_WAIT_CPU;
1147 atom->wake_up_time = timestamp;
1148}
1149
1150static void
1151latency_migrate_task_event(struct trace_migrate_task_event *migrate_task_event,
1152 struct machine *machine,
1153 struct event_format *__event __used,
1154 int cpu __used,
1155 u64 timestamp,
1156 struct thread *thread __used)
1157{
1158 struct work_atoms *atoms;
1159 struct work_atom *atom;
1160 struct thread *migrant;
1161
1162 /*
1163 * Only need to worry about migration when profiling one CPU.
1164 */
1165 if (profile_cpu == -1)
1166 return;
1167
1168 migrant = machine__findnew_thread(machine, migrate_task_event->pid);
1169 atoms = thread_atoms_search(&atom_root, migrant, &cmp_pid);
1170 if (!atoms) {
1171 thread_atoms_insert(migrant);
1172 register_pid(migrant->pid, migrant->comm);
1173 atoms = thread_atoms_search(&atom_root, migrant, &cmp_pid);
1174 if (!atoms)
1175 die("migration-event: Internal tree error");
1176 add_sched_out_event(atoms, 'R', timestamp);
1177 }
1178
1179 BUG_ON(list_empty(&atoms->work_list));
1180
1181 atom = list_entry(atoms->work_list.prev, struct work_atom, list);
1182 atom->sched_in_time = atom->sched_out_time = atom->wake_up_time = timestamp;
1183
1184 nr_timestamps++;
1185
1186 if (atom->sched_out_time > timestamp)
1187 nr_unordered_timestamps++;
1188}
1189
1190static struct trace_sched_handler lat_ops = {
1191 .wakeup_event = latency_wakeup_event,
1192 .switch_event = latency_switch_event,
1193 .runtime_event = latency_runtime_event,
1194 .fork_event = latency_fork_event,
1195 .migrate_task_event = latency_migrate_task_event,
1196};
1197
1198static void output_lat_thread(struct work_atoms *work_list)
1199{
1200 int i;
1201 int ret;
1202 u64 avg;
1203
1204 if (!work_list->nb_atoms)
1205 return;
1206 /*
1207 * Ignore idle threads:
1208 */
1209 if (!strcmp(work_list->thread->comm, "swapper"))
1210 return;
1211
1212 all_runtime += work_list->total_runtime;
1213 all_count += work_list->nb_atoms;
1214
1215 ret = printf(" %s:%d ", work_list->thread->comm, work_list->thread->pid);
1216
1217 for (i = 0; i < 24 - ret; i++)
1218 printf(" ");
1219
1220 avg = work_list->total_lat / work_list->nb_atoms;
1221
1222 printf("|%11.3f ms |%9" PRIu64 " | avg:%9.3f ms | max:%9.3f ms | max at: %9.6f s\n",
1223 (double)work_list->total_runtime / 1e6,
1224 work_list->nb_atoms, (double)avg / 1e6,
1225 (double)work_list->max_lat / 1e6,
1226 (double)work_list->max_lat_at / 1e9);
1227}
1228
1229static int pid_cmp(struct work_atoms *l, struct work_atoms *r)
1230{
1231 if (l->thread->pid < r->thread->pid)
1232 return -1;
1233 if (l->thread->pid > r->thread->pid)
1234 return 1;
1235
1236 return 0;
1237}
1238
1239static struct sort_dimension pid_sort_dimension = {
1240 .name = "pid",
1241 .cmp = pid_cmp,
1242};
1243
1244static int avg_cmp(struct work_atoms *l, struct work_atoms *r)
1245{
1246 u64 avgl, avgr;
1247
1248 if (!l->nb_atoms)
1249 return -1;
1250
1251 if (!r->nb_atoms)
1252 return 1;
1253
1254 avgl = l->total_lat / l->nb_atoms;
1255 avgr = r->total_lat / r->nb_atoms;
1256
1257 if (avgl < avgr)
1258 return -1;
1259 if (avgl > avgr)
1260 return 1;
1261
1262 return 0;
1263}
1264
