]> git.proxmox.com Git - mirror_ubuntu-jammy-kernel.git/blame - tools/perf/builtin-sched.c
projects / mirror_ubuntu-jammy-kernel.git / blame
? search:
summary | shortlog | log | commit | commitdiff | tree
blob | blame (incremental) | history | HEAD
Merge tag 'v5.6' into mips-next
[mirror_ubuntu-jammy-kernel.git] / tools / perf / builtin-sched.c
CommitLineData
b2441318 1// SPDX-License-Identifier: GPL-2.0
0a02ad93 2#include "builtin.h"
b1ffe8f3 3#include "perf.h"
91854f9a 4#include "perf-sys.h"
0a02ad93 5
87ffb6c6 6#include "util/cpumap.h"
ee29be62 7#include "util/evlist.h"
e3f42609 8#include "util/evsel.h"
ca125277 9#include "util/evsel_fprintf.h"
0a02ad93
IM		 10#include "util/symbol.h"
11#include "util/thread.h"
12#include "util/header.h"
94c744b6 13#include "util/session.h"
45694aa7 14#include "util/tool.h"
57480d2c 15#include "util/cloexec.h"
a151a37a 16#include "util/thread_map.h"
8cd91195 17#include "util/color.h"
49394a2a 18#include "util/stat.h"
6a9fa4e3 19#include "util/string2.h"
6c973c90 20#include "util/callchain.h"
853b7407 21#include "util/time-utils.h"
0a02ad93 22
fa0d9846 23#include <subcmd/pager.h>
4b6ab94e 24#include <subcmd/parse-options.h>
b1ffe8f3 25#include "util/trace-event.h"
0a02ad93 26
0a02ad93 27#include "util/debug.h"
f12be047 28#include "util/event.h"
0a02ad93 29
877a7a11 30#include <linux/kernel.h>
49394a2a 31#include <linux/log2.h>
7f7c536f 32#include <linux/zalloc.h>
b1ffe8f3 33#include <sys/prctl.h>
7b78f136 34#include <sys/resource.h>
fd20e811 35#include <inttypes.h>
0a02ad93 36
a43783ae 37#include <errno.h>
b1ffe8f3
IM		 38#include <semaphore.h>
39#include <pthread.h>
40#include <math.h>
cb06ac25 41#include <api/fs/fs.h>
87ffb6c6 42#include <perf/cpumap.h>
4fc76e49 43#include <linux/time64.h>
6ef81c55 44#include <linux/err.h>
419ab0d6 45
3052ba56 46#include <linux/ctype.h>
3d689ed6 47
b1ffe8f3
IM		 48#define PR_SET_NAME 15 /* Set process name */
49#define MAX_CPUS 4096
b1ffe8f3
IM		 50#define COMM_LEN 20
51#define SYM_LEN 129
a35e27d0 52#define MAX_PID 1024000
ec156764 53
c30d630d
DA		 54static const char *cpu_list;
55static DECLARE_BITMAP(cpu_bitmap, MAX_NR_CPUS);
56
39aeb52f 57struct sched_atom;
ec156764 58
b1ffe8f3
IM		 59struct task_desc {
60 unsigned long nr;
61 unsigned long pid;
62 char comm[COMM_LEN];
ec156764 63
b1ffe8f3
IM		 64 unsigned long nr_events;
65 unsigned long curr_event;
39aeb52f 66 struct sched_atom **atoms;
b1ffe8f3
IM		 67
68 pthread_t thread;
69 sem_t sleep_sem;
ec156764 70
b1ffe8f3
IM		 71 sem_t ready_for_work;
72 sem_t work_done_sem;
73
74 u64 cpu_usage;
75};
76
77enum sched_event_type {
78 SCHED_EVENT_RUN,
79 SCHED_EVENT_SLEEP,
80 SCHED_EVENT_WAKEUP,
55ffb7a6 81 SCHED_EVENT_MIGRATION,
b1ffe8f3
IM		 82};
83
39aeb52f 84struct sched_atom {
b1ffe8f3 85 enum sched_event_type type;
eed05fe7 86 int specific_wait;
b1ffe8f3
IM		 87 u64 timestamp;
88 u64 duration;
89 unsigned long nr;
b1ffe8f3
IM		 90 sem_t *wait_sem;
91 struct task_desc *wakee;
92};
93
e936e8e4 94#define TASK_STATE_TO_CHAR_STR "RSDTtZXxKWP"
b1ffe8f3 95
941bdea7
NK		 96/* task state bitmask, copied from include/linux/sched.h */
97#define TASK_RUNNING 0
98#define TASK_INTERRUPTIBLE 1
99#define TASK_UNINTERRUPTIBLE 2
100#define __TASK_STOPPED 4
101#define __TASK_TRACED 8
102/* in tsk->exit_state */
103#define EXIT_DEAD 16
104#define EXIT_ZOMBIE 32
105#define EXIT_TRACE (EXIT_ZOMBIE | EXIT_DEAD)
106/* in tsk->state again */
107#define TASK_DEAD 64
108#define TASK_WAKEKILL 128
109#define TASK_WAKING 256
110#define TASK_PARKED 512
111
b1ffe8f3
IM		 112enum thread_state {
113 THREAD_SLEEPING = 0,
114 THREAD_WAIT_CPU,
115 THREAD_SCHED_IN,
116 THREAD_IGNORE
117};
118
119struct work_atom {
120 struct list_head list;
121 enum thread_state state;
aa1ab9d2 122 u64 sched_out_time;
b1ffe8f3
IM		 123 u64 wake_up_time;
124 u64 sched_in_time;
125 u64 runtime;
126};
127
39aeb52f 128struct work_atoms {
129 struct list_head work_list;
b1ffe8f3
IM		 130 struct thread *thread;
131 struct rb_node node;
132 u64 max_lat;
3786310a 133 u64 max_lat_at;
b1ffe8f3
IM		 134 u64 total_lat;
135 u64 nb_atoms;
136 u64 total_runtime;
2f80dd44 137 int num_merged;
b1ffe8f3
IM		 138};
139
39aeb52f 140typedef int (*sort_fn_t)(struct work_atoms *, struct work_atoms *);
b1ffe8f3 141
9ec3f4e4 142struct perf_sched;
0e9b07e5 143
9ec3f4e4 144struct trace_sched_handler {
32dcd021 145 int (*switch_event)(struct perf_sched *sched, struct evsel *evsel,
9ec3f4e4 146 struct perf_sample *sample, struct machine *machine);
0e9b07e5 147
32dcd021 148 int (*runtime_event)(struct perf_sched *sched, struct evsel *evsel,
9ec3f4e4 149 struct perf_sample *sample, struct machine *machine);
0e9b07e5 150
32dcd021 151 int (*wakeup_event)(struct perf_sched *sched, struct evsel *evsel,
9ec3f4e4 152 struct perf_sample *sample, struct machine *machine);
0e9b07e5 153
cb627505
DA		 154 /* PERF_RECORD_FORK event, not sched_process_fork tracepoint */
155 int (*fork_event)(struct perf_sched *sched, union perf_event *event,
156 struct machine *machine);
0e9b07e5
ACM		 157
158 int (*migrate_task_event)(struct perf_sched *sched,
32dcd021 159 struct evsel *evsel,
9ec3f4e4
ACM		 160 struct perf_sample *sample,
161 struct machine *machine);
0e9b07e5
ACM		 162};
163
a151a37a 164#define COLOR_PIDS PERF_COLOR_BLUE
cf294f24 165#define COLOR_CPUS PERF_COLOR_BG_RED
a151a37a 166
99623c62
JO		 167struct perf_sched_map {
168 DECLARE_BITMAP(comp_cpus_mask, MAX_CPUS);
169 int *comp_cpus;
170 bool comp;
9749b90e 171 struct perf_thread_map *color_pids;
a151a37a 172 const char *color_pids_str;
f854839b 173 struct perf_cpu_map *color_cpus;
cf294f24 174 const char *color_cpus_str;
f854839b 175 struct perf_cpu_map *cpus;
73643bb6 176 const char *cpus_str;
99623c62
JO		 177};
178
0e9b07e5
ACM		 179struct perf_sched {
180 struct perf_tool tool;
0e9b07e5
ACM		 181 const char *sort_order;
182 unsigned long nr_tasks;
cb06ac25 183 struct task_desc **pid_to_task;
0e9b07e5
ACM		 184 struct task_desc **tasks;
185 const struct trace_sched_handler *tp_handler;
186 pthread_mutex_t start_work_mutex;
187 pthread_mutex_t work_done_wait_mutex;
188 int profile_cpu;
189/*
190 * Track the current task - that way we can know whether there's any
191 * weird events, such as a task being switched away that is not current.
192 */
193 int max_cpu;
194 u32 curr_pid[MAX_CPUS];
195 struct thread *curr_thread[MAX_CPUS];
196 char next_shortname1;
197 char next_shortname2;
198 unsigned int replay_repeat;
199 unsigned long nr_run_events;
200 unsigned long nr_sleep_events;
201 unsigned long nr_wakeup_events;
202 unsigned long nr_sleep_corrections;
203 unsigned long nr_run_events_optimized;
204 unsigned long targetless_wakeups;
205 unsigned long multitarget_wakeups;
206 unsigned long nr_runs;
207 unsigned long nr_timestamps;
208 unsigned long nr_unordered_timestamps;
0e9b07e5
ACM		 209 unsigned long nr_context_switch_bugs;
210 unsigned long nr_events;
211 unsigned long nr_lost_chunks;
212 unsigned long nr_lost_events;
213 u64 run_measurement_overhead;
214 u64 sleep_measurement_overhead;
215 u64 start_time;
216 u64 cpu_usage;
217 u64 runavg_cpu_usage;
218 u64 parent_cpu_usage;
219 u64 runavg_parent_cpu_usage;
220 u64 sum_runtime;
221 u64 sum_fluct;
222 u64 run_avg;
223 u64 all_runtime;
224 u64 all_count;
225 u64 cpu_last_switched[MAX_CPUS];
cb4c13a5 226 struct rb_root_cached atom_root, sorted_atom_root, merged_atom_root;
0e9b07e5 227 struct list_head sort_list, cmp_pid;
939cda52 228 bool force;
2f80dd44 229 bool skip_merge;
99623c62 230 struct perf_sched_map map;
52df138c
DA		 231
232 /* options for timehist command */
233 bool summary;
234 bool summary_only;
699b5b92 235 bool idle_hist;
6c973c90
DA		 236 bool show_callchain;
237 unsigned int max_stack;
a407b067 238 bool show_cpu_visual;
fc1469f1 239 bool show_wakeups;
292c4a8f 240 bool show_next;
350f54fa 241 bool show_migrations;
414e050c 242 bool show_state;
52df138c 243 u64 skipped_samples;
853b7407
DA		 244 const char *time_str;
245 struct perf_time_interval ptime;
9396c9cb 246 struct perf_time_interval hist_time;
0e9b07e5 247};
b1ffe8f3 248
49394a2a
DA		 249/* per thread run time data */
250struct thread_runtime {
251 u64 last_time; /* time of previous sched in/out event */
252 u64 dt_run; /* run time */
941bdea7
NK		 253 u64 dt_sleep; /* time between CPU access by sleep (off cpu) */
254 u64 dt_iowait; /* time between CPU access by iowait (off cpu) */
255 u64 dt_preempt; /* time between CPU access by preempt (off cpu) */
49394a2a
DA		 256 u64 dt_delay; /* time between wakeup and sched-in */
257 u64 ready_to_run; /* time of wakeup */
258
259 struct stats run_stats;
260 u64 total_run_time;
587782c5
NK		 261 u64 total_sleep_time;
262 u64 total_iowait_time;
263 u64 total_preempt_time;
264 u64 total_delay_time;
350f54fa 265
941bdea7 266 int last_state;
8640da9f
CD		 267
268 char shortname[3];
99a3c3a9
CD		 269 bool comm_changed;
270
350f54fa 271 u64 migrations;
49394a2a
DA		 272};
273
274/* per event run time data */
275struct evsel_runtime {
276 u64 *last_time; /* time this event was last seen per cpu */
277 u32 ncpu; /* highest cpu slot allocated */
278};
279
3bc2fa9c
NK		 280/* per cpu idle time data */
281struct idle_thread_runtime {
282 struct thread_runtime tr;
283 struct thread *last_thread;
cb4c13a5 284 struct rb_root_cached sorted_root;
3bc2fa9c
NK		 285 struct callchain_root callchain;
286 struct callchain_cursor cursor;
287};
288
49394a2a
DA		 289/* track idle times per cpu */
290static struct thread **idle_threads;
291static int idle_max_cpu;
292static char idle_comm[] = "<idle>";
293
b1ffe8f3 294static u64 get_nsecs(void)
ec156764
IM		 295{
296 struct timespec ts;
297
298 clock_gettime(CLOCK_MONOTONIC, &ts);
299
4fc76e49 300 return ts.tv_sec * NSEC_PER_SEC + ts.tv_nsec;
ec156764
IM		 301}
302
0e9b07e5 303static void burn_nsecs(struct perf_sched *sched, u64 nsecs)
ec156764 304{
b1ffe8f3 305 u64 T0 = get_nsecs(), T1;
ec156764
IM		 306
307 do {
308 T1 = get_nsecs();
0e9b07e5 309 } while (T1 + sched->run_measurement_overhead < T0 + nsecs);
ec156764
IM		 310}
311
b1ffe8f3 312static void sleep_nsecs(u64 nsecs)
ec156764
IM		 313{
314 struct timespec ts;
315
316 ts.tv_nsec = nsecs % 999999999;
317 ts.tv_sec = nsecs / 999999999;
318
319 nanosleep(&ts, NULL);
320}
321
0e9b07e5 322static void calibrate_run_measurement_overhead(struct perf_sched *sched)
ec156764 323{
4fc76e49 324 u64 T0, T1, delta, min_delta = NSEC_PER_SEC;
ec156764
IM		 325 int i;
326
327 for (i = 0; i < 10; i++) {
328 T0 = get_nsecs();
0e9b07e5 329 burn_nsecs(sched, 0);
ec156764
IM		 330 T1 = get_nsecs();
331 delta = T1-T0;
332 min_delta = min(min_delta, delta);
333 }
0e9b07e5 334 sched->run_measurement_overhead = min_delta;
ec156764 335
9486aa38 336 printf("run measurement overhead: %" PRIu64 " nsecs\n", min_delta);
ec156764
IM		 337}
338
0e9b07e5 339static void calibrate_sleep_measurement_overhead(struct perf_sched *sched)
ec156764 340{
4fc76e49 341 u64 T0, T1, delta, min_delta = NSEC_PER_SEC;
ec156764
IM		 342 int i;
343
344 for (i = 0; i < 10; i++) {
345 T0 = get_nsecs();
346 sleep_nsecs(10000);
347 T1 = get_nsecs();
348 delta = T1-T0;
349 min_delta = min(min_delta, delta);
350 }
351 min_delta -= 10000;
0e9b07e5 352 sched->sleep_measurement_overhead = min_delta;
ec156764 353
9486aa38 354 printf("sleep measurement overhead: %" PRIu64 " nsecs\n", min_delta);
ec156764
IM		 355}
356
39aeb52f 357static struct sched_atom *
b1ffe8f3 358get_new_event(struct task_desc *task, u64 timestamp)
ec156764 359{
36479484 360 struct sched_atom *event = zalloc(sizeof(*event));
ec156764
IM		 361 unsigned long idx = task->nr_events;
362 size_t size;
363
364 event->timestamp = timestamp;
365 event->nr = idx;
366
367 task->nr_events++;
39aeb52f 368 size = sizeof(struct sched_atom *) * task->nr_events;
369 task->atoms = realloc(task->atoms, size);
370 BUG_ON(!task->atoms);
ec156764 371
39aeb52f 372 task->atoms[idx] = event;
ec156764
IM		 373
374 return event;
375}
376
39aeb52f 377static struct sched_atom *last_event(struct task_desc *task)
ec156764
IM		 378{
379 if (!task->nr_events)
380 return NULL;
381
39aeb52f 382 return task->atoms[task->nr_events - 1];
ec156764
IM		 383}
384
0e9b07e5
ACM		 385static void add_sched_event_run(struct perf_sched *sched, struct task_desc *task,
386 u64 timestamp, u64 duration)
ec156764 387{
39aeb52f 388 struct sched_atom *event, *curr_event = last_event(task);
ec156764
IM		 389
390 /*
fbf94829
IM		 391 * optimize an existing RUN event by merging this one
392 * to it:
393 */
ec156764 394 if (curr_event && curr_event->type == SCHED_EVENT_RUN) {
0e9b07e5 395 sched->nr_run_events_optimized++;
ec156764
IM		 396 curr_event->duration += duration;
397 return;
398 }
399
400 event = get_new_event(task, timestamp);
401
402 event->type = SCHED_EVENT_RUN;
403 event->duration = duration;
404
0e9b07e5 405 sched->nr_run_events++;
ec156764
IM		 406}
407
0e9b07e5
ACM		 408static void add_sched_event_wakeup(struct perf_sched *sched, struct task_desc *task,
409 u64 timestamp, struct task_desc *wakee)
ec156764 410{
39aeb52f 411 struct sched_atom *event, *wakee_event;
ec156764
IM		 412
413 event = get_new_event(task, timestamp);
414 event->type = SCHED_EVENT_WAKEUP;
415 event->wakee = wakee;
416
417 wakee_event = last_event(wakee);
418 if (!wakee_event || wakee_event->type != SCHED_EVENT_SLEEP) {
0e9b07e5 419 sched->targetless_wakeups++;
ec156764
IM		 420 return;
421 }
422 if (wakee_event->wait_sem) {
0e9b07e5 423 sched->multitarget_wakeups++;
ec156764
IM		 424 return;
425 }
426
36479484 427 wakee_event->wait_sem = zalloc(sizeof(*wakee_event->wait_sem));
ec156764
IM		 428 sem_init(wakee_event->wait_sem, 0, 0);
429 wakee_event->specific_wait = 1;
430 event->wait_sem = wakee_event->wait_sem;
431
0e9b07e5 432 sched->nr_wakeup_events++;
ec156764
IM		 433}
434
0e9b07e5
ACM		 435static void add_sched_event_sleep(struct perf_sched *sched, struct task_desc *task,
436 u64 timestamp, u64 task_state __maybe_unused)
ec156764 437{
39aeb52f 438 struct sched_atom *event = get_new_event(task, timestamp);
ec156764
IM		 439
440 event->type = SCHED_EVENT_SLEEP;
441
0e9b07e5 442 sched->nr_sleep_events++;
ec156764
IM		 443}
444
0e9b07e5
ACM		 445static struct task_desc *register_pid(struct perf_sched *sched,
446 unsigned long pid, const char *comm)
ec156764
IM		 447{
448 struct task_desc *task;
cb06ac25 449 static int pid_max;
ec156764 450
cb06ac25
YS		 451 if (sched->pid_to_task == NULL) {
452 if (sysctl__read_int("kernel/pid_max", &pid_max) < 0)
453 pid_max = MAX_PID;
454 BUG_ON((sched->pid_to_task = calloc(pid_max, sizeof(struct task_desc *))) == NULL);
455 }
3a423a5c
YS		 456 if (pid >= (unsigned long)pid_max) {
457 BUG_ON((sched->pid_to_task = realloc(sched->pid_to_task, (pid + 1) *
458 sizeof(struct task_desc *))) == NULL);
459 while (pid >= (unsigned long)pid_max)
460 sched->pid_to_task[pid_max++] = NULL;
461 }
ec156764 462
0e9b07e5 463 task = sched->pid_to_task[pid];
ec156764
IM		 464
465 if (task)
466 return task;
467
36479484 468 task = zalloc(sizeof(*task));
ec156764 469 task->pid = pid;
0e9b07e5 470 task->nr = sched->nr_tasks;
ec156764
IM		 471 strcpy(task->comm, comm);
472 /*
473 * every task starts in sleeping state - this gets ignored
474 * if there's no wakeup pointing to this sleep state:
475 */
0e9b07e5 476 add_sched_event_sleep(sched, task, 0, 0);
ec156764 477
0e9b07e5
ACM		 478 sched->pid_to_task[pid] = task;
479 sched->nr_tasks++;
0755bc4d 480 sched->tasks = realloc(sched->tasks, sched->nr_tasks * sizeof(struct task_desc *));
0e9b07e5
ACM		 481 BUG_ON(!sched->tasks);
482 sched->tasks[task->nr] = task;
ec156764 483
bb963e16 484 if (verbose > 0)
0e9b07e5 485 printf("registered task #%ld, PID %ld (%s)\n", sched->nr_tasks, pid, comm);
ec156764
IM		 486
487 return task;
488}
489
490
0e9b07e5 491static void print_task_traces(struct perf_sched *sched)
ec156764
IM		 492{
493 struct task_desc *task;
494 unsigned long i;
495
0e9b07e5
ACM		 496 for (i = 0; i < sched->nr_tasks; i++) {
497 task = sched->tasks[i];
ad236fd2 498 printf("task %6ld (%20s:%10ld), nr_events: %ld\n",
ec156764
IM		 499 task->nr, task->comm, task->pid, task->nr_events);
500 }
501}
502
0e9b07e5 503static void add_cross_task_wakeups(struct perf_sched *sched)
ec156764
IM		 504{
505 struct task_desc *task1, *task2;
506 unsigned long i, j;
507
0e9b07e5
ACM		 508 for (i = 0; i < sched->nr_tasks; i++) {
509 task1 = sched->tasks[i];
ec156764 510 j = i + 1;
0e9b07e5 511 if (j == sched->nr_tasks)
ec156764 512 j = 0;
0e9b07e5
ACM		 513 task2 = sched->tasks[j];
514 add_sched_event_wakeup(sched, task1, 0, task2);
ec156764
IM		 515 }
516}
517
0e9b07e5
ACM		 518static void perf_sched__process_event(struct perf_sched *sched,
519 struct sched_atom *atom)
ec156764
IM		 520{
521 int ret = 0;
ec156764 522
39aeb52f 523 switch (atom->type) {
ec156764 524 case SCHED_EVENT_RUN:
0e9b07e5 525 burn_nsecs(sched, atom->duration);
ec156764
IM		 526 break;
527 case SCHED_EVENT_SLEEP:
39aeb52f 528 if (atom->wait_sem)
529 ret = sem_wait(atom->wait_sem);
ec156764
IM		 530 BUG_ON(ret);
531 break;
532 case SCHED_EVENT_WAKEUP:
39aeb52f 533 if (atom->wait_sem)
534 ret = sem_post(atom->wait_sem);
ec156764
IM		 535 BUG_ON(ret);
536 break;
55ffb7a6
MG		 537 case SCHED_EVENT_MIGRATION:
538 break;
ec156764
IM		 539 default:
540 BUG_ON(1);
541 }
542}
543
b1ffe8f3 544static u64 get_cpu_usage_nsec_parent(void)
ec156764
IM		 545{
546 struct rusage ru;
b1ffe8f3 547 u64 sum;
ec156764
IM		 548 int err;
549
550 err = getrusage(RUSAGE_SELF, &ru);
551 BUG_ON(err);
552
4fc76e49
ACM		 553 sum = ru.ru_utime.tv_sec * NSEC_PER_SEC + ru.ru_utime.tv_usec * NSEC_PER_USEC;
554 sum += ru.ru_stime.tv_sec * NSEC_PER_SEC + ru.ru_stime.tv_usec * NSEC_PER_USEC;
ec156764
IM		 555
556 return sum;
557}
558
939cda52 559static int self_open_counters(struct perf_sched *sched, unsigned long cur_task)
ec156764 560{
c0c9e721 561 struct perf_event_attr attr;
939cda52 562 char sbuf[STRERR_BUFSIZE], info[STRERR_BUFSIZE];
c0c9e721 563 int fd;
939cda52
YS		 564 struct rlimit limit;
565 bool need_privilege = false;
ec156764 566
c0c9e721 567 memset(&attr, 0, sizeof(attr));
ec156764 568
c0c9e721
XG		 569 attr.type = PERF_TYPE_SOFTWARE;
570 attr.config = PERF_COUNT_SW_TASK_CLOCK;
ec156764 571
939cda52 572force_again:
57480d2c
YD		 573 fd = sys_perf_event_open(&attr, 0, -1, -1,
574 perf_event_open_cloexec_flag());
c0c9e721 575
1aff59be 576 if (fd < 0) {
939cda52
YS		 577 if (errno == EMFILE) {
578 if (sched->force) {
579 BUG_ON(getrlimit(RLIMIT_NOFILE, &limit) == -1);
580 limit.rlim_cur += sched->nr_tasks - cur_task;
581 if (limit.rlim_cur > limit.rlim_max) {
582 limit.rlim_max = limit.rlim_cur;
583 need_privilege = true;
584 }
585 if (setrlimit(RLIMIT_NOFILE, &limit) == -1) {
586 if (need_privilege && errno == EPERM)
587 strcpy(info, "Need privilege\n");
588 } else
589 goto force_again;
590 } else
591 strcpy(info, "Have a try with -f option\n");
592 }
60b7d14a 593 pr_err("Error: sys_perf_event_open() syscall returned "
939cda52 594 "with %d (%s)\n%s", fd,
c8b5f2c9 595 str_error_r(errno, sbuf, sizeof(sbuf)), info);
1aff59be
YS		 596 exit(EXIT_FAILURE);
597 }
c0c9e721
XG		 598 return fd;
599}
600
601static u64 get_cpu_usage_nsec_self(int fd)
602{
603 u64 runtime;
604 int ret;
605
606 ret = read(fd, &runtime, sizeof(runtime));
607 BUG_ON(ret != sizeof(runtime));
608
609 return runtime;
ec156764
IM		 610}
611
0e9b07e5
ACM		 612struct sched_thread_parms {
613 struct task_desc *task;
614 struct perf_sched *sched;
08097abc 615 int fd;
0e9b07e5
ACM		 616};
617
ec156764
IM		 618static void *thread_func(void *ctx)
619{
0e9b07e5
ACM		 620 struct sched_thread_parms *parms = ctx;
621 struct task_desc *this_task = parms->task;
622 struct perf_sched *sched = parms->sched;
b1ffe8f3 623 u64 cpu_usage_0, cpu_usage_1;
ec156764
IM		 624 unsigned long i, ret;
625 char comm2[22];
08097abc 626 int fd = parms->fd;
ec156764 627
74cf249d 628 zfree(&parms);
0e9b07e5 629
ec156764
IM		 630 sprintf(comm2, ":%s", this_task->comm);
631 prctl(PR_SET_NAME, comm2);
a116e05d
ACM		 632 if (fd < 0)
633 return NULL;
ec156764
IM		 634again:
635 ret = sem_post(&this_task->ready_for_work);
636 BUG_ON(ret);
0e9b07e5 637 ret = pthread_mutex_lock(&sched->start_work_mutex);
ec156764 638 BUG_ON(ret);
0e9b07e5 639 ret = pthread_mutex_unlock(&sched->start_work_mutex);
ec156764 640 BUG_ON(ret);
ec156764 641
c0c9e721 642 cpu_usage_0 = get_cpu_usage_nsec_self(fd);
ec156764
IM		 643
644 for (i = 0; i < this_task->nr_events; i++) {
645 this_task->curr_event = i;
0e9b07e5 646 perf_sched__process_event(sched, this_task->atoms[i]);
ec156764
IM		 647 }
648
c0c9e721 649 cpu_usage_1 = get_cpu_usage_nsec_self(fd);
ec156764 650 this_task->cpu_usage = cpu_usage_1 - cpu_usage_0;
ec156764
IM		 651 ret = sem_post(&this_task->work_done_sem);
652 BUG_ON(ret);
ec156764 653
0e9b07e5 654 ret = pthread_mutex_lock(&sched->work_done_wait_mutex);
ec156764 655 BUG_ON(ret);
0e9b07e5 656 ret = pthread_mutex_unlock(&sched->work_done_wait_mutex);
ec156764 657 BUG_ON(ret);
ec156764
IM		 658
659 goto again;
660}
661
0e9b07e5 662static void create_tasks(struct perf_sched *sched)
ec156764
IM		 663{
664 struct task_desc *task;
665 pthread_attr_t attr;
666 unsigned long i;
667 int err;
668
669 err = pthread_attr_init(&attr);
670 BUG_ON(err);
12f7e036
JP		 671 err = pthread_attr_setstacksize(&attr,
672 (size_t) max(16 * 1024, PTHREAD_STACK_MIN));
ec156764 673 BUG_ON(err);
0e9b07e5 674 err = pthread_mutex_lock(&sched->start_work_mutex);
ec156764 675 BUG_ON(err);
0e9b07e5 676 err = pthread_mutex_lock(&sched->work_done_wait_mutex);
ec156764 677 BUG_ON(err);
0e9b07e5
ACM		 678 for (i = 0; i < sched->nr_tasks; i++) {
679 struct sched_thread_parms *parms = malloc(sizeof(*parms));
680 BUG_ON(parms == NULL);
681 parms->task = task = sched->tasks[i];
682 parms->sched = sched;
939cda52 683 parms->fd = self_open_counters(sched, i);
ec156764
IM		 684 sem_init(&task->sleep_sem, 0, 0);
685 sem_init(&task->ready_for_work, 0, 0);
686 sem_init(&task->work_done_sem, 0, 0);
687 task->curr_event = 0;
0e9b07e5 688 err = pthread_create(&task->thread, &attr, thread_func, parms);
ec156764
IM		 689 BUG_ON(err);
690 }
691}
692
0e9b07e5 693static void wait_for_tasks(struct perf_sched *sched)
ec156764 694{
b1ffe8f3 695 u64 cpu_usage_0, cpu_usage_1;
ec156764
IM		 696 struct task_desc *task;
697 unsigned long i, ret;
698
0e9b07e5
ACM		 699 sched->start_time = get_nsecs();
700 sched->cpu_usage = 0;
701 pthread_mutex_unlock(&sched->work_done_wait_mutex);
ec156764 702
0e9b07e5
ACM		 703 for (i = 0; i < sched->nr_tasks; i++) {
704 task = sched->tasks[i];
ec156764
IM		 705 ret = sem_wait(&task->ready_for_work);
706 BUG_ON(ret);
707 sem_init(&task->ready_for_work, 0, 0);
708 }
0e9b07e5 709 ret = pthread_mutex_lock(&sched->work_done_wait_mutex);
ec156764
IM		 710 BUG_ON(ret);
711
712 cpu_usage_0 = get_cpu_usage_nsec_parent();
713
0e9b07e5 714 pthread_mutex_unlock(&sched->start_work_mutex);
ec156764 715
0e9b07e5
ACM		 716 for (i = 0; i < sched->nr_tasks; i++) {
717 task = sched->tasks[i];
ec156764
IM		 718 ret = sem_wait(&task->work_done_sem);
719 BUG_ON(ret);
720 sem_init(&task->work_done_sem, 0, 0);
0e9b07e5 721 sched->cpu_usage += task->cpu_usage;
ec156764
IM		 722 task->cpu_usage = 0;
723 }
724
725 cpu_usage_1 = get_cpu_usage_nsec_parent();
0e9b07e5
ACM		 726 if (!sched->runavg_cpu_usage)
727 sched->runavg_cpu_usage = sched->cpu_usage;
ff5f3bbd 728 sched->runavg_cpu_usage = (sched->runavg_cpu_usage * (sched->replay_repeat - 1) + sched->cpu_usage) / sched->replay_repeat;
ec156764 729
0e9b07e5
ACM		 730 sched->parent_cpu_usage = cpu_usage_1 - cpu_usage_0;
731 if (!sched->runavg_parent_cpu_usage)
732 sched->runavg_parent_cpu_usage = sched->parent_cpu_usage;
ff5f3bbd
YS		 733 sched->runavg_parent_cpu_usage = (sched->runavg_parent_cpu_usage * (sched->replay_repeat - 1) +
734 sched->parent_cpu_usage)/sched->replay_repeat;
ec156764 735
0e9b07e5 736 ret = pthread_mutex_lock(&sched->start_work_mutex);
ec156764
IM		 737 BUG_ON(ret);
738
0e9b07e5
ACM		 739 for (i = 0; i < sched->nr_tasks; i++) {
740 task = sched->tasks[i];
ec156764
IM		 741 sem_init(&task->sleep_sem, 0, 0);
742 task->curr_event = 0;
743 }
744}
745
0e9b07e5 746static void run_one_test(struct perf_sched *sched)
ec156764 747{
fb7d0b3c 748 u64 T0, T1, delta, avg_delta, fluct;
ec156764
IM		 749
750 T0 = get_nsecs();
0e9b07e5 751 wait_for_tasks(sched);
ec156764
IM		 752 T1 = get_nsecs();
753
754 delta = T1 - T0;
0e9b07e5
ACM		 755 sched->sum_runtime += delta;
756 sched->nr_runs++;
ec156764 757
0e9b07e5 758 avg_delta = sched->sum_runtime / sched->nr_runs;
ec156764
IM		 759 if (delta < avg_delta)
760 fluct = avg_delta - delta;
761 else
762 fluct = delta - avg_delta;
0e9b07e5
ACM		 763 sched->sum_fluct += fluct;
764 if (!sched->run_avg)
765 sched->run_avg = delta;
ff5f3bbd 766 sched->run_avg = (sched->run_avg * (sched->replay_repeat - 1) + delta) / sched->replay_repeat;
ec156764 767
4fc76e49 768 printf("#%-3ld: %0.3f, ", sched->nr_runs, (double)delta / NSEC_PER_MSEC);
ec156764 769
4fc76e49 770 printf("ravg: %0.2f, ", (double)sched->run_avg / NSEC_PER_MSEC);
ec156764 771
ad236fd2 772 printf("cpu: %0.2f / %0.2f",
4fc76e49 773 (double)sched->cpu_usage / NSEC_PER_MSEC, (double)sched->runavg_cpu_usage / NSEC_PER_MSEC);
ec156764
IM		 774
775#if 0
776 /*
fbf94829 777 * rusage statistics done by the parent, these are less
0e9b07e5 778 * accurate than the sched->sum_exec_runtime based statistics:
fbf94829 779 */
ad236fd2 780 printf(" [%0.2f / %0.2f]",
4fc76e49
ACM		 781 (double)sched->parent_cpu_usage / NSEC_PER_MSEC,
782 (double)sched->runavg_parent_cpu_usage / NSEC_PER_MSEC);
ec156764
IM		 783#endif
784
ad236fd2 785 printf("\n");
ec156764 786
0e9b07e5
ACM		 787 if (sched->nr_sleep_corrections)
788 printf(" (%ld sleep corrections)\n", sched->nr_sleep_corrections);
789 sched->nr_sleep_corrections = 0;
ec156764
IM		 790}
791
0e9b07e5 792static void test_calibrations(struct perf_sched *sched)
ec156764 793{
b1ffe8f3 794 u64 T0, T1;
ec156764
IM		 795
796 T0 = get_nsecs();
4fc76e49 797 burn_nsecs(sched, NSEC_PER_MSEC);
ec156764
IM		 798 T1 = get_nsecs();
799
9486aa38 800 printf("the run test took %" PRIu64 " nsecs\n", T1 - T0);
ec156764
IM		 801
802 T0 = get_nsecs();
4fc76e49 803 sleep_nsecs(NSEC_PER_MSEC);
ec156764
IM		 804 T1 = get_nsecs();
805
9486aa38 806 printf("the sleep test took %" PRIu64 " nsecs\n", T1 - T0);
ec156764
IM		 807}
808
a116e05d 809static int
0e9b07e5 810replay_wakeup_event(struct perf_sched *sched,
32dcd021 811 struct evsel *evsel, struct perf_sample *sample,
9ec3f4e4 812 struct machine *machine __maybe_unused)
419ab0d6 813{
9ec3f4e4
ACM		 814 const char *comm = perf_evsel__strval(evsel, sample, "comm");
815 const u32 pid = perf_evsel__intval(evsel, sample, "pid");
419ab0d6 816 struct task_desc *waker, *wakee;
fbf94829 817
bb963e16 818 if (verbose > 0) {
2b7fcbc5 819 printf("sched_wakeup event %p\n", evsel);
fbf94829 820
9ec3f4e4 821 printf(" ... pid %d woke up %s/%d\n", sample->tid, comm, pid);
ad236fd2 822 }
fbf94829 823
2b7fcbc5 824 waker = register_pid(sched, sample->tid, "<unknown>");
9ec3f4e4 825 wakee = register_pid(sched, pid, comm);
fbf94829 826
0e9b07e5 827 add_sched_event_wakeup(sched, waker, sample->time, wakee);
a116e05d 828 return 0;
ec156764
IM		 829}
830
9ec3f4e4 831static int replay_switch_event(struct perf_sched *sched,
32dcd021 832 struct evsel *evsel,
9ec3f4e4
ACM		 833 struct perf_sample *sample,
834 struct machine *machine __maybe_unused)
ec156764 835{
9ec3f4e4
ACM		 836 const char *prev_comm = perf_evsel__strval(evsel, sample, "prev_comm"),
837 *next_comm = perf_evsel__strval(evsel, sample, "next_comm");
838 const u32 prev_pid = perf_evsel__intval(evsel, sample, "prev_pid"),
839 next_pid = perf_evsel__intval(evsel, sample, "next_pid");
840 const u64 prev_state = perf_evsel__intval(evsel, sample, "prev_state");
1d037ca1 841 struct task_desc *prev, __maybe_unused *next;
7f7f8d0b
ACM		 842 u64 timestamp0, timestamp = sample->time;
843 int cpu = sample->cpu;
fbf94829
IM		 844 s64 delta;
845
bb963e16 846 if (verbose > 0)
2b7fcbc5 847 printf("sched_switch event %p\n", evsel);
ad236fd2 848
fbf94829 849 if (cpu >= MAX_CPUS || cpu < 0)
a116e05d 850 return 0;
fbf94829 851
0e9b07e5 852 timestamp0 = sched->cpu_last_switched[cpu];
fbf94829
IM		 853 if (timestamp0)
854 delta = timestamp - timestamp0;
855 else
856 delta = 0;
857
a116e05d 858 if (delta < 0) {
60b7d14a 859 pr_err("hm, delta: %" PRIu64 " < 0 ?\n", delta);
a116e05d
ACM		 860 return -1;
861 }