1265static struct sort_dimension avg_sort_dimension = {
1266 .name = "avg",
1267 .cmp = avg_cmp,
1268};
1269
1270static int max_cmp(struct work_atoms *l, struct work_atoms *r)
1271{
1272 if (l->max_lat < r->max_lat)
1273 return -1;
1274 if (l->max_lat > r->max_lat)
1275 return 1;
1276
1277 return 0;
1278}
1279
1280static struct sort_dimension max_sort_dimension = {
1281 .name = "max",
1282 .cmp = max_cmp,
1283};
1284
1285static int switch_cmp(struct work_atoms *l, struct work_atoms *r)
1286{
1287 if (l->nb_atoms < r->nb_atoms)
1288 return -1;
1289 if (l->nb_atoms > r->nb_atoms)
1290 return 1;
1291
1292 return 0;
1293}
1294
1295static struct sort_dimension switch_sort_dimension = {
1296 .name = "switch",
1297 .cmp = switch_cmp,
1298};
1299
1300static int runtime_cmp(struct work_atoms *l, struct work_atoms *r)
1301{
1302 if (l->total_runtime < r->total_runtime)
1303 return -1;
1304 if (l->total_runtime > r->total_runtime)
1305 return 1;
1306
1307 return 0;
1308}
1309
1310static struct sort_dimension runtime_sort_dimension = {
1311 .name = "runtime",
1312 .cmp = runtime_cmp,
1313};
1314
1315static struct sort_dimension *available_sorts[] = {
1316 &pid_sort_dimension,
1317 &avg_sort_dimension,
1318 &max_sort_dimension,
1319 &switch_sort_dimension,
1320 &runtime_sort_dimension,
1321};
1322
1323#define NB_AVAILABLE_SORTS (int)(sizeof(available_sorts) / sizeof(struct sort_dimension *))
1324
1325static LIST_HEAD(sort_list);
1326
1327static int sort_dimension__add(const char *tok, struct list_head *list)
1328{
1329 int i;
1330
1331 for (i = 0; i < NB_AVAILABLE_SORTS; i++) {
1332 if (!strcmp(available_sorts[i]->name, tok)) {
1333 list_add_tail(&available_sorts[i]->list, list);
1334
1335 return 0;
1336 }
1337 }
1338
1339 return -1;
1340}
1341
1342static void setup_sorting(void);
1343
1344static void sort_lat(void)
1345{
1346 struct rb_node *node;
1347
1348 for (;;) {
1349 struct work_atoms *data;
1350 node = rb_first(&atom_root);
1351 if (!node)
1352 break;
1353
1354 rb_erase(node, &atom_root);
1355 data = rb_entry(node, struct work_atoms, node);
1356 __thread_latency_insert(&sorted_atom_root, data, &sort_list);
1357 }
1358}
1359
1360static struct trace_sched_handler *trace_handler;
1361
1362static void
1363process_sched_wakeup_event(struct perf_tool *tool __used,
1364 struct event_format *event,
1365 struct perf_sample *sample,
1366 struct machine *machine,
1367 struct thread *thread)
1368{
1369 void *data = sample->raw_data;
1370 struct trace_wakeup_event wakeup_event;
1371
1372 FILL_COMMON_FIELDS(wakeup_event, event, data);
1373
1374 FILL_ARRAY(wakeup_event, comm, event, data);
1375 FILL_FIELD(wakeup_event, pid, event, data);
1376 FILL_FIELD(wakeup_event, prio, event, data);
1377 FILL_FIELD(wakeup_event, success, event, data);
1378 FILL_FIELD(wakeup_event, cpu, event, data);
1379
1380 if (trace_handler->wakeup_event)
1381 trace_handler->wakeup_event(&wakeup_event, machine, event,
1382 sample->cpu, sample->time, thread);
1383}
1384
1385/*
1386 * Track the current task - that way we can know whether there's any
1387 * weird events, such as a task being switched away that is not current.