fbf94829 862
9ec3f4e4
ACM		 863 pr_debug(" ... switch from %s/%d to %s/%d [ran %" PRIu64 " nsecs]\n",
864 prev_comm, prev_pid, next_comm, next_pid, delta);
fbf94829 865
9ec3f4e4
ACM		 866 prev = register_pid(sched, prev_pid, prev_comm);
867 next = register_pid(sched, next_pid, next_comm);
fbf94829 868
0e9b07e5 869 sched->cpu_last_switched[cpu] = timestamp;
fbf94829 870
0e9b07e5 871 add_sched_event_run(sched, prev, timestamp, delta);
9ec3f4e4 872 add_sched_event_sleep(sched, prev, timestamp, prev_state);
a116e05d
ACM		 873
874 return 0;
fbf94829
IM		 875}
876
cb627505
DA		 877static int replay_fork_event(struct perf_sched *sched,
878 union perf_event *event,
879 struct machine *machine)
419ab0d6 880{
cb627505
DA		 881 struct thread *child, *parent;
882
314add6b
AH		 883 child = machine__findnew_thread(machine, event->fork.pid,
884 event->fork.tid);
885 parent = machine__findnew_thread(machine, event->fork.ppid,
886 event->fork.ptid);
cb627505
DA		 887
888 if (child == NULL || parent == NULL) {
889 pr_debug("thread does not exist on fork event: child %p, parent %p\n",
890 child, parent);
b91fc39f 891 goto out_put;
cb627505 892 }
9ec3f4e4 893
bb963e16 894 if (verbose > 0) {
cb627505 895 printf("fork event\n");
b9c5143a
FW		 896 printf("... parent: %s/%d\n", thread__comm_str(parent), parent->tid);
897 printf("... child: %s/%d\n", thread__comm_str(child), child->tid);
419ab0d6 898 }
9ec3f4e4 899
b9c5143a
FW		 900 register_pid(sched, parent->tid, thread__comm_str(parent));
901 register_pid(sched, child->tid, thread__comm_str(child));
b91fc39f
ACM		 902out_put:
903 thread__put(child);
904 thread__put(parent);
a116e05d 905 return 0;
419ab0d6 906}
fbf94829 907
b1ffe8f3
IM		 908struct sort_dimension {
909 const char *name;
b5fae128 910 sort_fn_t cmp;
b1ffe8f3
IM		 911 struct list_head list;
912};
913
8640da9f
CD		 914/*
915 * handle runtime stats saved per thread
916 */
917static struct thread_runtime *thread__init_runtime(struct thread *thread)
918{
919 struct thread_runtime *r;
920
921 r = zalloc(sizeof(struct thread_runtime));
922 if (!r)
923 return NULL;
924
925 init_stats(&r->run_stats);
926 thread__set_priv(thread, r);
927
928 return r;
929}
930
931static struct thread_runtime *thread__get_runtime(struct thread *thread)
932{
933 struct thread_runtime *tr;
934
935 tr = thread__priv(thread);
936 if (tr == NULL) {
937 tr = thread__init_runtime(thread);
938 if (tr == NULL)
939 pr_debug("Failed to malloc memory for runtime data.\n");
940 }
941
942 return tr;
943}
944
daa1d7a5 945static int
39aeb52f 946thread_lat_cmp(struct list_head *list, struct work_atoms *l, struct work_atoms *r)
daa1d7a5
FW		 947{
948 struct sort_dimension *sort;
949 int ret = 0;
950
b5fae128
IM		 951 BUG_ON(list_empty(list));
952
daa1d7a5
FW		 953 list_for_each_entry(sort, list, list) {
954 ret = sort->cmp(l, r);
955 if (ret)
956 return ret;
957 }
958
959 return ret;
960}
961
39aeb52f 962static struct work_atoms *
cb4c13a5 963thread_atoms_search(struct rb_root_cached *root, struct thread *thread,
b5fae128
IM		 964 struct list_head *sort_list)
965{
cb4c13a5 966 struct rb_node *node = root->rb_root.rb_node;
39aeb52f 967 struct work_atoms key = { .thread = thread };
b5fae128
IM		 968
969 while (node) {
39aeb52f 970 struct work_atoms *atoms;
b5fae128
IM		 971 int cmp;
972
39aeb52f 973 atoms = container_of(node, struct work_atoms, node);
b5fae128
IM		 974
975 cmp = thread_lat_cmp(sort_list, &key, atoms);
976 if (cmp > 0)
977 node = node->rb_left;
978 else if (cmp < 0)
979 node = node->rb_right;
980 else {
981 BUG_ON(thread != atoms->thread);
982 return atoms;
983 }
984 }
985 return NULL;
986}
987
cdce9d73 988static void
cb4c13a5 989__thread_latency_insert(struct rb_root_cached *root, struct work_atoms *data,
daa1d7a5 990 struct list_head *sort_list)
cdce9d73 991{
cb4c13a5
DB		 992 struct rb_node **new = &(root->rb_root.rb_node), *parent = NULL;
993 bool leftmost = true;
cdce9d73
FW		 994
995 while (*new) {
39aeb52f 996 struct work_atoms *this;
daa1d7a5 997 int cmp;
cdce9d73 998
39aeb52f 999 this = container_of(*new, struct work_atoms, node);
cdce9d73 1000 parent = *new;
daa1d7a5
FW		 1001
1002 cmp = thread_lat_cmp(sort_list, data, this);
1003
1004 if (cmp > 0)
cdce9d73 1005 new = &((*new)->rb_left);
cb4c13a5 1006 else {
daa1d7a5 1007 new = &((*new)->rb_right);
cb4c13a5
DB		 1008 leftmost = false;
1009 }
cdce9d73
FW		 1010 }
1011
1012 rb_link_node(&data->node, parent, new);
cb4c13a5 1013 rb_insert_color_cached(&data->node, root, leftmost);
cdce9d73
FW		 1014}
1015
0e9b07e5 1016static int thread_atoms_insert(struct perf_sched *sched, struct thread *thread)
cdce9d73 1017{
36479484 1018 struct work_atoms *atoms = zalloc(sizeof(*atoms));
a116e05d
ACM		 1019 if (!atoms) {
1020 pr_err("No memory at %s\n", __func__);
1021 return -1;
1022 }
cdce9d73 1023
f3b623b8 1024 atoms->thread = thread__get(thread);
39aeb52f 1025 INIT_LIST_HEAD(&atoms->work_list);
0e9b07e5 1026 __thread_latency_insert(&sched->atom_root, atoms, &sched->cmp_pid);
a116e05d 1027 return 0;
cdce9d73
FW		 1028}
1029
9ec3f4e4 1030static char sched_out_state(u64 prev_state)
cdce9d73
FW		 1031{
1032 const char *str = TASK_STATE_TO_CHAR_STR;
1033
9ec3f4e4 1034 return str[prev_state];
cdce9d73
FW		 1035}
1036
a116e05d 1037static int
39aeb52f 1038add_sched_out_event(struct work_atoms *atoms,
1039 char run_state,
1040 u64 timestamp)
cdce9d73 1041{
36479484 1042 struct work_atom *atom = zalloc(sizeof(*atom));
a116e05d
ACM		 1043 if (!atom) {
1044 pr_err("Non memory at %s", __func__);
1045 return -1;
1046 }
cdce9d73 1047
aa1ab9d2
FW		 1048 atom->sched_out_time = timestamp;
1049
39aeb52f 1050 if (run_state == 'R') {
b1ffe8f3 1051 atom->state = THREAD_WAIT_CPU;
aa1ab9d2 1052 atom->wake_up_time = atom->sched_out_time;
c6ced611
FW		 1053 }
1054
39aeb52f 1055 list_add_tail(&atom->list, &atoms->work_list);
a116e05d 1056 return 0;
cdce9d73
FW		 1057}
1058
1059static void
1d037ca1
IT		 1060add_runtime_event(struct work_atoms *atoms, u64 delta,
1061 u64 timestamp __maybe_unused)
39aeb52f 1062{
1063 struct work_atom *atom;
1064
1065 BUG_ON(list_empty(&atoms->work_list));
1066
1067 atom = list_entry(atoms->work_list.prev, struct work_atom, list);
1068
1069 atom->runtime += delta;
1070 atoms->total_runtime += delta;
1071}
1072
1073static void
1074add_sched_in_event(struct work_atoms *atoms, u64 timestamp)
cdce9d73 1075{
b1ffe8f3 1076 struct work_atom *atom;
66685678 1077 u64 delta;
cdce9d73 1078
39aeb52f 1079 if (list_empty(&atoms->work_list))
cdce9d73
FW		 1080 return;
1081
39aeb52f 1082 atom = list_entry(atoms->work_list.prev, struct work_atom, list);
cdce9d73 1083
b1ffe8f3 1084 if (atom->state != THREAD_WAIT_CPU)
cdce9d73
FW		 1085 return;
1086
b1ffe8f3
IM		 1087 if (timestamp < atom->wake_up_time) {
1088 atom->state = THREAD_IGNORE;
cdce9d73
FW		 1089 return;
1090 }
1091
b1ffe8f3
IM		 1092 atom->state = THREAD_SCHED_IN;
1093 atom->sched_in_time = timestamp;
66685678 1094
b1ffe8f3 1095 delta = atom->sched_in_time - atom->wake_up_time;
66685678 1096 atoms->total_lat += delta;
3786310a 1097 if (delta > atoms->max_lat) {
66685678 1098 atoms->max_lat = delta;
3786310a
FW		 1099 atoms->max_lat_at = timestamp;
1100 }
66685678 1101 atoms->nb_atoms++;
cdce9d73
FW		 1102}
1103
9ec3f4e4 1104static int latency_switch_event(struct perf_sched *sched,
32dcd021 1105 struct evsel *evsel,
9ec3f4e4
ACM		 1106 struct perf_sample *sample,
1107 struct machine *machine)
cdce9d73 1108{
9ec3f4e4
ACM		 1109 const u32 prev_pid = perf_evsel__intval(evsel, sample, "prev_pid"),
1110 next_pid = perf_evsel__intval(evsel, sample, "next_pid");
1111 const u64 prev_state = perf_evsel__intval(evsel, sample, "prev_state");
39aeb52f 1112 struct work_atoms *out_events, *in_events;
cdce9d73 1113 struct thread *sched_out, *sched_in;
7f7f8d0b 1114 u64 timestamp0, timestamp = sample->time;
b91fc39f 1115 int cpu = sample->cpu, err = -1;
ea92ed5a
IM		 1116 s64 delta;
1117
39aeb52f 1118 BUG_ON(cpu >= MAX_CPUS || cpu < 0);
ea92ed5a 1119
0e9b07e5
ACM		 1120 timestamp0 = sched->cpu_last_switched[cpu];
1121 sched->cpu_last_switched[cpu] = timestamp;
ea92ed5a
IM		 1122 if (timestamp0)
1123 delta = timestamp - timestamp0;
1124 else
1125 delta = 0;
1126
a116e05d
ACM		 1127 if (delta < 0) {
1128 pr_err("hm, delta: %" PRIu64 " < 0 ?\n", delta);
1129 return -1;
1130 }
cdce9d73 1131
1fcb8768
AH		 1132 sched_out = machine__findnew_thread(machine, -1, prev_pid);
1133 sched_in = machine__findnew_thread(machine, -1, next_pid);
b91fc39f
ACM		 1134 if (sched_out == NULL || sched_in == NULL)
1135 goto out_put;
cdce9d73 1136
0e9b07e5 1137 out_events = thread_atoms_search(&sched->atom_root, sched_out, &sched->cmp_pid);
39aeb52f 1138 if (!out_events) {
0e9b07e5 1139 if (thread_atoms_insert(sched, sched_out))
b91fc39f 1140 goto out_put;
0e9b07e5 1141 out_events = thread_atoms_search(&sched->atom_root, sched_out, &sched->cmp_pid);
a116e05d
ACM		 1142 if (!out_events) {
1143 pr_err("out-event: Internal tree error");
b91fc39f 1144 goto out_put;
a116e05d 1145 }
39aeb52f 1146 }
9ec3f4e4 1147 if (add_sched_out_event(out_events, sched_out_state(prev_state), timestamp))
a116e05d 1148 return -1;
39aeb52f 1149
0e9b07e5 1150 in_events = thread_atoms_search(&sched->atom_root, sched_in, &sched->cmp_pid);
39aeb52f 1151 if (!in_events) {
0e9b07e5 1152 if (thread_atoms_insert(sched, sched_in))
b91fc39f 1153 goto out_put;
0e9b07e5 1154 in_events = thread_atoms_search(&sched->atom_root, sched_in, &sched->cmp_pid);
a116e05d
ACM		 1155 if (!in_events) {
1156 pr_err("in-event: Internal tree error");
b91fc39f 1157 goto out_put;
a116e05d 1158 }
39aeb52f 1159 /*
1160 * Take came in we have not heard about yet,
1161 * add in an initial atom in runnable state:
1162 */
a116e05d 1163 if (add_sched_out_event(in_events, 'R', timestamp))
b91fc39f 1164 goto out_put;
cdce9d73 1165 }
39aeb52f 1166 add_sched_in_event(in_events, timestamp);
b91fc39f
ACM		 1167 err = 0;
1168out_put:
1169 thread__put(sched_out);
1170 thread__put(sched_in);
1171 return err;
39aeb52f 1172}
cdce9d73 1173
9ec3f4e4 1174static int latency_runtime_event(struct perf_sched *sched,
32dcd021 1175 struct evsel *evsel,
9ec3f4e4
ACM		 1176 struct perf_sample *sample,
1177 struct machine *machine)
39aeb52f 1178{
9ec3f4e4
ACM		 1179 const u32 pid = perf_evsel__intval(evsel, sample, "pid");
1180 const u64 runtime = perf_evsel__intval(evsel, sample, "runtime");
1fcb8768 1181 struct thread *thread = machine__findnew_thread(machine, -1, pid);
0e9b07e5 1182 struct work_atoms *atoms = thread_atoms_search(&sched->atom_root, thread, &sched->cmp_pid);
7f7f8d0b 1183 u64 timestamp = sample->time;
b91fc39f
ACM		 1184 int cpu = sample->cpu, err = -1;
1185
1186 if (thread == NULL)
1187 return -1;
39aeb52f 1188
1189 BUG_ON(cpu >= MAX_CPUS || cpu < 0);
39aeb52f 1190 if (!atoms) {
0e9b07e5 1191 if (thread_atoms_insert(sched, thread))
b91fc39f 1192 goto out_put;
0e9b07e5 1193 atoms = thread_atoms_search(&sched->atom_root, thread, &sched->cmp_pid);
a116e05d 1194 if (!atoms) {
60b7d14a 1195 pr_err("in-event: Internal tree error");
b91fc39f 1196 goto out_put;
a116e05d
ACM		 1197 }
1198 if (add_sched_out_event(atoms, 'R', timestamp))
b91fc39f 1199 goto out_put;
cdce9d73
FW		 1200 }
1201
9ec3f4e4 1202 add_runtime_event(atoms, runtime, timestamp);
b91fc39f
ACM		 1203 err = 0;
1204out_put:
1205 thread__put(thread);
1206 return err;
cdce9d73
FW		 1207}
1208
9ec3f4e4 1209static int latency_wakeup_event(struct perf_sched *sched,
32dcd021 1210 struct evsel *evsel,
9ec3f4e4
ACM		 1211 struct perf_sample *sample,
1212 struct machine *machine)
cdce9d73 1213{
0680ee7d 1214 const u32 pid = perf_evsel__intval(evsel, sample, "pid");
39aeb52f 1215 struct work_atoms *atoms;
b1ffe8f3 1216 struct work_atom *atom;
cdce9d73 1217 struct thread *wakee;
7f7f8d0b 1218 u64 timestamp = sample->time;
b91fc39f 1219 int err = -1;
cdce9d73 1220
1fcb8768 1221 wakee = machine__findnew_thread(machine, -1, pid);
b91fc39f
ACM		 1222 if (wakee == NULL)
1223 return -1;
0e9b07e5 1224 atoms = thread_atoms_search(&sched->atom_root, wakee, &sched->cmp_pid);
17562205 1225 if (!atoms) {
0e9b07e5 1226 if (thread_atoms_insert(sched, wakee))
b91fc39f 1227 goto out_put;
0e9b07e5 1228 atoms = thread_atoms_search(&sched->atom_root, wakee, &sched->cmp_pid);
a116e05d 1229 if (!atoms) {
60b7d14a 1230 pr_err("wakeup-event: Internal tree error");
b91fc39f 1231 goto out_put;
a116e05d
ACM		 1232 }
1233 if (add_sched_out_event(atoms, 'S', timestamp))
b91fc39f 1234 goto out_put;
cdce9d73
FW		 1235 }
1236
39aeb52f 1237 BUG_ON(list_empty(&atoms->work_list));
cdce9d73 1238
39aeb52f 1239 atom = list_entry(atoms->work_list.prev, struct work_atom, list);
cdce9d73 1240
55ffb7a6 1241 /*
67d6259d
DY		 1242 * As we do not guarantee the wakeup event happens when
1243 * task is out of run queue, also may happen when task is
1244 * on run queue and wakeup only change ->state to TASK_RUNNING,
1245 * then we should not set the ->wake_up_time when wake up a
1246 * task which is on run queue.
1247 *
55ffb7a6
MG		 1248 * You WILL be missing events if you've recorded only
1249 * one CPU, or are only looking at only one, so don't
67d6259d 1250 * skip in this case.
55ffb7a6 1251 */
0e9b07e5 1252 if (sched->profile_cpu == -1 && atom->state != THREAD_SLEEPING)
b91fc39f 1253 goto out_ok;
cdce9d73 1254
0e9b07e5 1255 sched->nr_timestamps++;
ea57c4f5 1256 if (atom->sched_out_time > timestamp) {
0e9b07e5 1257 sched->nr_unordered_timestamps++;
b91fc39f 1258 goto out_ok;
ea57c4f5 1259 }
aa1ab9d2 1260
b1ffe8f3
IM		 1261 atom->state = THREAD_WAIT_CPU;
1262 atom->wake_up_time = timestamp;
b91fc39f
ACM		 1263out_ok:
1264 err = 0;
1265out_put:
1266 thread__put(wakee);
1267 return err;
cdce9d73
FW		 1268}
1269
9ec3f4e4 1270static int latency_migrate_task_event(struct perf_sched *sched,
32dcd021 1271 struct evsel *evsel,
9ec3f4e4
ACM		 1272 struct perf_sample *sample,
1273 struct machine *machine)
55ffb7a6 1274{
9ec3f4e4 1275 const u32 pid = perf_evsel__intval(evsel, sample, "pid");
7f7f8d0b 1276 u64 timestamp = sample->time;
55ffb7a6
MG		 1277 struct work_atoms *atoms;
1278 struct work_atom *atom;
1279 struct thread *migrant;
b91fc39f 1280 int err = -1;
55ffb7a6
MG		 1281
1282 /*
1283 * Only need to worry about migration when profiling one CPU.