1388 */
1389static int max_cpu;
1390
1391static u32 curr_pid[MAX_CPUS] = { [0 ... MAX_CPUS-1] = -1 };
1392
1393static struct thread *curr_thread[MAX_CPUS];
1394
1395static char next_shortname1 = 'A';
1396static char next_shortname2 = '0';
1397
1398static void
1399map_switch_event(struct trace_switch_event *switch_event,
1400 struct machine *machine,
1401 struct event_format *event __used,
1402 int this_cpu,
1403 u64 timestamp,
1404 struct thread *thread __used)
1405{
1406 struct thread *sched_out __used, *sched_in;
1407 int new_shortname;
1408 u64 timestamp0;
1409 s64 delta;
1410 int cpu;
1411
1412 BUG_ON(this_cpu >= MAX_CPUS || this_cpu < 0);
1413
1414 if (this_cpu > max_cpu)
1415 max_cpu = this_cpu;
1416
1417 timestamp0 = cpu_last_switched[this_cpu];
1418 cpu_last_switched[this_cpu] = timestamp;
1419 if (timestamp0)
1420 delta = timestamp - timestamp0;
1421 else
1422 delta = 0;
1423
1424 if (delta < 0)
1425 die("hm, delta: %" PRIu64 " < 0 ?\n", delta);
1426
1427
1428 sched_out = machine__findnew_thread(machine, switch_event->prev_pid);
1429 sched_in = machine__findnew_thread(machine, switch_event->next_pid);
1430
1431 curr_thread[this_cpu] = sched_in;
1432
1433 printf(" ");
1434
1435 new_shortname = 0;
1436 if (!sched_in->shortname[0]) {
1437 sched_in->shortname[0] = next_shortname1;
1438 sched_in->shortname[1] = next_shortname2;
1439
1440 if (next_shortname1 < 'Z') {
1441 next_shortname1++;
1442 } else {
1443 next_shortname1='A';
1444 if (next_shortname2 < '9') {
1445 next_shortname2++;
1446 } else {
1447 next_shortname2='0';
1448 }
1449 }
1450 new_shortname = 1;
1451 }
1452
1453 for (cpu = 0; cpu <= max_cpu; cpu++) {
1454 if (cpu != this_cpu)
1455 printf(" ");
1456 else
1457 printf("*");
1458
1459 if (curr_thread[cpu]) {
1460 if (curr_thread[cpu]->pid)
1461 printf("%2s ", curr_thread[cpu]->shortname);
1462 else
1463 printf(". ");
1464 } else
1465 printf(" ");
1466 }
1467
1468 printf(" %12.6f secs ", (double)timestamp/1e9);
1469 if (new_shortname) {
1470 printf("%s => %s:%d\n",
1471 sched_in->shortname, sched_in->comm, sched_in->pid);
1472 } else {
1473 printf("\n");
1474 }
1475}
1476
1477static void
1478process_sched_switch_event(struct perf_tool *tool __used,
1479 struct event_format *event,
1480 struct perf_sample *sample,
1481 struct machine *machine,
1482 struct thread *thread)
1483{
1484 int this_cpu = sample->cpu;
1485 void *data = sample->raw_data;
1486 struct trace_switch_event switch_event;
1487
1488 FILL_COMMON_FIELDS(switch_event, event, data);
1489
1490 FILL_ARRAY(switch_event, prev_comm, event, data);
1491 FILL_FIELD(switch_event, prev_pid, event, data);
1492 FILL_FIELD(switch_event, prev_prio, event, data);
1493 FILL_FIELD(switch_event, prev_state, event, data);
1494 FILL_ARRAY(switch_event, next_comm, event, data);
1495 FILL_FIELD(switch_event, next_pid, event, data);
1496 FILL_FIELD(switch_event, next_prio, event, data);
1497
1498 if (curr_pid[this_cpu] != (u32)-1) {
1499 /*
1500 * Are we trying to switch away a PID that is
1501 * not current?
1502 */
1503 if (curr_pid[this_cpu] != switch_event.prev_pid)
1504 nr_context_switch_bugs++;
1505 }
1506 if (trace_handler->switch_event)
1507 trace_handler->switch_event(&switch_event, machine, event,
1508 this_cpu, sample->time, thread);
1509
1510 curr_pid[this_cpu] = switch_event.next_pid;
1511}
1512
1513static void
1514process_sched_runtime_event(struct perf_tool *tool __used,
1515 struct event_format *event,
1516 struct perf_sample *sample,
1517 struct machine *machine,
1518 struct thread *thread)
1519{
1520 void *data = sample->raw_data;
1521 struct trace_runtime_event runtime_event;
1522
1523 FILL_ARRAY(runtime_event, comm, event, data);
1524 FILL_FIELD(runtime_event, pid, event, data);
1525 FILL_FIELD(runtime_event, runtime, event, data);
1526 FILL_FIELD(runtime_event, vruntime, event, data);
1527
1528 if (trace_handler->runtime_event)
1529 trace_handler->runtime_event(&runtime_event, machine, event,
1530 sample->cpu, sample->time, thread);
1531}
1532
1533static void
1534process_sched_fork_event(struct perf_tool *tool __used,
1535 struct event_format *event,
1536 struct perf_sample *sample,
1537 struct machine *machine __used,
1538 struct thread *thread)
1539{
1540 void *data = sample->raw_data;
1541 struct trace_fork_event fork_event;
1542
1543 FILL_COMMON_FIELDS(fork_event, event, data);
1544
1545 FILL_ARRAY(fork_event, parent_comm, event, data);
1546 FILL_FIELD(fork_event, parent_pid, event, data);
1547 FILL_ARRAY(fork_event, child_comm, event, data);
1548 FILL_FIELD(fork_event, child_pid, event, data);
1549
1550 if (trace_handler->fork_event)
1551 trace_handler->fork_event(&fork_event, event,
1552 sample->cpu, sample->time, thread);
1553}
1554
1555static void
1556process_sched_exit_event(struct perf_tool *tool __used,
1557 struct event_format *event,
1558 struct perf_sample *sample __used,
1559 struct machine *machine __used,
1560 struct thread *thread __used)
1561{
1562 if (verbose)
1563 printf("sched_exit event %p\n", event);
1564}
1565
1566static void
1567process_sched_migrate_task_event(struct perf_tool *tool __used,
1568 struct event_format *event,
1569 struct perf_sample *sample,
1570 struct machine *machine,
1571 struct thread *thread)
1572{
1573 void *data = sample->raw_data;
1574 struct trace_migrate_task_event migrate_task_event;
1575
1576 FILL_COMMON_FIELDS(migrate_task_event, event, data);
1577
1578 FILL_ARRAY(migrate_task_event, comm, event, data);
1579 FILL_FIELD(migrate_task_event, pid, event, data);
1580 FILL_FIELD(migrate_task_event, prio, event, data);
1581 FILL_FIELD(migrate_task_event, cpu, event, data);
1582
1583 if (trace_handler->migrate_task_event)
1584 trace_handler->migrate_task_event(&migrate_task_event, machine,
1585 event, sample->cpu,
1586 sample->time, thread);
1587}
1588
1589typedef void (*tracepoint_handler)(struct perf_tool *tool, struct event_format *event,
1590 struct perf_sample *sample,
1591 struct machine *machine,
1592 struct thread *thread);
1593
1594static int perf_sched__process_tracepoint_sample(struct perf_tool *tool,
1595 union perf_event *event __used,
1596 struct perf_sample *sample,
1597 struct perf_evsel *evsel,
1598 struct machine *machine)
1599{
1600 struct thread *thread = machine__findnew_thread(machine, sample->pid);
1601
1602 if (thread == NULL) {
1603 pr_debug("problem processing %s event, skipping it.\n",