1284 */
0e9b07e5 1285 if (sched->profile_cpu == -1)
a116e05d 1286 return 0;
55ffb7a6 1287
1fcb8768 1288 migrant = machine__findnew_thread(machine, -1, pid);
b91fc39f
ACM		 1289 if (migrant == NULL)
1290 return -1;
0e9b07e5 1291 atoms = thread_atoms_search(&sched->atom_root, migrant, &sched->cmp_pid);
55ffb7a6 1292 if (!atoms) {
0e9b07e5 1293 if (thread_atoms_insert(sched, migrant))
b91fc39f 1294 goto out_put;
b9c5143a 1295 register_pid(sched, migrant->tid, thread__comm_str(migrant));
0e9b07e5 1296 atoms = thread_atoms_search(&sched->atom_root, migrant, &sched->cmp_pid);
a116e05d 1297 if (!atoms) {
60b7d14a 1298 pr_err("migration-event: Internal tree error");
b91fc39f 1299 goto out_put;
a116e05d
ACM		 1300 }
1301 if (add_sched_out_event(atoms, 'R', timestamp))
b91fc39f 1302 goto out_put;
55ffb7a6
MG		 1303 }
1304
1305 BUG_ON(list_empty(&atoms->work_list));
1306
1307 atom = list_entry(atoms->work_list.prev, struct work_atom, list);
1308 atom->sched_in_time = atom->sched_out_time = atom->wake_up_time = timestamp;
1309
0e9b07e5 1310 sched->nr_timestamps++;
55ffb7a6
MG		 1311
1312 if (atom->sched_out_time > timestamp)
0e9b07e5 1313 sched->nr_unordered_timestamps++;
b91fc39f
ACM		 1314 err = 0;
1315out_put:
1316 thread__put(migrant);
1317 return err;
55ffb7a6
MG		 1318}
1319
0e9b07e5 1320static void output_lat_thread(struct perf_sched *sched, struct work_atoms *work_list)
cdce9d73 1321{
cdce9d73
FW		 1322 int i;
1323 int ret;
66685678 1324 u64 avg;
99620a5d 1325 char max_lat_at[32];
cdce9d73 1326
39aeb52f 1327 if (!work_list->nb_atoms)
cdce9d73 1328 return;
ea57c4f5
IM		 1329 /*
1330 * Ignore idle threads:
1331 */
b9c5143a 1332 if (!strcmp(thread__comm_str(work_list->thread), "swapper"))
ea57c4f5 1333 return;
cdce9d73 1334
0e9b07e5
ACM		 1335 sched->all_runtime += work_list->total_runtime;
1336 sched->all_count += work_list->nb_atoms;
66685678 1337
2f80dd44
JB		 1338 if (work_list->num_merged > 1)
1339 ret = printf(" %s:(%d) ", thread__comm_str(work_list->thread), work_list->num_merged);
1340 else
1341 ret = printf(" %s:%d ", thread__comm_str(work_list->thread), work_list->thread->tid);
cdce9d73 1342
08f69e6c 1343 for (i = 0; i < 24 - ret; i++)
cdce9d73
FW		 1344 printf(" ");
1345
39aeb52f 1346 avg = work_list->total_lat / work_list->nb_atoms;
99620a5d 1347 timestamp__scnprintf_usec(work_list->max_lat_at, max_lat_at, sizeof(max_lat_at));
cdce9d73 1348
99620a5d 1349 printf("|%11.3f ms |%9" PRIu64 " | avg:%9.3f ms | max:%9.3f ms | max at: %13s s\n",
4fc76e49
ACM		 1350 (double)work_list->total_runtime / NSEC_PER_MSEC,
1351 work_list->nb_atoms, (double)avg / NSEC_PER_MSEC,
1352 (double)work_list->max_lat / NSEC_PER_MSEC,
99620a5d 1353 max_lat_at);
cdce9d73
FW		 1354}
1355
39aeb52f 1356static int pid_cmp(struct work_atoms *l, struct work_atoms *r)
daa1d7a5 1357{
0014de17
JO		 1358 if (l->thread == r->thread)
1359 return 0;
38051234 1360 if (l->thread->tid < r->thread->tid)
daa1d7a5 1361 return -1;
38051234 1362 if (l->thread->tid > r->thread->tid)
daa1d7a5 1363 return 1;
0014de17 1364 return (int)(l->thread - r->thread);
daa1d7a5
FW		 1365}
1366
39aeb52f 1367static int avg_cmp(struct work_atoms *l, struct work_atoms *r)
daa1d7a5
FW		 1368{
1369 u64 avgl, avgr;
1370
1371 if (!l->nb_atoms)
1372 return -1;
1373
1374 if (!r->nb_atoms)
1375 return 1;
1376
1377 avgl = l->total_lat / l->nb_atoms;
1378 avgr = r->total_lat / r->nb_atoms;
1379
1380 if (avgl < avgr)
1381 return -1;
1382 if (avgl > avgr)
1383 return 1;
1384
1385 return 0;
1386}
1387
39aeb52f 1388static int max_cmp(struct work_atoms *l, struct work_atoms *r)
daa1d7a5
FW		 1389{
1390 if (l->max_lat < r->max_lat)
1391 return -1;
1392 if (l->max_lat > r->max_lat)
1393 return 1;
1394
1395 return 0;
1396}
1397
39aeb52f 1398static int switch_cmp(struct work_atoms *l, struct work_atoms *r)
daa1d7a5
FW		 1399{
1400 if (l->nb_atoms < r->nb_atoms)
1401 return -1;
1402 if (l->nb_atoms > r->nb_atoms)
1403 return 1;
1404
1405 return 0;
1406}
1407
39aeb52f 1408static int runtime_cmp(struct work_atoms *l, struct work_atoms *r)
daa1d7a5
FW		 1409{
1410 if (l->total_runtime < r->total_runtime)
1411 return -1;
1412 if (l->total_runtime > r->total_runtime)
1413 return 1;
1414
1415 return 0;
1416}
1417
cbef79a8 1418static int sort_dimension__add(const char *tok, struct list_head *list)
daa1d7a5 1419{
0e9b07e5
ACM		 1420 size_t i;
1421 static struct sort_dimension avg_sort_dimension = {
1422 .name = "avg",
1423 .cmp = avg_cmp,
1424 };
1425 static struct sort_dimension max_sort_dimension = {
1426 .name = "max",
1427 .cmp = max_cmp,
1428 };
1429 static struct sort_dimension pid_sort_dimension = {
1430 .name = "pid",
1431 .cmp = pid_cmp,
1432 };
1433 static struct sort_dimension runtime_sort_dimension = {
1434 .name = "runtime",
1435 .cmp = runtime_cmp,
1436 };
1437 static struct sort_dimension switch_sort_dimension = {
1438 .name = "switch",
1439 .cmp = switch_cmp,
1440 };
1441 struct sort_dimension *available_sorts[] = {
1442 &pid_sort_dimension,
1443 &avg_sort_dimension,
1444 &max_sort_dimension,
1445 &switch_sort_dimension,
1446 &runtime_sort_dimension,
1447 };
daa1d7a5 1448
0e9b07e5 1449 for (i = 0; i < ARRAY_SIZE(available_sorts); i++) {
daa1d7a5
FW		 1450 if (!strcmp(available_sorts[i]->name, tok)) {
1451 list_add_tail(&available_sorts[i]->list, list);
1452
1453 return 0;
1454 }
1455 }
1456
1457 return -1;
1458}
1459
0e9b07e5 1460static void perf_sched__sort_lat(struct perf_sched *sched)
daa1d7a5
FW		 1461{
1462 struct rb_node *node;
cb4c13a5 1463 struct rb_root_cached *root = &sched->atom_root;
2f80dd44 1464again:
daa1d7a5 1465 for (;;) {
39aeb52f 1466 struct work_atoms *data;
cb4c13a5 1467 node = rb_first_cached(root);
daa1d7a5
FW		 1468 if (!node)
1469 break;
1470
cb4c13a5 1471 rb_erase_cached(node, root);
39aeb52f 1472 data = rb_entry(node, struct work_atoms, node);
0e9b07e5 1473 __thread_latency_insert(&sched->sorted_atom_root, data, &sched->sort_list);
daa1d7a5 1474 }
2f80dd44
JB		 1475 if (root == &sched->atom_root) {
1476 root = &sched->merged_atom_root;
1477 goto again;
1478 }
daa1d7a5
FW		 1479}
1480
0e9b07e5 1481static int process_sched_wakeup_event(struct perf_tool *tool,
32dcd021 1482 struct evsel *evsel,
1d037ca1 1483 struct perf_sample *sample,
4218e673 1484 struct machine *machine)
419ab0d6 1485{
0e9b07e5 1486 struct perf_sched *sched = container_of(tool, struct perf_sched, tool);
419ab0d6 1487
9ec3f4e4
ACM		 1488 if (sched->tp_handler->wakeup_event)
1489 return sched->tp_handler->wakeup_event(sched, evsel, sample, machine);
a116e05d 1490
2b7fcbc5 1491 return 0;
419ab0d6
FW		 1492}
1493
a151a37a
JO		 1494union map_priv {
1495 void *ptr;
1496 bool color;
1497};
1498
1499static bool thread__has_color(struct thread *thread)
1500{
1501 union map_priv priv = {
1502 .ptr = thread__priv(thread),
1503 };
1504
1505 return priv.color;
1506}
1507
1508static struct thread*
1509map__findnew_thread(struct perf_sched *sched, struct machine *machine, pid_t pid, pid_t tid)
1510{
1511 struct thread *thread = machine__findnew_thread(machine, pid, tid);
1512 union map_priv priv = {
1513 .color = false,
1514 };
1515
1516 if (!sched->map.color_pids || !thread || thread__priv(thread))
1517 return thread;
1518
1519 if (thread_map__has(sched->map.color_pids, tid))
1520 priv.color = true;
1521
1522 thread__set_priv(thread, priv.ptr);
1523 return thread;
1524}
1525
32dcd021 1526static int map_switch_event(struct perf_sched *sched, struct evsel *evsel,
9ec3f4e4 1527 struct perf_sample *sample, struct machine *machine)
0ec04e16 1528{
9d372ca5
DY		 1529 const u32 next_pid = perf_evsel__intval(evsel, sample, "next_pid");
1530 struct thread *sched_in;
8640da9f 1531 struct thread_runtime *tr;
0ec04e16 1532 int new_shortname;
7f7f8d0b 1533 u64 timestamp0, timestamp = sample->time;
0ec04e16 1534 s64 delta;
99623c62
JO		 1535 int i, this_cpu = sample->cpu;
1536 int cpus_nr;
1537 bool new_cpu = false;
8cd91195 1538 const char *color = PERF_COLOR_NORMAL;
99620a5d 1539 char stimestamp[32];
0ec04e16
IM		 1540
1541 BUG_ON(this_cpu >= MAX_CPUS || this_cpu < 0);
1542
0e9b07e5
ACM		 1543 if (this_cpu > sched->max_cpu)
1544 sched->max_cpu = this_cpu;
0ec04e16 1545
99623c62
JO		 1546 if (sched->map.comp) {
1547 cpus_nr = bitmap_weight(sched->map.comp_cpus_mask, MAX_CPUS);
1548 if (!test_and_set_bit(this_cpu, sched->map.comp_cpus_mask)) {
1549 sched->map.comp_cpus[cpus_nr++] = this_cpu;
1550 new_cpu = true;
1551 }
1552 } else
1553 cpus_nr = sched->max_cpu;
1554
0e9b07e5
ACM		 1555 timestamp0 = sched->cpu_last_switched[this_cpu];
1556 sched->cpu_last_switched[this_cpu] = timestamp;
0ec04e16
IM		 1557 if (timestamp0)
1558 delta = timestamp - timestamp0;
1559 else
1560 delta = 0;
1561
a116e05d 1562 if (delta < 0) {
60b7d14a 1563 pr_err("hm, delta: %" PRIu64 " < 0 ?\n", delta);
a116e05d
ACM		 1564 return -1;
1565 }
0ec04e16 1566
a151a37a 1567 sched_in = map__findnew_thread(sched, machine, -1, next_pid);
b91fc39f
ACM		 1568 if (sched_in == NULL)
1569 return -1;
0ec04e16 1570
8640da9f
CD		 1571 tr = thread__get_runtime(sched_in);
1572 if (tr == NULL) {
1573 thread__put(sched_in);
1574 return -1;
1575 }
1576
b91fc39f 1577 sched->curr_thread[this_cpu] = thread__get(sched_in);
0ec04e16
IM		 1578
1579 printf(" ");
1580
1581 new_shortname = 0;
8640da9f 1582 if (!tr->shortname[0]) {
6bcab4e1
D		 1583 if (!strcmp(thread__comm_str(sched_in), "swapper")) {
1584 /*
1585 * Don't allocate a letter-number for swapper:0
1586 * as a shortname. Instead, we use '.' for it.
1587 */
8640da9f
CD		 1588 tr->shortname[0] = '.';
1589 tr->shortname[1] = ' ';
0ec04e16 1590 } else {
8640da9f
CD		 1591 tr->shortname[0] = sched->next_shortname1;
1592 tr->shortname[1] = sched->next_shortname2;
6bcab4e1
D		 1593
1594 if (sched->next_shortname1 < 'Z') {
1595 sched->next_shortname1++;
0ec04e16 1596 } else {
6bcab4e1
D		 1597 sched->next_shortname1 = 'A';
1598 if (sched->next_shortname2 < '9')
1599 sched->next_shortname2++;
1600 else
1601 sched->next_shortname2 = '0';
0ec04e16
IM		 1602 }
1603 }
1604 new_shortname = 1;
1605 }
1606
99623c62
JO		 1607 for (i = 0; i < cpus_nr; i++) {
1608 int cpu = sched->map.comp ? sched->map.comp_cpus[i] : i;
a151a37a 1609 struct thread *curr_thread = sched->curr_thread[cpu];
8640da9f 1610 struct thread_runtime *curr_tr;
a151a37a 1611 const char *pid_color = color;
cf294f24 1612 const char *cpu_color = color;
a151a37a
JO		 1613
1614 if (curr_thread && thread__has_color(curr_thread))
1615 pid_color = COLOR_PIDS;
99623c62 1616
73643bb6
JO		 1617 if (sched->map.cpus && !cpu_map__has(sched->map.cpus, cpu))
1618 continue;
1619
cf294f24
JO		 1620 if (sched->map.color_cpus && cpu_map__has(sched->map.color_cpus, cpu))
1621 cpu_color = COLOR_CPUS;
1622
0ec04e16 1623 if (cpu != this_cpu)
1208bb27 1624 color_fprintf(stdout, color, " ");
0ec04e16 1625 else
cf294f24 1626 color_fprintf(stdout, cpu_color, "*");
0ec04e16 1627
8640da9f
CD		 1628 if (sched->curr_thread[cpu]) {
1629 curr_tr = thread__get_runtime(sched->curr_thread[cpu]);
1630 if (curr_tr == NULL) {
1631 thread__put(sched_in);
1632 return -1;
1633 }
1634 color_fprintf(stdout, pid_color, "%2s ", curr_tr->shortname);
1635 } else
8cd91195 1636 color_fprintf(stdout, color, " ");
0ec04e16
IM		 1637 }
1638
73643bb6
JO		 1639 if (sched->map.cpus && !cpu_map__has(sched->map.cpus, this_cpu))
1640 goto out;
1641
99620a5d
NK		 1642 timestamp__scnprintf_usec(timestamp, stimestamp, sizeof(stimestamp));
1643 color_fprintf(stdout, color, " %12s secs ", stimestamp);
99a3c3a9 1644 if (new_shortname || tr->comm_changed || (verbose > 0 && sched_in->tid)) {
a151a37a
JO		 1645 const char *pid_color = color;
1646
1647 if (thread__has_color(sched_in))
1648 pid_color = COLOR_PIDS;
1649
1650 color_fprintf(stdout, pid_color, "%s => %s:%d",
8640da9f 1651 tr->shortname, thread__comm_str(sched_in), sched_in->tid);
99a3c3a9 1652 tr->comm_changed = false;
0ec04e16 1653 }
a116e05d 1654
99623c62 1655 if (sched->map.comp && new_cpu)
8cd91195 1656 color_fprintf(stdout, color, " (CPU %d)", this_cpu);
99623c62 1657
73643bb6 1658out:
8cd91195 1659 color_fprintf(stdout, color, "\n");
99623c62 1660
b91fc39f
ACM		 1661 thread__put(sched_in);
1662
a116e05d 1663 return 0;
0ec04e16
IM		 1664}
1665
0e9b07e5 1666static int process_sched_switch_event(struct perf_tool *tool,
32dcd021 1667 struct evsel *evsel,
1d037ca1 1668 struct perf_sample *sample,
4218e673 1669 struct machine *machine)
419ab0d6 1670{
0e9b07e5 1671 struct perf_sched *sched = container_of(tool, struct perf_sched, tool);
a116e05d 1672 int this_cpu = sample->cpu, err = 0;
2b7fcbc5
ACM		 1673 u32 prev_pid = perf_evsel__intval(evsel, sample, "prev_pid"),
1674 next_pid = perf_evsel__intval(evsel, sample, "next_pid");
419ab0d6 1675
0e9b07e5 1676 if (sched->curr_pid[this_cpu] != (u32)-1) {
c8a37751
IM		 1677 /*
1678 * Are we trying to switch away a PID that is
1679 * not current?
1680 */
2b7fcbc5 1681 if (sched->curr_pid[this_cpu] != prev_pid)
0e9b07e5 1682 sched->nr_context_switch_bugs++;
c8a37751 1683 }
c8a37751 1684
9ec3f4e4
ACM		 1685 if (sched->tp_handler->switch_event)
1686 err = sched->tp_handler->switch_event(sched, evsel, sample, machine);
2b7fcbc5
ACM		 1687
1688 sched->curr_pid[this_cpu] = next_pid;
a116e05d 1689 return err;
419ab0d6
FW		 1690}
1691
0e9b07e5 1692static int process_sched_runtime_event(struct perf_tool *tool,
32dcd021 1693 struct evsel *evsel,
1d037ca1 1694 struct perf_sample *sample,
4218e673 1695 struct machine *machine)
39aeb52f 1696{
0e9b07e5 1697 struct perf_sched *sched = container_of(tool, struct perf_sched, tool);
39aeb52f 1698
9ec3f4e4
ACM		 1699 if (sched->tp_handler->runtime_event)
1700 return sched->tp_handler->runtime_event(sched, evsel, sample, machine);
a116e05d 1701
2b7fcbc5 1702 return 0;
39aeb52f 1703}
1704
cb627505
DA		 1705static int perf_sched__process_fork_event(struct perf_tool *tool,
1706 union perf_event *event,
1707 struct perf_sample *sample,
1708 struct machine *machine)
fbf94829 1709{
0e9b07e5 1710 struct perf_sched *sched = container_of(tool, struct perf_sched, tool);
46538818 1711
cb627505
DA		 1712 /* run the fork event through the perf machineruy */
1713 perf_event__process_fork(tool, event, sample, machine);
1714
1715 /* and then run additional processing needed for this command */
9ec3f4e4 1716 if (sched->tp_handler->fork_event)
cb627505 1717 return sched->tp_handler->fork_event(sched, event, machine);
a116e05d 1718
2b7fcbc5 1719 return 0;
fbf94829
IM		 1720}
1721
0e9b07e5 1722static int process_sched_migrate_task_event(struct perf_tool *tool,
32dcd021 1723 struct evsel *evsel,
1d037ca1 1724 struct perf_sample *sample,
4218e673 1725 struct machine *machine)
55ffb7a6 1726{
0e9b07e5 1727 struct perf_sched *sched = container_of(tool, struct perf_sched, tool);
55ffb7a6 1728
9ec3f4e4
ACM		 1729 if (sched->tp_handler->migrate_task_event)
1730 return sched->tp_handler->migrate_task_event(sched, evsel, sample, machine);
a116e05d 1731
2b7fcbc5 1732 return 0;
55ffb7a6
MG		 1733}
1734
a116e05d 1735typedef int (*tracepoint_handler)(struct perf_tool *tool,
32dcd021 1736 struct evsel *evsel,
a116e05d 1737 struct perf_sample *sample,
4218e673 1738 struct machine *machine);
ec156764 1739
1d037ca1
IT		 1740static int perf_sched__process_tracepoint_sample(struct perf_tool *tool __maybe_unused,
1741 union perf_event *event __maybe_unused,
ee29be62 1742 struct perf_sample *sample,
32dcd021 1743 struct evsel *evsel,
ee29be62 1744 struct machine *machine)
0a02ad93 1745{
a116e05d 1746 int err = 0;
0a02ad93 1747
744a9719
ACM		 1748 if (evsel->handler != NULL) {
1749 tracepoint_handler f = evsel->handler;
2b7fcbc5 1750 err = f(tool, evsel, sample, machine);
ee29be62 1751 }
0a02ad93 1752
a116e05d 1753 return err;
0a02ad93
IM		 1754}
1755
99a3c3a9
CD		 1756static int perf_sched__process_comm(struct perf_tool *tool __maybe_unused,
1757 union perf_event *event,
1758 struct perf_sample *sample,
1759 struct machine *machine)
1760{
1761 struct thread *thread;
1762 struct thread_runtime *tr;
1763 int err;
1764
1765 err = perf_event__process_comm(tool, event, sample, machine);
1766 if (err)
1767 return err;
1768
1769 thread = machine__find_thread(machine, sample->pid, sample->tid);
1770 if (!thread) {
1771 pr_err("Internal error: can't find thread\n");
1772 return -1;
1773 }
1774
1775 tr = thread__get_runtime(thread);
1776 if (tr == NULL) {
1777 thread__put(thread);
1778 return -1;
1779 }
1780
1781 tr->comm_changed = true;
1782 thread__put(thread);
1783
1784 return 0;
1785}
1786
ae536acf 1787static int perf_sched__read_events(struct perf_sched *sched)
0a02ad93 1788{
32dcd021 1789 const struct evsel_str_handler handlers[] = {
ee29be62
ACM		 1790 { "sched:sched_switch", process_sched_switch_event, },
1791 { "sched:sched_stat_runtime", process_sched_runtime_event, },
1792 { "sched:sched_wakeup", process_sched_wakeup_event, },
1793 { "sched:sched_wakeup_new", process_sched_wakeup_event, },
ee29be62
ACM		 1794 { "sched:sched_migrate_task", process_sched_migrate_task_event, },
1795 };
da378962 1796 struct perf_session *session;
8ceb41d7 1797 struct perf_data data = {
2d4f2799
JO		 1798 .path = input_name,
1799 .mode = PERF_DATA_MODE_READ,
1800 .force = sched->force,
f5fc1412 1801 };
ae536acf 1802 int rc = -1;
da378962 1803
8ceb41d7 1804 session = perf_session__new(&data, false, &sched->tool);
6ef81c55
MI		 1805 if (IS_ERR(session)) {
1806 pr_debug("Error creating perf session");
1807 return PTR_ERR(session);
a116e05d 1808 }
94c744b6 1809
0a7e6d1b 1810 symbol__init(&session->header.env);
04934106 1811
a116e05d
ACM		 1812 if (perf_session__set_tracepoints_handlers(session, handlers))
1813 goto out_delete;
ee29be62 1814
cee75ac7 1815 if (perf_session__has_traces(session, "record -R")) {
b7b61cbe 1816 int err = perf_session__process_events(session);
a116e05d
ACM		 1817 if (err) {
1818 pr_err("Failed to process events, error %d", err);
1819 goto out_delete;
1820 }
4c09bafa 1821
75be989a
ACM		 1822 sched->nr_events = session->evlist->stats.nr_events[0];
1823 sched->nr_lost_events = session->evlist->stats.total_lost;
1824 sched->nr_lost_chunks = session->evlist->stats.nr_events[PERF_RECORD_LOST];
cee75ac7 1825 }
d549c769 1826
ae536acf 1827 rc = 0;
a116e05d
ACM		 1828out_delete:
1829 perf_session__delete(session);
ae536acf 1830 return rc;
0a02ad93
IM		 1831}
1832
49394a2a
DA		 1833/*
1834 * scheduling times are printed as msec.usec
1835 */
1836static inline void print_sched_time(unsigned long long nsecs, int width)
1837{
1838 unsigned long msecs;
1839 unsigned long usecs;
1840
1841 msecs = nsecs / NSEC_PER_MSEC;
1842 nsecs -= msecs * NSEC_PER_MSEC;
1843 usecs = nsecs / NSEC_PER_USEC;
1844 printf("%*lu.%03lu ", width, msecs, usecs);
1845}
1846
1847/*
1848 * returns runtime data for event, allocating memory for it the
1849 * first time it is used.
1850 */
32dcd021 1851static struct evsel_runtime *perf_evsel__get_runtime(struct evsel *evsel)
49394a2a
DA		 1852{
1853 struct evsel_runtime *r = evsel->priv;
1854
1855 if (r == NULL) {
1856 r = zalloc(sizeof(struct evsel_runtime));
1857 evsel->priv = r;
1858 }
1859
1860 return r;
1861}
1862
1863/*
1864 * save last time event was seen per cpu
1865 */
32dcd021 1866static void perf_evsel__save_time(struct evsel *evsel,
49394a2a
DA		 1867 u64 timestamp, u32 cpu)
1868{
1869 struct evsel_runtime *r = perf_evsel__get_runtime(evsel);
1870
1871 if (r == NULL)
1872 return;
1873
1874 if ((cpu >= r->ncpu) || (r->last_time == NULL)) {
1875 int i, n = __roundup_pow_of_two(cpu+1);
1876 void *p = r->last_time;
1877
1878 p = realloc(r->last_time, n * sizeof(u64));
1879 if (!p)
1880 return;
1881
1882 r->last_time = p;
1883 for (i = r->ncpu; i < n; ++i)
1884 r->last_time[i] = (u64) 0;
1885
1886 r->ncpu = n;
1887 }
1888
1889 r->last_time[cpu] = timestamp;
1890}
1891
1892/* returns last time this event was seen on the given cpu */
32dcd021 1893static u64 perf_evsel__get_time(struct evsel *evsel, u32 cpu)
49394a2a
DA		 1894{
1895 struct evsel_runtime *r = perf_evsel__get_runtime(evsel);
1896
1897 if ((r == NULL) || (r->last_time == NULL) || (cpu >= r->ncpu))
1898 return 0;
1899
1900 return r->last_time[cpu];
1901}
1902
9b8087d7 1903static int comm_width = 30;
49394a2a
DA		 1904
1905static char *timehist_get_commstr(struct thread *thread)
1906{
1907 static char str[32];
1908 const char *comm = thread__comm_str(thread);
1909 pid_t tid = thread->tid;
1910 pid_t pid = thread->pid_;
1911 int n;
1912
1913 if (pid == 0)
1914 n = scnprintf(str, sizeof(str), "%s", comm);
1915
1916 else if (tid != pid)
1917 n = scnprintf(str, sizeof(str), "%s[%d/%d]", comm, tid, pid);
1918
1919 else
1920 n = scnprintf(str, sizeof(str), "%s[%d]", comm, tid);
1921
1922 if (n > comm_width)
1923 comm_width = n;
1924
1925 return str;
1926}
1927
a407b067 1928static void timehist_header(struct perf_sched *sched)
49394a2a 1929{
a407b067
DA		 1930 u32 ncpus = sched->max_cpu + 1;
1931 u32 i, j;
1932
49394a2a
DA		 1933 printf("%15s %6s ", "time", "cpu");
1934
a407b067
DA		 1935 if (sched->show_cpu_visual) {
1936 printf(" ");
1937 for (i = 0, j = 0; i < ncpus; ++i) {
1938 printf("%x", j++);
1939 if (j > 15)
1940 j = 0;
1941 }
1942 printf(" ");
1943 }
1944
0e6758e8 1945 printf(" %-*s %9s %9s %9s", comm_width,
49394a2a
DA		 1946 "task name", "wait time", "sch delay", "run time");
1947
414e050c
NK		 1948 if (sched->show_state)
1949 printf(" %s", "state");
1950
49394a2a
DA		 1951 printf("\n");
1952
1953 /*
1954 * units row
1955 */
1956 printf("%15s %-6s ", "", "");
1957
a407b067
DA		 1958 if (sched->show_cpu_visual)
1959 printf(" %*s ", ncpus, "");
1960
414e050c 1961 printf(" %-*s %9s %9s %9s", comm_width,
0e6758e8 1962 "[tid/pid]", "(msec)", "(msec)", "(msec)");
49394a2a 1963
414e050c
NK		 1964 if (sched->show_state)
1965 printf(" %5s", "");
1966
1967 printf("\n");
1968
49394a2a
DA		 1969 /*
1970 * separator
1971 */
1972 printf("%.15s %.6s ", graph_dotted_line, graph_dotted_line);
1973
a407b067
DA		 1974 if (sched->show_cpu_visual)
1975 printf(" %.*s ", ncpus, graph_dotted_line);
1976
0e6758e8 1977 printf(" %.*s %.9s %.9s %.9s", comm_width,
49394a2a
DA		 1978 graph_dotted_line, graph_dotted_line, graph_dotted_line,
1979 graph_dotted_line);
1980
414e050c
NK		 1981 if (sched->show_state)
1982 printf(" %.5s", graph_dotted_line);
1983
49394a2a
DA		 1984 printf("\n");
1985}
1986
414e050c
NK		 1987static char task_state_char(struct thread *thread, int state)
1988{
1989 static const char state_to_char[] = TASK_STATE_TO_CHAR_STR;
1990 unsigned bit = state ? ffs(state) : 0;
1991
1992 /* 'I' for idle */
1993 if (thread->tid == 0)
1994 return 'I';
1995
1996 return bit < sizeof(state_to_char) - 1 ? state_to_char[bit] : '?';
1997}
1998
fc1469f1 1999static void timehist_print_sample(struct perf_sched *sched,
32dcd021 2000 struct evsel *evsel,
fc1469f1 2001 struct perf_sample *sample,
6c973c90 2002 struct addr_location *al,
853b7407 2003 struct thread *thread,
414e050c 2004 u64 t, int state)
49394a2a
DA		 2005{
2006 struct thread_runtime *tr = thread__priv(thread);
292c4a8f
BG		 2007 const char *next_comm = perf_evsel__strval(evsel, sample, "next_comm");
2008 const u32 next_pid = perf_evsel__intval(evsel, sample, "next_pid");
a407b067 2009 u32 max_cpus = sched->max_cpu + 1;
49394a2a 2010 char tstr[64];
292c4a8f 2011 char nstr[30];
941bdea7 2012 u64 wait_time;
49394a2a 2013
c30d630d
DA		 2014 if (cpu_list && !test_bit(sample->cpu, cpu_bitmap))
2015 return;
2016
853b7407 2017 timestamp__scnprintf_usec(t, tstr, sizeof(tstr));
49394a2a
DA		 2018 printf("%15s [%04d] ", tstr, sample->cpu);
2019
a407b067
DA		 2020 if (sched->show_cpu_visual) {
2021 u32 i;
2022 char c;
2023
2024 printf(" ");
2025 for (i = 0; i < max_cpus; ++i) {
2026 /* flag idle times with 'i'; others are sched events */
2027 if (i == sample->cpu)
2028 c = (thread->tid == 0) ? 'i' : 's';
2029 else
2030 c = ' ';
2031 printf("%c", c);
2032 }
2033 printf(" ");
2034 }
2035
49394a2a
DA		 2036 printf(" %-*s ", comm_width, timehist_get_commstr(thread));
2037
941bdea7
NK		 2038 wait_time = tr->dt_sleep + tr->dt_iowait + tr->dt_preempt;
2039 print_sched_time(wait_time, 6);
2040
49394a2a
DA		 2041 print_sched_time(tr->dt_delay, 6);
2042 print_sched_time(tr->dt_run, 6);
fc1469f1 2043
414e050c
NK		 2044 if (sched->show_state)
2045 printf(" %5c ", task_state_char(thread, state));
2046
292c4a8f
BG		 2047 if (sched->show_next) {
2048 snprintf(nstr, sizeof(nstr), "next: %s[%d]", next_comm, next_pid);
2049 printf(" %-*s", comm_width, nstr);
2050 }
2051
2052 if (sched->show_wakeups && !sched->show_next)
fc1469f1
DA		 2053 printf(" %-*s", comm_width, "");
2054
6c973c90
DA		 2055 if (thread->tid == 0)
2056 goto out;
2057
2058 if (sched->show_callchain)
2059 printf(" ");
2060
2061 sample__fprintf_sym(sample, al, 0,
2062 EVSEL__PRINT_SYM | EVSEL__PRINT_ONELINE |
2d9bbf6e
NK		 2063 EVSEL__PRINT_CALLCHAIN_ARROW |
2064 EVSEL__PRINT_SKIP_IGNORED,
9620bc36 2065 &callchain_cursor, symbol_conf.bt_stop_list, stdout);
6c973c90
DA		 2066
2067out:
49394a2a
DA		 2068 printf("\n");
2069}
2070
2071/*
2072 * Explanation of delta-time stats:
2073 *
2074 * t = time of current schedule out event
2075 * tprev = time of previous sched out event
2076 * also time of schedule-in event for current task
2077 * last_time = time of last sched change event for current task
2078 * (i.e, time process was last scheduled out)
2079 * ready_to_run = time of wakeup for current task
2080 *
2081 * -----|------------|------------|------------|------
2082 * last ready tprev t
2083 * time to run
2084 *
2085 * |-------- dt_wait --------|
2086 * |- dt_delay -|-- dt_run --|
2087 *
2088 * dt_run = run time of current task
2089 * dt_wait = time between last schedule out event for task and tprev
2090 * represents time spent off the cpu
2091 * dt_delay = time between wakeup and schedule-in of task
2092 */
2093
2094static void timehist_update_runtime_stats(struct thread_runtime *r,
2095 u64 t, u64 tprev)
2096{
2097 r->dt_delay = 0;
941bdea7
NK		 2098 r->dt_sleep = 0;
2099 r->dt_iowait = 0;
2100 r->dt_preempt = 0;
49394a2a 2101 r->dt_run = 0;
941bdea7 2102
49394a2a
DA		 2103 if (tprev) {
2104 r->dt_run = t - tprev;
2105 if (r->ready_to_run) {
2106 if (r->ready_to_run > tprev)
2107 pr_debug("time travel: wakeup time for task > previous sched_switch event\n");
2108 else
2109 r->dt_delay = tprev - r->ready_to_run;
2110 }
2111
2112 if (r->last_time > tprev)
2113 pr_debug("time travel: last sched out time for task > previous sched_switch event\n");
941bdea7
NK		 2114 else if (r->last_time) {
2115 u64 dt_wait = tprev - r->last_time;
2116
2117 if (r->last_state == TASK_RUNNING)
2118 r->dt_preempt = dt_wait;
2119 else if (r->last_state == TASK_UNINTERRUPTIBLE)
2120 r->dt_iowait = dt_wait;
2121 else
2122 r->dt_sleep = dt_wait;
2123 }
49394a2a
DA		 2124 }
2125
2126 update_stats(&r->run_stats, r->dt_run);
587782c5
NK		 2127
2128 r->total_run_time += r->dt_run;
2129 r->total_delay_time += r->dt_delay;
2130 r->total_sleep_time += r->dt_sleep;
2131 r->total_iowait_time += r->dt_iowait;
2132 r->total_preempt_time += r->dt_preempt;
49394a2a
DA		 2133}
2134
96039c7c 2135static bool is_idle_sample(struct perf_sample *sample,
32dcd021 2136 struct evsel *evsel)
49394a2a
DA		 2137{
2138 /* pid 0 == swapper == idle task */
96039c7c
NK		 2139 if (strcmp(perf_evsel__name(evsel), "sched:sched_switch") == 0)
2140 return perf_evsel__intval(evsel, sample, "prev_pid") == 0;
49394a2a 2141
96039c7c
NK		 2142 return sample->pid == 0;
2143}
2144
2145static void save_task_callchain(struct perf_sched *sched,
2146 struct perf_sample *sample,
32dcd021 2147 struct evsel *evsel,
96039c7c
NK		 2148 struct machine *machine)
2149{
2150 struct callchain_cursor *cursor = &callchain_cursor;
2151 struct thread *thread;
6c973c90
DA		 2152
2153 /* want main thread for process - has maps */
2154 thread = machine__findnew_thread(machine, sample->pid, sample->pid);
2155 if (thread == NULL) {
2156 pr_debug("Failed to get thread for pid %d.\n", sample->pid);
96039c7c 2157 return;
6c973c90
DA		 2158 }
2159
4c50563d 2160 if (!sched->show_callchain || sample->callchain == NULL)
96039c7c 2161 return;
6c973c90
DA		 2162
2163 if (thread__resolve_callchain(thread, cursor, evsel, sample,
8388deb3 2164 NULL, NULL, sched->max_stack + 2) != 0) {
bb963e16 2165 if (verbose > 0)
62d94b00 2166 pr_err("Failed to resolve callchain. Skipping\n");
6c973c90 2167
96039c7c 2168 return;
6c973c90 2169 }
cdeb01bf 2170
6c973c90 2171 callchain_cursor_commit(cursor);
cdeb01bf
NK		 2172
2173 while (true) {
2174 struct callchain_cursor_node *node;
2175 struct symbol *sym;
2176
2177 node = callchain_cursor_current(cursor);
2178 if (node == NULL)
2179 break;
2180
5f0fef8a 2181 sym = node->ms.sym;
a7c3899c 2182 if (sym) {
cdeb01bf
NK		 2183 if (!strcmp(sym->name, "schedule") ||
2184 !strcmp(sym->name, "__schedule") ||
2185 !strcmp(sym->name, "preempt_schedule"))
2186 sym->ignore = 1;
2187 }
2188
2189 callchain_cursor_advance(cursor);
2190 }
49394a2a
DA		 2191}
2192
3bc2fa9c
NK		 2193static int init_idle_thread(struct thread *thread)
2194{
2195 struct idle_thread_runtime *itr;
2196
2197 thread__set_comm(thread, idle_comm, 0);
2198
2199 itr = zalloc(sizeof(*itr));
2200 if (itr == NULL)
2201 return -ENOMEM;
2202
2203 init_stats(&itr->tr.run_stats);
2204 callchain_init(&itr->callchain);
2205 callchain_cursor_reset(&itr->cursor);
2206 thread__set_priv(thread, itr);
2207
2208 return 0;
2209}
2210
49394a2a
DA		 2211/*
2212 * Track idle stats per cpu by maintaining a local thread
2213 * struct for the idle task on each cpu.