1604 evsel->name);
1605 return -1;
1606 }
1607
1608 evsel->hists.stats.total_period += sample->period;
1609 hists__inc_nr_events(&evsel->hists, PERF_RECORD_SAMPLE);
1610
1611 if (evsel->handler.func != NULL) {
1612 tracepoint_handler f = evsel->handler.func;
1613
1614 if (evsel->handler.data == NULL)
1615 evsel->handler.data = trace_find_event(evsel->attr.config);
1616
1617 f(tool, evsel->handler.data, sample, machine, thread);
1618 }
1619
1620 return 0;
1621}
1622
1623static struct perf_tool perf_sched = {
1624 .sample = perf_sched__process_tracepoint_sample,
1625 .comm = perf_event__process_comm,
1626 .lost = perf_event__process_lost,
1627 .fork = perf_event__process_task,
1628 .ordered_samples = true,
1629};
1630
1631static void read_events(bool destroy, struct perf_session **psession)
1632{
1633 int err = -EINVAL;
1634 const struct perf_evsel_str_handler handlers[] = {
1635 { "sched:sched_switch", process_sched_switch_event, },
1636 { "sched:sched_stat_runtime", process_sched_runtime_event, },
1637 { "sched:sched_wakeup", process_sched_wakeup_event, },
1638 { "sched:sched_wakeup_new", process_sched_wakeup_event, },
1639 { "sched:sched_process_fork", process_sched_fork_event, },
1640 { "sched:sched_process_exit", process_sched_exit_event, },
1641 { "sched:sched_migrate_task", process_sched_migrate_task_event, },
1642 };
1643 struct perf_session *session = perf_session__new(input_name, O_RDONLY,
1644 0, false, &perf_sched);
1645 if (session == NULL)
1646 die("No Memory");
1647
1648 err = perf_evlist__set_tracepoints_handlers_array(session->evlist, handlers);
1649 assert(err == 0);
1650
1651 if (perf_session__has_traces(session, "record -R")) {
1652 err = perf_session__process_events(session, &perf_sched);
1653 if (err)
1654 die("Failed to process events, error %d", err);
1655
1656 nr_events = session->hists.stats.nr_events[0];
1657 nr_lost_events = session->hists.stats.total_lost;
1658 nr_lost_chunks = session->hists.stats.nr_events[PERF_RECORD_LOST];
1659 }
1660
1661 if (destroy)
1662 perf_session__delete(session);
1663
1664 if (psession)
1665 *psession = session;
1666}
1667
1668static void print_bad_events(void)
1669{
1670 if (nr_unordered_timestamps && nr_timestamps) {
1671 printf(" INFO: %.3f%% unordered timestamps (%ld out of %ld)\n",
1672 (double)nr_unordered_timestamps/(double)nr_timestamps*100.0,
1673 nr_unordered_timestamps, nr_timestamps);
1674 }
1675 if (nr_lost_events && nr_events) {
1676 printf(" INFO: %.3f%% lost events (%ld out of %ld, in %ld chunks)\n",
1677 (double)nr_lost_events/(double)nr_events*100.0,
1678 nr_lost_events, nr_events, nr_lost_chunks);
1679 }
1680 if (nr_state_machine_bugs && nr_timestamps) {
1681 printf(" INFO: %.3f%% state machine bugs (%ld out of %ld)",
1682 (double)nr_state_machine_bugs/(double)nr_timestamps*100.0,
1683 nr_state_machine_bugs, nr_timestamps);
1684 if (nr_lost_events)
1685 printf(" (due to lost events?)");
1686 printf("\n");
1687 }
1688 if (nr_context_switch_bugs && nr_timestamps) {
1689 printf(" INFO: %.3f%% context switch bugs (%ld out of %ld)",
1690 (double)nr_context_switch_bugs/(double)nr_timestamps*100.0,
1691 nr_context_switch_bugs, nr_timestamps);
1692 if (nr_lost_events)
1693 printf(" (due to lost events?)");
1694 printf("\n");
1695 }
1696}
1697
1698static void __cmd_lat(void)
1699{
1700 struct rb_node *next;