2214 */
2215static int init_idle_threads(int ncpu)
2216{
3bc2fa9c 2217 int i, ret;
49394a2a
DA		 2218
2219 idle_threads = zalloc(ncpu * sizeof(struct thread *));
2220 if (!idle_threads)
2221 return -ENOMEM;
2222
b336352b 2223 idle_max_cpu = ncpu;
49394a2a
DA		 2224
2225 /* allocate the actual thread struct if needed */
2226 for (i = 0; i < ncpu; ++i) {
2227 idle_threads[i] = thread__new(0, 0);
2228 if (idle_threads[i] == NULL)
2229 return -ENOMEM;
2230
3bc2fa9c
NK		 2231 ret = init_idle_thread(idle_threads[i]);
2232 if (ret < 0)
2233 return ret;
49394a2a
DA		 2234 }
2235
2236 return 0;
2237}
2238
2239static void free_idle_threads(void)
2240{
2241 int i;
2242
2243 if (idle_threads == NULL)
2244 return;
2245
b336352b 2246 for (i = 0; i < idle_max_cpu; ++i) {
49394a2a
DA		 2247 if ((idle_threads[i]))
2248 thread__delete(idle_threads[i]);
2249 }
2250
2251 free(idle_threads);
2252}
2253
2254static struct thread *get_idle_thread(int cpu)
2255{
2256 /*
2257 * expand/allocate array of pointers to local thread
2258 * structs if needed
2259 */
2260 if ((cpu >= idle_max_cpu) || (idle_threads == NULL)) {
2261 int i, j = __roundup_pow_of_two(cpu+1);
2262 void *p;
2263
2264 p = realloc(idle_threads, j * sizeof(struct thread *));
2265 if (!p)
2266 return NULL;
2267
2268 idle_threads = (struct thread **) p;
b336352b 2269 for (i = idle_max_cpu; i < j; ++i)
49394a2a
DA		 2270 idle_threads[i] = NULL;
2271
2272 idle_max_cpu = j;
2273 }
2274
2275 /* allocate a new thread struct if needed */
2276 if (idle_threads[cpu] == NULL) {
2277 idle_threads[cpu] = thread__new(0, 0);
2278
2279 if (idle_threads[cpu]) {
3bc2fa9c
NK		 2280 if (init_idle_thread(idle_threads[cpu]) < 0)
2281 return NULL;
49394a2a
DA		 2282 }
2283 }
2284
2285 return idle_threads[cpu];
2286}
2287
4c50563d
ACM		 2288static void save_idle_callchain(struct perf_sched *sched,
2289 struct idle_thread_runtime *itr,
699b5b92
NK		 2290 struct perf_sample *sample)
2291{
4c50563d 2292 if (!sched->show_callchain || sample->callchain == NULL)
699b5b92
NK		 2293 return;
2294
2295 callchain_cursor__copy(&itr->cursor, &callchain_cursor);
2296}
2297
6c973c90
DA		 2298static struct thread *timehist_get_thread(struct perf_sched *sched,
2299 struct perf_sample *sample,
49394a2a 2300 struct machine *machine,
32dcd021 2301 struct evsel *evsel)
49394a2a
DA		 2302{
2303 struct thread *thread;
2304
96039c7c 2305 if (is_idle_sample(sample, evsel)) {
49394a2a
DA		 2306 thread = get_idle_thread(sample->cpu);
2307 if (thread == NULL)
2308 pr_err("Failed to get idle thread for cpu %d.\n", sample->cpu);
2309
2310 } else {
5d92d96a
NK		 2311 /* there were samples with tid 0 but non-zero pid */
2312 thread = machine__findnew_thread(machine, sample->pid,
2313 sample->tid ?: sample->pid);
49394a2a
DA		 2314 if (thread == NULL) {
2315 pr_debug("Failed to get thread for tid %d. skipping sample.\n",
2316 sample->tid);
2317 }
96039c7c
NK		 2318
2319 save_task_callchain(sched, sample, evsel, machine);
699b5b92
NK		 2320 if (sched->idle_hist) {
2321 struct thread *idle;
2322 struct idle_thread_runtime *itr;
2323
2324 idle = get_idle_thread(sample->cpu);
2325 if (idle == NULL) {
2326 pr_err("Failed to get idle thread for cpu %d.\n", sample->cpu);
2327 return NULL;
2328 }
2329
2330 itr = thread__priv(idle);
2331 if (itr == NULL)
2332 return NULL;
2333
2334 itr->last_thread = thread;
2335
2336 /* copy task callchain when entering to idle */
2337 if (perf_evsel__intval(evsel, sample, "next_pid") == 0)
4c50563d 2338 save_idle_callchain(sched, itr, sample);
699b5b92 2339 }
49394a2a
DA		 2340 }
2341
2342 return thread;
2343}
2344
52df138c 2345static bool timehist_skip_sample(struct perf_sched *sched,
a4b2b6f5 2346 struct thread *thread,
32dcd021 2347 struct evsel *evsel,
a4b2b6f5 2348 struct perf_sample *sample)
49394a2a
DA		 2349{
2350 bool rc = false;
2351
52df138c 2352 if (thread__is_filtered(thread)) {
49394a2a 2353 rc = true;
52df138c
DA		 2354 sched->skipped_samples++;
2355 }
49394a2a 2356
a4b2b6f5
NK		 2357 if (sched->idle_hist) {
2358 if (strcmp(perf_evsel__name(evsel), "sched:sched_switch"))
2359 rc = true;
2360 else if (perf_evsel__intval(evsel, sample, "prev_pid") != 0 &&
2361 perf_evsel__intval(evsel, sample, "next_pid") != 0)
2362 rc = true;
2363 }
2364
49394a2a
DA		 2365 return rc;
2366}
2367
fc1469f1 2368static void timehist_print_wakeup_event(struct perf_sched *sched,
32dcd021 2369 struct evsel *evsel,
fc1469f1
DA		 2370 struct perf_sample *sample,
2371 struct machine *machine,
2372 struct thread *awakened)
2373{
2374 struct thread *thread;
2375 char tstr[64];
2376
2377 thread = machine__findnew_thread(machine, sample->pid, sample->tid);
2378 if (thread == NULL)
2379 return;
2380
2381 /* show wakeup unless both awakee and awaker are filtered */
a4b2b6f5
NK		 2382 if (timehist_skip_sample(sched, thread, evsel, sample) &&
2383 timehist_skip_sample(sched, awakened, evsel, sample)) {
fc1469f1
DA		 2384 return;
2385 }
2386
2387 timestamp__scnprintf_usec(sample->time, tstr, sizeof(tstr));
2388 printf("%15s [%04d] ", tstr, sample->cpu);
a407b067
DA		 2389 if (sched->show_cpu_visual)
2390 printf(" %*s ", sched->max_cpu + 1, "");
fc1469f1
DA		 2391
2392 printf(" %-*s ", comm_width, timehist_get_commstr(thread));
2393
2394 /* dt spacer */
2395 printf(" %9s %9s %9s ", "", "", "");
2396
2397 printf("awakened: %s", timehist_get_commstr(awakened));
2398
2399 printf("\n");
2400}
2401
2402static int timehist_sched_wakeup_event(struct perf_tool *tool,
49394a2a 2403 union perf_event *event __maybe_unused,
32dcd021 2404 struct evsel *evsel,
49394a2a
DA		 2405 struct perf_sample *sample,
2406 struct machine *machine)
2407{
fc1469f1 2408 struct perf_sched *sched = container_of(tool, struct perf_sched, tool);
49394a2a
DA		 2409 struct thread *thread;
2410 struct thread_runtime *tr = NULL;
2411 /* want pid of awakened task not pid in sample */
2412 const u32 pid = perf_evsel__intval(evsel, sample, "pid");
2413
2414 thread = machine__findnew_thread(machine, 0, pid);
2415 if (thread == NULL)
2416 return -1;
2417
2418 tr = thread__get_runtime(thread);
2419 if (tr == NULL)
2420 return -1;
2421
2422 if (tr->ready_to_run == 0)
2423 tr->ready_to_run = sample->time;
2424
fc1469f1 2425 /* show wakeups if requested */
853b7407
DA		 2426 if (sched->show_wakeups &&
2427 !perf_time__skip_sample(&sched->ptime, sample->time))
a4b2b6f5 2428 timehist_print_wakeup_event(sched, evsel, sample, machine, thread);
fc1469f1 2429
49394a2a
DA		 2430 return 0;
2431}
2432
350f54fa 2433static void timehist_print_migration_event(struct perf_sched *sched,
32dcd021 2434 struct evsel *evsel,
350f54fa
DA		 2435 struct perf_sample *sample,
2436 struct machine *machine,
2437 struct thread *migrated)
2438{
2439 struct thread *thread;
2440 char tstr[64];
2441 u32 max_cpus = sched->max_cpu + 1;
2442 u32 ocpu, dcpu;
2443
2444 if (sched->summary_only)
2445 return;
2446
2447 max_cpus = sched->max_cpu + 1;
2448 ocpu = perf_evsel__intval(evsel, sample, "orig_cpu");
2449 dcpu = perf_evsel__intval(evsel, sample, "dest_cpu");
2450
2451 thread = machine__findnew_thread(machine, sample->pid, sample->tid);
2452 if (thread == NULL)
2453 return;
2454
a4b2b6f5
NK		 2455 if (timehist_skip_sample(sched, thread, evsel, sample) &&
2456 timehist_skip_sample(sched, migrated, evsel, sample)) {
350f54fa
DA		 2457 return;
2458 }
2459
2460 timestamp__scnprintf_usec(sample->time, tstr, sizeof(tstr));
2461 printf("%15s [%04d] ", tstr, sample->cpu);
2462
2463 if (sched->show_cpu_visual) {
2464 u32 i;
2465 char c;
2466
2467 printf(" ");
2468 for (i = 0; i < max_cpus; ++i) {
2469 c = (i == sample->cpu) ? 'm' : ' ';
2470 printf("%c", c);
2471 }
2472 printf(" ");
2473 }
2474
2475 printf(" %-*s ", comm_width, timehist_get_commstr(thread));
2476
2477 /* dt spacer */
2478 printf(" %9s %9s %9s ", "", "", "");
2479
2480 printf("migrated: %s", timehist_get_commstr(migrated));
2481 printf(" cpu %d => %d", ocpu, dcpu);
2482
2483 printf("\n");
2484}
2485
2486static int timehist_migrate_task_event(struct perf_tool *tool,
2487 union perf_event *event __maybe_unused,
32dcd021 2488 struct evsel *evsel,
350f54fa
DA		 2489 struct perf_sample *sample,
2490 struct machine *machine)
2491{
2492 struct perf_sched *sched = container_of(tool, struct perf_sched, tool);
2493 struct thread *thread;
2494 struct thread_runtime *tr = NULL;
2495 /* want pid of migrated task not pid in sample */
2496 const u32 pid = perf_evsel__intval(evsel, sample, "pid");
2497
2498 thread = machine__findnew_thread(machine, 0, pid);
2499 if (thread == NULL)
2500 return -1;
2501
2502 tr = thread__get_runtime(thread);
2503 if (tr == NULL)
2504 return -1;
2505
2506 tr->migrations++;
2507
2508 /* show migrations if requested */
2509 timehist_print_migration_event(sched, evsel, sample, machine, thread);
2510
2511 return 0;
2512}
2513
52df138c 2514static int timehist_sched_change_event(struct perf_tool *tool,
49394a2a 2515 union perf_event *event,
32dcd021 2516 struct evsel *evsel,
49394a2a
DA		 2517 struct perf_sample *sample,
2518 struct machine *machine)
2519{
fc1469f1 2520 struct perf_sched *sched = container_of(tool, struct perf_sched, tool);
853b7407 2521 struct perf_time_interval *ptime = &sched->ptime;
49394a2a
DA		 2522 struct addr_location al;
2523 struct thread *thread;
2524 struct thread_runtime *tr = NULL;
853b7407 2525 u64 tprev, t = sample->time;
49394a2a 2526 int rc = 0;
414e050c
NK		 2527 int state = perf_evsel__intval(evsel, sample, "prev_state");
2528
49394a2a
DA		 2529
2530 if (machine__resolve(machine, &al, sample) < 0) {
2531 pr_err("problem processing %d event. skipping it\n",
2532 event->header.type);
2533 rc = -1;
2534 goto out;
2535 }
2536
6c973c90 2537 thread = timehist_get_thread(sched, sample, machine, evsel);
49394a2a
DA		 2538 if (thread == NULL) {
2539 rc = -1;
2540 goto out;
2541 }
2542
a4b2b6f5 2543 if (timehist_skip_sample(sched, thread, evsel, sample))
49394a2a
DA		 2544 goto out;
2545
2546 tr = thread__get_runtime(thread);
2547 if (tr == NULL) {
2548 rc = -1;
2549 goto out;
2550 }
2551
2552 tprev = perf_evsel__get_time(evsel, sample->cpu);
2553
853b7407
DA		 2554 /*
2555 * If start time given:
2556 * - sample time is under window user cares about - skip sample
2557 * - tprev is under window user cares about - reset to start of window
2558 */
2559 if (ptime->start && ptime->start > t)
2560 goto out;
2561
bdd75729 2562 if (tprev && ptime->start > tprev)
853b7407
DA		 2563 tprev = ptime->start;
2564
2565 /*
2566 * If end time given:
2567 * - previous sched event is out of window - we are done
2568 * - sample time is beyond window user cares about - reset it
2569 * to close out stats for time window interest
2570 */
2571 if (ptime->end) {
2572 if (tprev > ptime->end)
2573 goto out;
2574
2575 if (t > ptime->end)
2576 t = ptime->end;
2577 }
2578
07235f84
NK		 2579 if (!sched->idle_hist || thread->tid == 0) {
2580 timehist_update_runtime_stats(tr, t, tprev);
2581
2582 if (sched->idle_hist) {
2583 struct idle_thread_runtime *itr = (void *)tr;
2584 struct thread_runtime *last_tr;
2585
2586 BUG_ON(thread->tid != 0);
2587
2588 if (itr->last_thread == NULL)
2589 goto out;
2590
2591 /* add current idle time as last thread's runtime */
2592 last_tr = thread__get_runtime(itr->last_thread);
2593 if (last_tr == NULL)
2594 goto out;
2595
2596 timehist_update_runtime_stats(last_tr, t, tprev);
2597 /*
2598 * remove delta time of last thread as it's not updated
2599 * and otherwise it will show an invalid value next
2600 * time. we only care total run time and run stat.
2601 */
2602 last_tr->dt_run = 0;
07235f84 2603 last_tr->dt_delay = 0;
941bdea7
NK		 2604 last_tr->dt_sleep = 0;
2605 last_tr->dt_iowait = 0;
2606 last_tr->dt_preempt = 0;
07235f84 2607
ba957ebb
NK		 2608 if (itr->cursor.nr)
2609 callchain_append(&itr->callchain, &itr->cursor, t - tprev);
2610
07235f84
NK		 2611 itr->last_thread = NULL;
2612 }
2613 }
853b7407 2614
52df138c 2615 if (!sched->summary_only)
292c4a8f 2616 timehist_print_sample(sched, evsel, sample, &al, thread, t, state);
49394a2a
DA		 2617
2618out:
9396c9cb
NK		 2619 if (sched->hist_time.start == 0 && t >= ptime->start)
2620 sched->hist_time.start = t;
2621 if (ptime->end == 0 || t <= ptime->end)
2622 sched->hist_time.end = t;
2623
49394a2a
DA		 2624 if (tr) {
2625 /* time of this sched_switch event becomes last time task seen */
2626 tr->last_time = sample->time;
2627
941bdea7 2628 /* last state is used to determine where to account wait time */
414e050c 2629 tr->last_state = state;
941bdea7 2630
49394a2a
DA		 2631 /* sched out event for task so reset ready to run time */
2632 tr->ready_to_run = 0;
2633 }
2634
2635 perf_evsel__save_time(evsel, sample->time, sample->cpu);
2636
2637 return rc;
2638}
2639
2640static int timehist_sched_switch_event(struct perf_tool *tool,
2641 union perf_event *event,
32dcd021 2642 struct evsel *evsel,
49394a2a
DA		 2643 struct perf_sample *sample,
2644 struct machine *machine __maybe_unused)
2645{
2646 return timehist_sched_change_event(tool, event, evsel, sample, machine);
2647}
2648
2649static int process_lost(struct perf_tool *tool __maybe_unused,
2650 union perf_event *event,
2651 struct perf_sample *sample,
2652 struct machine *machine __maybe_unused)
2653{
2654 char tstr[64];
2655
2656 timestamp__scnprintf_usec(sample->time, tstr, sizeof(tstr));
2657 printf("%15s ", tstr);
5290ed69 2658 printf("lost %" PRI_lu64 " events on cpu %d\n", event->lost.lost, sample->cpu);
49394a2a
DA		 2659
2660 return 0;
2661}
2662
2663
52df138c
DA		 2664static void print_thread_runtime(struct thread *t,
2665 struct thread_runtime *r)
2666{
2667 double mean = avg_stats(&r->run_stats);
2668 float stddev;
2669
2670 printf("%*s %5d %9" PRIu64 " ",
2671 comm_width, timehist_get_commstr(t), t->ppid,
2672 (u64) r->run_stats.n);
2673
2674 print_sched_time(r->total_run_time, 8);
2675 stddev = rel_stddev_stats(stddev_stats(&r->run_stats), mean);
2676 print_sched_time(r->run_stats.min, 6);
2677 printf(" ");
2678 print_sched_time((u64) mean, 6);
2679 printf(" ");
2680 print_sched_time(r->run_stats.max, 6);
2681 printf(" ");
2682 printf("%5.2f", stddev);
350f54fa 2683 printf(" %5" PRIu64, r->migrations);
52df138c
DA		 2684 printf("\n");
2685}
2686
587782c5
NK		 2687static void print_thread_waittime(struct thread *t,
2688 struct thread_runtime *r)
2689{
2690 printf("%*s %5d %9" PRIu64 " ",
2691 comm_width, timehist_get_commstr(t), t->ppid,
2692 (u64) r->run_stats.n);
2693
2694 print_sched_time(r->total_run_time, 8);
2695 print_sched_time(r->total_sleep_time, 6);
2696 printf(" ");
2697 print_sched_time(r->total_iowait_time, 6);
2698 printf(" ");
2699 print_sched_time(r->total_preempt_time, 6);
2700 printf(" ");
2701 print_sched_time(r->total_delay_time, 6);
2702 printf("\n");
2703}
2704
52df138c 2705struct total_run_stats {
587782c5 2706 struct perf_sched *sched;
52df138c
DA		 2707 u64 sched_count;
2708 u64 task_count;
2709 u64 total_run_time;
2710};
2711
2712static int __show_thread_runtime(struct thread *t, void *priv)
2713{
2714 struct total_run_stats *stats = priv;
2715 struct thread_runtime *r;
2716
2717 if (thread__is_filtered(t))
2718 return 0;
2719
2720 r = thread__priv(t);
2721 if (r && r->run_stats.n) {
2722 stats->task_count++;
2723 stats->sched_count += r->run_stats.n;
2724 stats->total_run_time += r->total_run_time;
587782c5
NK		 2725
2726 if (stats->sched->show_state)
2727 print_thread_waittime(t, r);
2728 else
2729 print_thread_runtime(t, r);
52df138c
DA		 2730 }
2731
2732 return 0;
2733}
2734
2735static int show_thread_runtime(struct thread *t, void *priv)
2736{
2737 if (t->dead)
2738 return 0;
2739
2740 return __show_thread_runtime(t, priv);
2741}
2742
2743static int show_deadthread_runtime(struct thread *t, void *priv)
2744{
2745 if (!t->dead)
2746 return 0;
2747
2748 return __show_thread_runtime(t, priv);
2749}
2750
ba957ebb
NK		 2751static size_t callchain__fprintf_folded(FILE *fp, struct callchain_node *node)
2752{
2753 const char *sep = " <- ";
2754 struct callchain_list *chain;
2755 size_t ret = 0;
2756 char bf[1024];
2757 bool first;
2758
2759 if (node == NULL)
2760 return 0;
2761
2762 ret = callchain__fprintf_folded(fp, node->parent);
2763 first = (ret == 0);
2764
2765 list_for_each_entry(chain, &node->val, list) {
2766 if (chain->ip >= PERF_CONTEXT_MAX)
2767 continue;
2768 if (chain->ms.sym && chain->ms.sym->ignore)
2769 continue;
2770 ret += fprintf(fp, "%s%s", first ? "" : sep,
2771 callchain_list__sym_name(chain, bf, sizeof(bf),
2772 false));
2773 first = false;
2774 }
2775
2776 return ret;
2777}
2778
cb4c13a5 2779static size_t timehist_print_idlehist_callchain(struct rb_root_cached *root)
ba957ebb
NK		 2780{
2781 size_t ret = 0;
2782 FILE *fp = stdout;
2783 struct callchain_node *chain;
cb4c13a5 2784 struct rb_node *rb_node = rb_first_cached(root);
ba957ebb
NK		 2785
2786 printf(" %16s %8s %s\n", "Idle time (msec)", "Count", "Callchains");
2787 printf(" %.16s %.8s %.50s\n", graph_dotted_line, graph_dotted_line,
2788 graph_dotted_line);
2789
2790 while (rb_node) {
2791 chain = rb_entry(rb_node, struct callchain_node, rb_node);
2792 rb_node = rb_next(rb_node);
2793
2794 ret += fprintf(fp, " ");
2795 print_sched_time(chain->hit, 12);
2796 ret += 16; /* print_sched_time returns 2nd arg + 4 */
2797 ret += fprintf(fp, " %8d ", chain->count);
2798 ret += callchain__fprintf_folded(fp, chain);
2799 ret += fprintf(fp, "\n");
2800 }
2801
2802 return ret;
2803}
2804
52df138c
DA		 2805static void timehist_print_summary(struct perf_sched *sched,
2806 struct perf_session *session)
2807{
2808 struct machine *m = &session->machines.host;
2809 struct total_run_stats totals;
2810 u64 task_count;
2811 struct thread *t;
2812 struct thread_runtime *r;
2813 int i;
9396c9cb 2814 u64 hist_time = sched->hist_time.end - sched->hist_time.start;
52df138c
DA		 2815
2816 memset(&totals, 0, sizeof(totals));
587782c5 2817 totals.sched = sched;
52df138c 2818
07235f84
NK		 2819 if (sched->idle_hist) {
2820 printf("\nIdle-time summary\n");
2821 printf("%*s parent sched-out ", comm_width, "comm");
2822 printf(" idle-time min-idle avg-idle max-idle stddev migrations\n");
587782c5
NK		 2823 } else if (sched->show_state) {
2824 printf("\nWait-time summary\n");
2825 printf("%*s parent sched-in ", comm_width, "comm");
2826 printf(" run-time sleep iowait preempt delay\n");
07235f84
NK		 2827 } else {
2828 printf("\nRuntime summary\n");
2829 printf("%*s parent sched-in ", comm_width, "comm");
2830 printf(" run-time min-run avg-run max-run stddev migrations\n");
2831 }
52df138c 2832 printf("%*s (count) ", comm_width, "");
587782c5
NK		 2833 printf(" (msec) (msec) (msec) (msec) %s\n",
2834 sched->show_state ? "(msec)" : "%");
350f54fa 2835 printf("%.117s\n", graph_dotted_line);
52df138c
DA		 2836
2837 machine__for_each_thread(m, show_thread_runtime, &totals);
2838 task_count = totals.task_count;
2839 if (!task_count)
2840 printf("<no still running tasks>\n");
2841
2842 printf("\nTerminated tasks:\n");
2843 machine__for_each_thread(m, show_deadthread_runtime, &totals);
2844 if (task_count == totals.task_count)
2845 printf("<no terminated tasks>\n");
2846
2847 /* CPU idle stats not tracked when samples were skipped */
07235f84 2848 if (sched->skipped_samples && !sched->idle_hist)
52df138c
DA		 2849 return;
2850
2851 printf("\nIdle stats:\n");
b336352b 2852 for (i = 0; i < idle_max_cpu; ++i) {
52df138c
DA		 2853 t = idle_threads[i];
2854 if (!t)
2855 continue;
2856
2857 r = thread__priv(t);
2858 if (r && r->run_stats.n) {
2859 totals.sched_count += r->run_stats.n;
2860 printf(" CPU %2d idle for ", i);
2861 print_sched_time(r->total_run_time, 6);
9396c9cb 2862 printf(" msec (%6.2f%%)\n", 100.0 * r->total_run_time / hist_time);
52df138c
DA		 2863 } else
2864 printf(" CPU %2d idle entire time window\n", i);
2865 }
2866
4c50563d 2867 if (sched->idle_hist && sched->show_callchain) {
ba957ebb
NK		 2868 callchain_param.mode = CHAIN_FOLDED;
2869 callchain_param.value = CCVAL_PERIOD;
2870
2871 callchain_register_param(&callchain_param);
2872
2873 printf("\nIdle stats by callchain:\n");
2874 for (i = 0; i < idle_max_cpu; ++i) {
2875 struct idle_thread_runtime *itr;
2876
2877 t = idle_threads[i];
2878 if (!t)
2879 continue;
2880
2881 itr = thread__priv(t);
2882 if (itr == NULL)
2883 continue;
2884
cb4c13a5 2885 callchain_param.sort(&itr->sorted_root.rb_root, &itr->callchain,
ba957ebb
NK		 2886 0, &callchain_param);
2887
2888 printf(" CPU %2d:", i);
2889 print_sched_time(itr->tr.total_run_time, 6);
2890 printf(" msec\n");
2891 timehist_print_idlehist_callchain(&itr->sorted_root);
2892 printf("\n");
2893 }
2894 }
2895
52df138c
DA		 2896 printf("\n"
2897 " Total number of unique tasks: %" PRIu64 "\n"
9396c9cb 2898 "Total number of context switches: %" PRIu64 "\n",
52df138c
DA		 2899 totals.task_count, totals.sched_count);
2900
9396c9cb 2901 printf(" Total run time (msec): ");
52df138c
DA		 2902 print_sched_time(totals.total_run_time, 2);
2903 printf("\n");
9396c9cb
NK		 2904
2905 printf(" Total scheduling time (msec): ");
2906 print_sched_time(hist_time, 2);
2907 printf(" (x %d)\n", sched->max_cpu);
52df138c
DA		 2908}
2909
49394a2a
DA		 2910typedef int (*sched_handler)(struct perf_tool *tool,
2911 union perf_event *event,
32dcd021 2912 struct evsel *evsel,
49394a2a
DA		 2913 struct perf_sample *sample,
2914 struct machine *machine);
2915
2916static int perf_timehist__process_sample(struct perf_tool *tool,
2917 union perf_event *event,
2918 struct perf_sample *sample,
32dcd021 2919 struct evsel *evsel,
49394a2a
DA		 2920 struct machine *machine)
2921{
2922 struct perf_sched *sched = container_of(tool, struct perf_sched, tool);
2923 int err = 0;
2924 int this_cpu = sample->cpu;
2925
2926 if (this_cpu > sched->max_cpu)
2927 sched->max_cpu = this_cpu;
2928
2929 if (evsel->handler != NULL) {
2930 sched_handler f = evsel->handler;
2931
2932 err = f(tool, event, evsel, sample, machine);
2933 }
2934
2935 return err;
2936}
2937
6c973c90 2938static int timehist_check_attr(struct perf_sched *sched,
63503dba 2939 struct evlist *evlist)
6c973c90 2940{
32dcd021 2941 struct evsel *evsel;
6c973c90
DA		 2942 struct evsel_runtime *er;
2943
ce9036a6 2944 list_for_each_entry(evsel, &evlist->core.entries, core.node) {
6c973c90
DA		 2945 er = perf_evsel__get_runtime(evsel);
2946 if (er == NULL) {
2947 pr_err("Failed to allocate memory for evsel runtime data\n");
2948 return -1;
2949 }
2950
27de9b2b 2951 if (sched->show_callchain && !evsel__has_callchain(evsel)) {
6c973c90
DA		 2952 pr_info("Samples do not have callchains.\n");
2953 sched->show_callchain = 0;
2954 symbol_conf.use_callchain = 0;
2955 }
2956 }
2957
2958 return 0;
2959}
2960
49394a2a
DA		 2961static int perf_sched__timehist(struct perf_sched *sched)
2962{
32dcd021 2963 const struct evsel_str_handler handlers[] = {
49394a2a
DA		 2964 { "sched:sched_switch", timehist_sched_switch_event, },
2965 { "sched:sched_wakeup", timehist_sched_wakeup_event, },
2966 { "sched:sched_wakeup_new", timehist_sched_wakeup_event, },
2967 };
32dcd021 2968 const struct evsel_str_handler migrate_handlers[] = {
350f54fa
DA		 2969 { "sched:sched_migrate_task", timehist_migrate_task_event, },
2970 };
8ceb41d7 2971 struct perf_data data = {
2d4f2799
JO		 2972 .path = input_name,
2973 .mode = PERF_DATA_MODE_READ,
2974 .force = sched->force,
49394a2a
DA		 2975 };
2976
2977 struct perf_session *session;
63503dba 2978 struct evlist *evlist;
49394a2a
DA		 2979 int err = -1;
2980
2981 /*
2982 * event handlers for timehist option
2983 */
2984 sched->tool.sample = perf_timehist__process_sample;
2985 sched->tool.mmap = perf_event__process_mmap;
2986 sched->tool.comm = perf_event__process_comm;
2987 sched->tool.exit = perf_event__process_exit;
2988 sched->tool.fork = perf_event__process_fork;
2989 sched->tool.lost = process_lost;
2990 sched->tool.attr = perf_event__process_attr;
2991 sched->tool.tracing_data = perf_event__process_tracing_data;
2992 sched->tool.build_id = perf_event__process_build_id;
2993
2994 sched->tool.ordered_events = true;
2995 sched->tool.ordering_requires_timestamps = true;
2996
6c973c90
DA		 2997 symbol_conf.use_callchain = sched->show_callchain;
2998
8ceb41d7 2999 session = perf_session__new(&data, false, &sched->tool);
6ef81c55
MI		 3000 if (IS_ERR(session))
3001 return PTR_ERR(session);
49394a2a 3002
c30d630d
DA		 3003 if (cpu_list) {
3004 err = perf_session__cpu_bitmap(session, cpu_list, cpu_bitmap);
3005 if (err < 0)
3006 goto out;
3007 }
3008
52df138c
DA		 3009 evlist = session->evlist;
3010
49394a2a
DA		 3011 symbol__init(&session->header.env);
3012
853b7407
DA		 3013 if (perf_time__parse_str(&sched->ptime, sched->time_str) != 0) {
3014 pr_err("Invalid time string\n");
3015 return -EINVAL;
3016 }
3017
6c973c90
DA		 3018 if (timehist_check_attr(sched, evlist) != 0)
3019 goto out;
3020
49394a2a
DA		 3021 setup_pager();
3022
3023 /* setup per-evsel handlers */
3024 if (perf_session__set_tracepoints_handlers(session, handlers))
3025 goto out;
3026
f45bf8d3
DA		 3027 /* sched_switch event at a minimum needs to exist */
3028 if (!perf_evlist__find_tracepoint_by_name(session->evlist,
3029 "sched:sched_switch")) {
3030 pr_err("No sched_switch events found. Have you run 'perf sched record'?\n");
49394a2a 3031 goto out;
f45bf8d3 3032 }
49394a2a 3033
350f54fa
DA		 3034 if (sched->show_migrations &&
3035 perf_session__set_tracepoints_handlers(session, migrate_handlers))
3036 goto out;
3037
49394a2a
DA		 3038 /* pre-allocate struct for per-CPU idle stats */
3039 sched->max_cpu = session->header.env.nr_cpus_online;
3040 if (sched->max_cpu == 0)
3041 sched->max_cpu = 4;
3042 if (init_idle_threads(sched->max_cpu))
3043 goto out;
3044
52df138c
DA		 3045 /* summary_only implies summary option, but don't overwrite summary if set */
3046 if (sched->summary_only)
3047 sched->summary = sched->summary_only;
3048
3049 if (!sched->summary_only)
a407b067 3050 timehist_header(sched);
49394a2a
DA		 3051
3052 err = perf_session__process_events(session);
3053 if (err) {
3054 pr_err("Failed to process events, error %d", err);
3055 goto out;
3056 }
3057
52df138c
DA		 3058 sched->nr_events = evlist->stats.nr_events[0];
3059 sched->nr_lost_events = evlist->stats.total_lost;
3060 sched->nr_lost_chunks = evlist->stats.nr_events[PERF_RECORD_LOST];
3061
3062 if (sched->summary)
3063 timehist_print_summary(sched, session);
3064
49394a2a
DA		 3065out:
3066 free_idle_threads();
3067 perf_session__delete(session);
3068
3069 return err;
3070}
3071
3072
0e9b07e5 3073static void print_bad_events(struct perf_sched *sched)
0ec04e16 3074{
0e9b07e5 3075 if (sched->nr_unordered_timestamps && sched->nr_timestamps) {
0ec04e16 3076 printf(" INFO: %.3f%% unordered timestamps (%ld out of %ld)\n",
0e9b07e5
ACM		 3077 (double)sched->nr_unordered_timestamps/(double)sched->nr_timestamps*100.0,
3078 sched->nr_unordered_timestamps, sched->nr_timestamps);
0ec04e16 3079 }
0e9b07e5 3080 if (sched->nr_lost_events && sched->nr_events) {
0ec04e16 3081 printf(" INFO: %.3f%% lost events (%ld out of %ld, in %ld chunks)\n",
0e9b07e5
ACM		 3082 (double)sched->nr_lost_events/(double)sched->nr_events * 100.0,
3083 sched->nr_lost_events, sched->nr_events, sched->nr_lost_chunks);
0ec04e16 3084 }
0e9b07e5 3085 if (sched->nr_context_switch_bugs && sched->nr_timestamps) {
0ec04e16 3086 printf(" INFO: %.3f%% context switch bugs (%ld out of %ld)",
0e9b07e5
ACM		 3087 (double)sched->nr_context_switch_bugs/(double)sched->nr_timestamps*100.0,
3088 sched->nr_context_switch_bugs, sched->nr_timestamps);
3089 if (sched->nr_lost_events)
0ec04e16
IM		 3090 printf(" (due to lost events?)");
3091 printf("\n");
3092 }
3093}
3094
cb4c13a5 3095static void __merge_work_atoms(struct rb_root_cached *root, struct work_atoms *data)
2f80dd44 3096{
cb4c13a5 3097 struct rb_node **new = &(root->rb_root.rb_node), *parent = NULL;
2f80dd44
JB		 3098 struct work_atoms *this;
3099 const char *comm = thread__comm_str(data->thread), *this_comm;
cb4c13a5 3100 bool leftmost = true;
2f80dd44
JB		 3101
3102 while (*new) {
3103 int cmp;
3104
3105 this = container_of(*new, struct work_atoms, node);
3106 parent = *new;
3107
3108 this_comm = thread__comm_str(this->thread);