1701 struct perf_session *session;
1702
1703 setup_pager();
1704 read_events(false, &session);
1705 sort_lat();
1706
1707 printf("\n ---------------------------------------------------------------------------------------------------------------\n");
1708 printf(" Task | Runtime ms | Switches | Average delay ms | Maximum delay ms | Maximum delay at |\n");
1709 printf(" ---------------------------------------------------------------------------------------------------------------\n");
1710
1711 next = rb_first(&sorted_atom_root);
1712
1713 while (next) {
1714 struct work_atoms *work_list;
1715
1716 work_list = rb_entry(next, struct work_atoms, node);
1717 output_lat_thread(work_list);
1718 next = rb_next(next);
1719 }
1720
1721 printf(" -----------------------------------------------------------------------------------------\n");
1722 printf(" TOTAL: |%11.3f ms |%9" PRIu64 " |\n",
1723 (double)all_runtime/1e6, all_count);
1724
1725 printf(" ---------------------------------------------------\n");
1726
1727 print_bad_events();
1728 printf("\n");
1729
1730 perf_session__delete(session);
1731}
1732
1733static struct trace_sched_handler map_ops = {
1734 .wakeup_event = NULL,
1735 .switch_event = map_switch_event,
1736 .runtime_event = NULL,
1737 .fork_event = NULL,
1738};
1739
1740static void __cmd_map(void)
1741{
1742 max_cpu = sysconf(_SC_NPROCESSORS_CONF);
1743
1744 setup_pager();
1745 read_events(true, NULL);
1746 print_bad_events();
1747}
1748
1749static void __cmd_replay(void)
1750{
1751 unsigned long i;
1752
1753 calibrate_run_measurement_overhead();
1754 calibrate_sleep_measurement_overhead();
1755
1756 test_calibrations();
1757
1758 read_events(true, NULL);
1759
1760 printf("nr_run_events: %ld\n", nr_run_events);
1761 printf("nr_sleep_events: %ld\n", nr_sleep_events);
1762 printf("nr_wakeup_events: %ld\n", nr_wakeup_events);
1763
1764 if (targetless_wakeups)
1765 printf("target-less wakeups: %ld\n", targetless_wakeups);
1766 if (multitarget_wakeups)
1767 printf("multi-target wakeups: %ld\n", multitarget_wakeups);
1768 if (nr_run_events_optimized)
1769 printf("run atoms optimized: %ld\n",
1770 nr_run_events_optimized);
1771
1772 print_task_traces();
1773 add_cross_task_wakeups();
1774
1775 create_tasks();
1776 printf("------------------------------------------------------------\n");
1777 for (i = 0; i < replay_repeat; i++)
1778 run_one_test();
1779}
1780
1781
1782static const char * const sched_usage[] = {
1783 "perf sched [<options>] {record|latency|map|replay|script}",
1784 NULL
1785};
1786
1787static const struct option sched_options[] = {
1788 OPT_STRING('i', "input", &input_name, "file",
1789 "input file name"),
1790 OPT_INCR('v', "verbose", &verbose,
1791 "be more verbose (show symbol address, etc)"),
1792 OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace,
1793 "dump raw trace in ASCII"),
1794 OPT_END()
1795};
1796
1797static const char * const latency_usage[] = {
1798 "perf sched latency [<options>]",
1799 NULL
1800};
1801
1802static const struct option latency_options[] = {
1803 OPT_STRING('s', "sort", &sort_order, "key[,key2...]",
1804 "sort by key(s): runtime, switch, avg, max"),
1805 OPT_INCR('v', "verbose", &verbose,
1806 "be more verbose (show symbol address, etc)"),
1807 OPT_INTEGER('C', "CPU", &profile_cpu,
1808 "CPU to profile on"),
1809 OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace,
1810 "dump raw trace in ASCII"),
1811 OPT_END()
1812};
1813
1814static const char * const replay_usage[] = {
1815 "perf sched replay [<options>]",
1816 NULL
1817};
1818
1819static const struct option replay_options[] = {
1820 OPT_UINTEGER('r', "repeat", &replay_repeat,
1821 "repeat the workload replay N times (-1: infinite)"),
1822 OPT_INCR('v', "verbose", &verbose,
1823 "be more verbose (show symbol address, etc)"),
1824 OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace,
1825 "dump raw trace in ASCII"),
1826 OPT_END()
1827};
1828
1829static void setup_sorting(void)
1830{
1831 char *tmp, *tok, *str = strdup(sort_order);
1832
1833 for (tok = strtok_r(str, ", ", &tmp);
1834 tok; tok = strtok_r(NULL, ", ", &tmp)) {
1835 if (sort_dimension__add(tok, &sort_list) < 0) {
1836 error("Unknown --sort key: `%s'", tok);
1837 usage_with_options(latency_usage, latency_options);
1838 }
1839 }
1840
1841 free(str);
1842
1843 sort_dimension__add("pid", &cmp_pid);
1844}
1845
1846static const char *record_args[] = {
1847 "record",
1848 "-a",
1849 "-R",
1850 "-f",
1851 "-m", "1024",
1852 "-c", "1",
1853 "-e", "sched:sched_switch",
1854 "-e", "sched:sched_stat_wait",
1855 "-e", "sched:sched_stat_sleep",
1856 "-e", "sched:sched_stat_iowait",
1857 "-e", "sched:sched_stat_runtime",
1858 "-e", "sched:sched_process_exit",
1859 "-e", "sched:sched_process_fork",
1860 "-e", "sched:sched_wakeup",
1861 "-e", "sched:sched_migrate_task",
1862};
1863
1864static int __cmd_record(int argc, const char **argv)
1865{
1866 unsigned int rec_argc, i, j;
1867 const char **rec_argv;
1868
1869 rec_argc = ARRAY_SIZE(record_args) + argc - 1;
1870 rec_argv = calloc(rec_argc + 1, sizeof(char *));
1871
1872 if (rec_argv == NULL)
1873 return -ENOMEM;
1874
1875 for (i = 0; i < ARRAY_SIZE(record_args); i++)
1876 rec_argv[i] = strdup(record_args[i]);
1877
1878 for (j = 1; j < (unsigned int)argc; j++, i++)
1879 rec_argv[i] = argv[j];
1880
1881 BUG_ON(i != rec_argc);
1882
1883 return cmd_record(i, rec_argv, NULL);
1884}
1885
1886int cmd_sched(int argc, const char **argv, const char *prefix __used)
1887{
1888 argc = parse_options(argc, argv, sched_options, sched_usage,
1889 PARSE_OPT_STOP_AT_NON_OPTION);
1890 if (!argc)
1891 usage_with_options(sched_usage, sched_options);
1892
1893 /*
1894 * Aliased to 'perf script' for now:
1895 */
1896 if (!strcmp(argv[0], "script"))
1897 return cmd_script(argc, argv, prefix);
1898
1899 symbol__init();
1900 if (!strncmp(argv[0], "rec", 3)) {
1901 return __cmd_record(argc, argv);
1902 } else if (!strncmp(argv[0], "lat", 3)) {
1903 trace_handler = &lat_ops;
1904 if (argc > 1) {
1905 argc = parse_options(argc, argv, latency_options, latency_usage, 0);
1906 if (argc)
1907 usage_with_options(latency_usage, latency_options);
1908 }
1909 setup_sorting();
1910 __cmd_lat();
1911 } else if (!strcmp(argv[0], "map")) {
1912 trace_handler = &map_ops;
1913 setup_sorting();
1914 __cmd_map();
1915 } else if (!strncmp(argv[0], "rep", 3)) {
1916 trace_handler = &replay_ops;
1917 if (argc) {
1918 argc = parse_options(argc, argv, replay_options, replay_usage, 0);
1919 if (argc)
1920 usage_with_options(replay_usage, replay_options);
1921 }
1922 __cmd_replay();
1923 } else {
1924 usage_with_options(sched_usage, sched_options);
1925 }
1926
1927 return 0;
1928}