3109 cmp = strcmp(comm, this_comm);
3110 if (cmp > 0) {
3111 new = &((*new)->rb_left);
3112 } else if (cmp < 0) {
3113 new = &((*new)->rb_right);
cb4c13a5 3114 leftmost = false;
2f80dd44
JB		 3115 } else {
3116 this->num_merged++;
3117 this->total_runtime += data->total_runtime;
3118 this->nb_atoms += data->nb_atoms;
3119 this->total_lat += data->total_lat;
3120 list_splice(&data->work_list, &this->work_list);
3121 if (this->max_lat < data->max_lat) {
3122 this->max_lat = data->max_lat;
3123 this->max_lat_at = data->max_lat_at;
3124 }
3125 zfree(&data);
3126 return;
3127 }
3128 }
3129
3130 data->num_merged++;
3131 rb_link_node(&data->node, parent, new);
cb4c13a5 3132 rb_insert_color_cached(&data->node, root, leftmost);
2f80dd44
JB		 3133}
3134
3135static void perf_sched__merge_lat(struct perf_sched *sched)
3136{
3137 struct work_atoms *data;
3138 struct rb_node *node;
3139
3140 if (sched->skip_merge)
3141 return;
3142
cb4c13a5
DB		 3143 while ((node = rb_first_cached(&sched->atom_root))) {
3144 rb_erase_cached(node, &sched->atom_root);
2f80dd44
JB		 3145 data = rb_entry(node, struct work_atoms, node);
3146 __merge_work_atoms(&sched->merged_atom_root, data);
3147 }
3148}
3149
0e9b07e5 3150static int perf_sched__lat(struct perf_sched *sched)
0ec04e16
IM		 3151{
3152 struct rb_node *next;
3153
3154 setup_pager();
ad9def7c 3155
ae536acf 3156 if (perf_sched__read_events(sched))
a116e05d 3157 return -1;
ad9def7c 3158
2f80dd44 3159 perf_sched__merge_lat(sched);
0e9b07e5 3160 perf_sched__sort_lat(sched);
0ec04e16 3161
80790e0b
RR		 3162 printf("\n -----------------------------------------------------------------------------------------------------------------\n");
3163 printf(" Task | Runtime ms | Switches | Average delay ms | Maximum delay ms | Maximum delay at |\n");
3164 printf(" -----------------------------------------------------------------------------------------------------------------\n");
0ec04e16 3165
cb4c13a5 3166 next = rb_first_cached(&sched->sorted_atom_root);
0ec04e16
IM		 3167
3168 while (next) {
3169 struct work_atoms *work_list;
3170
3171 work_list = rb_entry(next, struct work_atoms, node);
0e9b07e5 3172 output_lat_thread(sched, work_list);
0ec04e16 3173 next = rb_next(next);
ae536acf 3174 thread__zput(work_list->thread);
0ec04e16
IM		 3175 }
3176
80790e0b 3177 printf(" -----------------------------------------------------------------------------------------------------------------\n");
9486aa38 3178 printf(" TOTAL: |%11.3f ms |%9" PRIu64 " |\n",
4fc76e49 3179 (double)sched->all_runtime / NSEC_PER_MSEC, sched->all_count);
0ec04e16
IM		 3180
3181 printf(" ---------------------------------------------------\n");
3182
0e9b07e5 3183 print_bad_events(sched);
0ec04e16
IM		 3184 printf("\n");
3185
a116e05d 3186 return 0;
0ec04e16
IM		 3187}
3188
99623c62
JO		 3189static int setup_map_cpus(struct perf_sched *sched)
3190{
f854839b 3191 struct perf_cpu_map *map;
73643bb6 3192
99623c62
JO		 3193 sched->max_cpu = sysconf(_SC_NPROCESSORS_CONF);
3194
3195 if (sched->map.comp) {
3196 sched->map.comp_cpus = zalloc(sched->max_cpu * sizeof(int));
cf294f24
JO		 3197 if (!sched->map.comp_cpus)
3198 return -1;
99623c62
JO		 3199 }
3200
73643bb6
JO		 3201 if (!sched->map.cpus_str)
3202 return 0;
3203
9c3516d1 3204 map = perf_cpu_map__new(sched->map.cpus_str);
73643bb6
JO		 3205 if (!map) {
3206 pr_err("failed to get cpus map from %s\n", sched->map.cpus_str);
3207 return -1;
3208 }
3209
3210 sched->map.cpus = map;
99623c62
JO		 3211 return 0;
3212}
3213
a151a37a
JO		 3214static int setup_color_pids(struct perf_sched *sched)
3215{
9749b90e 3216 struct perf_thread_map *map;
a151a37a
JO		 3217
3218 if (!sched->map.color_pids_str)
3219 return 0;
3220
3221 map = thread_map__new_by_tid_str(sched->map.color_pids_str);
3222 if (!map) {
3223 pr_err("failed to get thread map from %s\n", sched->map.color_pids_str);
3224 return -1;
3225 }
3226
3227 sched->map.color_pids = map;
3228 return 0;
3229}
3230
cf294f24
JO		 3231static int setup_color_cpus(struct perf_sched *sched)
3232{
f854839b 3233 struct perf_cpu_map *map;
cf294f24
JO		 3234
3235 if (!sched->map.color_cpus_str)
3236 return 0;
3237
9c3516d1 3238 map = perf_cpu_map__new(sched->map.color_cpus_str);
cf294f24
JO		 3239 if (!map) {
3240 pr_err("failed to get thread map from %s\n", sched->map.color_cpus_str);
3241 return -1;
3242 }
3243
3244 sched->map.color_cpus = map;
3245 return 0;
3246}
3247
0e9b07e5 3248static int perf_sched__map(struct perf_sched *sched)
0ec04e16 3249{
99623c62
JO		 3250 if (setup_map_cpus(sched))
3251 return -1;
40749d0f 3252
a151a37a
JO		 3253 if (setup_color_pids(sched))
3254 return -1;
3255
cf294f24
JO		 3256 if (setup_color_cpus(sched))
3257 return -1;
3258
0ec04e16 3259 setup_pager();
ae536acf 3260 if (perf_sched__read_events(sched))
a116e05d 3261 return -1;
0e9b07e5 3262 print_bad_events(sched);
a116e05d 3263 return 0;
0ec04e16
IM		 3264}
3265
0e9b07e5 3266static int perf_sched__replay(struct perf_sched *sched)
0ec04e16
IM		 3267{
3268 unsigned long i;
3269
0e9b07e5
ACM		 3270 calibrate_run_measurement_overhead(sched);
3271 calibrate_sleep_measurement_overhead(sched);
0ec04e16 3272
0e9b07e5 3273 test_calibrations(sched);
0ec04e16 3274
ae536acf 3275 if (perf_sched__read_events(sched))
a116e05d 3276 return -1;
0ec04e16 3277
0e9b07e5
ACM		 3278 printf("nr_run_events: %ld\n", sched->nr_run_events);
3279 printf("nr_sleep_events: %ld\n", sched->nr_sleep_events);
3280 printf("nr_wakeup_events: %ld\n", sched->nr_wakeup_events);
0ec04e16 3281
0e9b07e5
ACM		 3282 if (sched->targetless_wakeups)
3283 printf("target-less wakeups: %ld\n", sched->targetless_wakeups);
3284 if (sched->multitarget_wakeups)
3285 printf("multi-target wakeups: %ld\n", sched->multitarget_wakeups);
3286 if (sched->nr_run_events_optimized)
0ec04e16 3287 printf("run atoms optimized: %ld\n",
0e9b07e5 3288 sched->nr_run_events_optimized);
0ec04e16 3289
0e9b07e5
ACM		 3290 print_task_traces(sched);
3291 add_cross_task_wakeups(sched);
0ec04e16 3292
0e9b07e5 3293 create_tasks(sched);
0ec04e16 3294 printf("------------------------------------------------------------\n");
0e9b07e5
ACM		 3295 for (i = 0; i < sched->replay_repeat; i++)
3296 run_one_test(sched);
a116e05d
ACM		 3297
3298 return 0;
0ec04e16
IM		 3299}
3300
0e9b07e5
ACM		 3301static void setup_sorting(struct perf_sched *sched, const struct option *options,
3302 const char * const usage_msg[])
daa1d7a5 3303{
0e9b07e5 3304 char *tmp, *tok, *str = strdup(sched->sort_order);
daa1d7a5
FW		 3305
3306 for (tok = strtok_r(str, ", ", &tmp);
3307 tok; tok = strtok_r(NULL, ", ", &tmp)) {
0e9b07e5 3308 if (sort_dimension__add(tok, &sched->sort_list) < 0) {
c7118369
NK		 3309 usage_with_options_msg(usage_msg, options,
3310 "Unknown --sort key: `%s'", tok);
daa1d7a5
FW		 3311 }
3312 }
3313
3314 free(str);
3315
0e9b07e5 3316 sort_dimension__add("pid", &sched->cmp_pid);
daa1d7a5
FW		 3317}
3318
1fc35b29
IM		 3319static int __cmd_record(int argc, const char **argv)
3320{
3321 unsigned int rec_argc, i, j;
3322 const char **rec_argv;
0e9b07e5
ACM		 3323 const char * const record_args[] = {
3324 "record",
3325 "-a",
3326 "-R",
0e9b07e5
ACM		 3327 "-m", "1024",
3328 "-c", "1",
3329 "-e", "sched:sched_switch",
3330 "-e", "sched:sched_stat_wait",
3331 "-e", "sched:sched_stat_sleep",
3332 "-e", "sched:sched_stat_iowait",
3333 "-e", "sched:sched_stat_runtime",
0e9b07e5
ACM		 3334 "-e", "sched:sched_process_fork",
3335 "-e", "sched:sched_wakeup",
7fff9597 3336 "-e", "sched:sched_wakeup_new",
0e9b07e5
ACM		 3337 "-e", "sched:sched_migrate_task",
3338 };
1fc35b29
IM		 3339
3340 rec_argc = ARRAY_SIZE(record_args) + argc - 1;
3341 rec_argv = calloc(rec_argc + 1, sizeof(char *));
3342
e462dc55 3343 if (rec_argv == NULL)
ce47dc56
CS		 3344 return -ENOMEM;
3345
1fc35b29
IM		 3346 for (i = 0; i < ARRAY_SIZE(record_args); i++)
3347 rec_argv[i] = strdup(record_args[i]);
3348
3349 for (j = 1; j < (unsigned int)argc; j++, i++)
3350 rec_argv[i] = argv[j];
3351
3352 BUG_ON(i != rec_argc);
3353
b0ad8ea6 3354 return cmd_record(i, rec_argv);
1fc35b29
IM		 3355}
3356
b0ad8ea6 3357int cmd_sched(int argc, const char **argv)
0a02ad93 3358{
49b8e2be 3359 static const char default_sort_order[] = "avg, max, switch, runtime";
8a39df8f
AH		 3360 struct perf_sched sched = {
3361 .tool = {
3362 .sample = perf_sched__process_tracepoint_sample,
99a3c3a9 3363 .comm = perf_sched__process_comm,
f3b3614a 3364 .namespaces = perf_event__process_namespaces,
8a39df8f
AH		 3365 .lost = perf_event__process_lost,
3366 .fork = perf_sched__process_fork_event,
0a8cb85c 3367 .ordered_events = true,
8a39df8f
AH		 3368 },
3369 .cmp_pid = LIST_HEAD_INIT(sched.cmp_pid),
3370 .sort_list = LIST_HEAD_INIT(sched.sort_list),
3371 .start_work_mutex = PTHREAD_MUTEX_INITIALIZER,
3372 .work_done_wait_mutex = PTHREAD_MUTEX_INITIALIZER,
8a39df8f
AH		 3373 .sort_order = default_sort_order,
3374 .replay_repeat = 10,
3375 .profile_cpu = -1,
3376 .next_shortname1 = 'A',
3377 .next_shortname2 = '0',
2f80dd44 3378 .skip_merge = 0,
6c973c90
DA		 3379 .show_callchain = 1,
3380 .max_stack = 5,
8a39df8f 3381 };
77f02f44
NK		 3382 const struct option sched_options[] = {
3383 OPT_STRING('i', "input", &input_name, "file",
3384 "input file name"),
3385 OPT_INCR('v', "verbose", &verbose,
3386 "be more verbose (show symbol address, etc)"),
3387 OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace,
3388 "dump raw trace in ASCII"),
6fa94258 3389 OPT_BOOLEAN('f', "force", &sched.force, "don't complain, do it"),
77f02f44
NK		 3390 OPT_END()
3391 };
0e9b07e5
ACM		 3392 const struct option latency_options[] = {
3393 OPT_STRING('s', "sort", &sched.sort_order, "key[,key2...]",
3394 "sort by key(s): runtime, switch, avg, max"),
0e9b07e5
ACM		 3395 OPT_INTEGER('C', "CPU", &sched.profile_cpu,
3396 "CPU to profile on"),
2f80dd44
JB		 3397 OPT_BOOLEAN('p', "pids", &sched.skip_merge,
3398 "latency stats per pid instead of per comm"),
77f02f44 3399 OPT_PARENT(sched_options)
0e9b07e5
ACM		 3400 };
3401 const struct option replay_options[] = {
3402 OPT_UINTEGER('r', "repeat", &sched.replay_repeat,
3403 "repeat the workload replay N times (-1: infinite)"),
77f02f44 3404 OPT_PARENT(sched_options)
0e9b07e5 3405 };
99623c62
JO		 3406 const struct option map_options[] = {
3407 OPT_BOOLEAN(0, "compact", &sched.map.comp,
3408 "map output in compact mode"),
a151a37a
JO		 3409 OPT_STRING(0, "color-pids", &sched.map.color_pids_str, "pids",
3410 "highlight given pids in map"),
cf294f24
JO		 3411 OPT_STRING(0, "color-cpus", &sched.map.color_cpus_str, "cpus",
3412 "highlight given CPUs in map"),
73643bb6
JO		 3413 OPT_STRING(0, "cpus", &sched.map.cpus_str, "cpus",
3414 "display given CPUs in map"),
77f02f44 3415 OPT_PARENT(sched_options)
99623c62 3416 };
49394a2a
DA		 3417 const struct option timehist_options[] = {
3418 OPT_STRING('k', "vmlinux", &symbol_conf.vmlinux_name,
3419 "file", "vmlinux pathname"),
3420 OPT_STRING(0, "kallsyms", &symbol_conf.kallsyms_name,
3421 "file", "kallsyms pathname"),
6c973c90
DA		 3422 OPT_BOOLEAN('g', "call-graph", &sched.show_callchain,
3423 "Display call chains if present (default on)"),
3424 OPT_UINTEGER(0, "max-stack", &sched.max_stack,
3425 "Maximum number of functions to display backtrace."),
49394a2a
DA		 3426 OPT_STRING(0, "symfs", &symbol_conf.symfs, "directory",
3427 "Look for files with symbols relative to this directory"),
52df138c
DA		 3428 OPT_BOOLEAN('s', "summary", &sched.summary_only,
3429 "Show only syscall summary with statistics"),
3430 OPT_BOOLEAN('S', "with-summary", &sched.summary,
3431 "Show all syscalls and summary with statistics"),
fc1469f1 3432 OPT_BOOLEAN('w', "wakeups", &sched.show_wakeups, "Show wakeup events"),
292c4a8f 3433 OPT_BOOLEAN('n', "next", &sched.show_next, "Show next task"),
350f54fa 3434 OPT_BOOLEAN('M', "migrations", &sched.show_migrations, "Show migration events"),
a407b067 3435 OPT_BOOLEAN('V', "cpu-visual", &sched.show_cpu_visual, "Add CPU visual"),
07235f84 3436 OPT_BOOLEAN('I', "idle-hist", &sched.idle_hist, "Show idle events only"),
853b7407
DA		 3437 OPT_STRING(0, "time", &sched.time_str, "str",
3438 "Time span for analysis (start,stop)"),
414e050c 3439 OPT_BOOLEAN(0, "state", &sched.show_state, "Show task state when sched-out"),
0f59d7a3
DA		 3440 OPT_STRING('p', "pid", &symbol_conf.pid_list_str, "pid[,pid...]",
3441 "analyze events only for given process id(s)"),
3442 OPT_STRING('t', "tid", &symbol_conf.tid_list_str, "tid[,tid...]",
3443 "analyze events only for given thread id(s)"),
c30d630d 3444 OPT_STRING('C', "cpu", &cpu_list, "cpu", "list of cpus to profile"),
49394a2a
DA		 3445 OPT_PARENT(sched_options)
3446 };
3447
0e9b07e5
ACM		 3448 const char * const latency_usage[] = {
3449 "perf sched latency [<options>]",
3450 NULL
3451 };
3452 const char * const replay_usage[] = {
3453 "perf sched replay [<options>]",
3454 NULL
3455 };
99623c62
JO		 3456 const char * const map_usage[] = {
3457 "perf sched map [<options>]",
3458 NULL
3459 };
49394a2a
DA		 3460 const char * const timehist_usage[] = {
3461 "perf sched timehist [<options>]",
3462 NULL
3463 };
a83edb2d 3464 const char *const sched_subcommands[] = { "record", "latency", "map",
49394a2a
DA		 3465 "replay", "script",
3466 "timehist", NULL };
a83edb2d
RR		 3467 const char *sched_usage[] = {
3468 NULL,
0e9b07e5
ACM		 3469 NULL
3470 };
3471 struct trace_sched_handler lat_ops = {
3472 .wakeup_event = latency_wakeup_event,
3473 .switch_event = latency_switch_event,
3474 .runtime_event = latency_runtime_event,
0e9b07e5
ACM		 3475 .migrate_task_event = latency_migrate_task_event,
3476 };
3477 struct trace_sched_handler map_ops = {
3478 .switch_event = map_switch_event,
3479 };
3480 struct trace_sched_handler replay_ops = {
3481 .wakeup_event = replay_wakeup_event,
3482 .switch_event = replay_switch_event,
3483 .fork_event = replay_fork_event,
3484 };
156a2b02
AH		 3485 unsigned int i;
3486
3487 for (i = 0; i < ARRAY_SIZE(sched.curr_pid); i++)
3488 sched.curr_pid[i] = -1;
0e9b07e5 3489
a83edb2d
RR		 3490 argc = parse_options_subcommand(argc, argv, sched_options, sched_subcommands,
3491 sched_usage, PARSE_OPT_STOP_AT_NON_OPTION);
f2858d8a
IM		 3492 if (!argc)
3493 usage_with_options(sched_usage, sched_options);
0a02ad93 3494
c0777c5a 3495 /*
133dc4c3 3496 * Aliased to 'perf script' for now:
c0777c5a 3497 */
133dc4c3 3498 if (!strcmp(argv[0], "script"))
b0ad8ea6 3499 return cmd_script(argc, argv);
c0777c5a 3500
1fc35b29
IM		 3501 if (!strncmp(argv[0], "rec", 3)) {
3502 return __cmd_record(argc, argv);
3503 } else if (!strncmp(argv[0], "lat", 3)) {
0e9b07e5 3504 sched.tp_handler = &lat_ops;
f2858d8a
IM		 3505 if (argc > 1) {
3506 argc = parse_options(argc, argv, latency_options, latency_usage, 0);
3507 if (argc)
3508 usage_with_options(latency_usage, latency_options);
f2858d8a 3509 }
0e9b07e5
ACM		 3510 setup_sorting(&sched, latency_options, latency_usage);
3511 return perf_sched__lat(&sched);
0ec04e16 3512 } else if (!strcmp(argv[0], "map")) {
99623c62 3513 if (argc) {
a151a37a 3514 argc = parse_options(argc, argv, map_options, map_usage, 0);
99623c62
JO		 3515 if (argc)
3516 usage_with_options(map_usage, map_options);
3517 }
0e9b07e5
ACM		 3518 sched.tp_handler = &map_ops;
3519 setup_sorting(&sched, latency_options, latency_usage);
3520 return perf_sched__map(&sched);
f2858d8a 3521 } else if (!strncmp(argv[0], "rep", 3)) {
0e9b07e5 3522 sched.tp_handler = &replay_ops;
f2858d8a
IM		 3523 if (argc) {
3524 argc = parse_options(argc, argv, replay_options, replay_usage, 0);
3525 if (argc)
3526 usage_with_options(replay_usage, replay_options);
3527 }
0e9b07e5 3528 return perf_sched__replay(&sched);
49394a2a
DA		 3529 } else if (!strcmp(argv[0], "timehist")) {
3530 if (argc) {
3531 argc = parse_options(argc, argv, timehist_options,
3532 timehist_usage, 0);
3533 if (argc)
3534 usage_with_options(timehist_usage, timehist_options);
3535 }
292c4a8f
BG		 3536 if ((sched.show_wakeups || sched.show_next) &&
3537 sched.summary_only) {
3538 pr_err(" Error: -s and -[n|w] are mutually exclusive.\n");
fc1469f1 3539 parse_options_usage(timehist_usage, timehist_options, "s", true);
292c4a8f
BG		 3540 if (sched.show_wakeups)
3541 parse_options_usage(NULL, timehist_options, "w", true);
3542 if (sched.show_next)
3543 parse_options_usage(NULL, timehist_options, "n", true);
fc1469f1
DA		 3544 return -EINVAL;
3545 }
3546
49394a2a 3547 return perf_sched__timehist(&sched);
f2858d8a
IM		 3548 } else {
3549 usage_with_options(sched_usage, sched_options);
3550 }
3551
ec156764 3552 return 0;
0a02ad93 3553}
Ubuntu Jammy Linux kernel mirror