aboutsummaryrefslogtreecommitdiff
path: root/tools/perf/builtin-timechart.c
blob: dd4d82ac7aa4ac67a51e6d2eedfc653c8c35832c (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
/*
 * builtin-timechart.c - make an svg timechart of system activity
 *
 * (C) Copyright 2009 Intel Corporation
 *
 * Authors:
 *     Arjan van de Ven <arjan@linux.intel.com>
 *
 * This program is free software; you can redistribute it and/or
 * modify it under the terms of the GNU General Public License
 * as published by the Free Software Foundation; version 2
 * of the License.
 */

#include "builtin.h"

#include "util/util.h"

#include "util/color.h"
#include <linux/list.h>
#include "util/cache.h"
#include <linux/rbtree.h>
#include "util/symbol.h"
#include "util/string.h"
#include "util/callchain.h"
#include "util/strlist.h"

#include "perf.h"
#include "util/header.h"
#include "util/parse-options.h"
#include "util/parse-events.h"
#include "util/svghelper.h"

static char		const *input_name = "perf.data";
static char		const *output_name = "output.svg";


static unsigned long	page_size;
static unsigned long	mmap_window = 32;
static u64		sample_type;

static unsigned int	numcpus;
static u64		min_freq;	/* Lowest CPU frequency seen */
static u64		max_freq;	/* Highest CPU frequency seen */
static u64		turbo_frequency;

static u64		first_time, last_time;

static int		power_only;


static struct perf_header	*header;

struct per_pid;
struct per_pidcomm;

struct cpu_sample;
struct power_event;
struct wake_event;

struct sample_wrapper;

/*
 * Datastructure layout:
 * We keep an list of "pid"s, matching the kernels notion of a task struct.
 * Each "pid" entry, has a list of "comm"s.
 *	this is because we want to track different programs different, while
 *	exec will reuse the original pid (by design).
 * Each comm has a list of samples that will be used to draw
 * final graph.
 */

struct per_pid {
	struct per_pid *next;

	int		pid;
	int		ppid;

	u64		start_time;
	u64		end_time;
	u64		total_time;
	int		display;

	struct per_pidcomm *all;
	struct per_pidcomm *current;

	int painted;
};


struct per_pidcomm {
	struct per_pidcomm *next;

	u64		start_time;
	u64		end_time;
	u64		total_time;

	int		Y;
	int		display;

	long		state;
	u64		state_since;

	char		*comm;

	struct cpu_sample *samples;
};

struct sample_wrapper {
	struct sample_wrapper *next;

	u64		timestamp;
	unsigned char	data[0];
};

#define TYPE_NONE	0
#define TYPE_RUNNING	1
#define TYPE_WAITING	2
#define TYPE_BLOCKED	3

struct cpu_sample {
	struct cpu_sample *next;

	u64 start_time;
	u64 end_time;
	int type;
	int cpu;
};

static struct per_pid *all_data;

#define CSTATE 1
#define PSTATE 2

struct power_event {
	struct power_event *next;
	int type;
	int state;
	u64 start_time;
	u64 end_time;
	int cpu;
};

struct wake_event {
	struct wake_event *next;
	int waker;
	int wakee;
	u64 time;
};

static struct power_event    *power_events;
static struct wake_event     *wake_events;

struct sample_wrapper *all_samples;


struct process_filter;
struct process_filter {
	char 			*name;
	int  			pid;
	struct process_filter 	*next;
};

static struct process_filter *process_filter;


static struct per_pid *find_create_pid(int pid)
{
	struct per_pid *cursor = all_data;

	while (cursor) {
		if (cursor->pid == pid)
			return cursor;
		cursor = cursor->next;
	}
	cursor = malloc(sizeof(struct per_pid));
	assert(cursor != NULL);
	memset(cursor, 0, sizeof(struct per_pid));
	cursor->pid = pid;
	cursor->next = all_data;
	all_data = cursor;
	return cursor;
}

static void pid_set_comm(int pid, char *comm)
{
	struct per_pid *p;
	struct per_pidcomm *c;
	p = find_create_pid(pid);
	c = p->all;
	while (c) {
		if (c->comm && strcmp(c->comm, comm) == 0) {
			p->current = c;
			return;
		}
		if (!c->comm) {
			c->comm = strdup(comm);
			p->current = c;
			return;
		}
		c = c->next;
	}
	c = malloc(sizeof(struct per_pidcomm));
	assert(c != NULL);
	memset(c, 0, sizeof(struct per_pidcomm));
	c->comm = strdup(comm);
	p->current = c;
	c->next = p->all;
	p->all = c;
}

static void pid_fork(int pid, int ppid, u64 timestamp)
{
	struct per_pid *p, *pp;
	p = find_create_pid(pid);
	pp = find_create_pid(ppid);
	p->ppid = ppid;
	if (pp->current && pp->current->comm && !p->current)
		pid_set_comm(pid, pp->current->comm);

	p->start_time = timestamp;
	if (p->current) {
		p->current->start_time = timestamp;
		p->current->state_since = timestamp;
	}
}

static void pid_exit(int pid, u64 timestamp)
{
	struct per_pid *p;
	p = find_create_pid(pid);
	p->end_time = timestamp;
	if (p->current)
		p->current->end_time = timestamp;
}

static void
pid_put_sample(int pid, int type, unsigned int cpu, u64 start, u64 end)
{
	struct per_pid *p;
	struct per_pidcomm *c;
	struct cpu_sample *sample;

	p = find_create_pid(pid);
	c = p->current;
	if (!c) {
		c = malloc(sizeof(struct per_pidcomm));
		assert(c != NULL);
		memset(c, 0, sizeof(struct per_pidcomm));
		p->current = c;
		c->next = p->all;
		p->all = c;
	}

	sample = malloc(sizeof(struct cpu_sample));
	assert(sample != NULL);
	memset(sample, 0, sizeof(struct cpu_sample));
	sample->start_time = start;
	sample->end_time = end;
	sample->type = type;
	sample->next = c->samples;
	sample->cpu = cpu;
	c->samples = sample;

	if (sample->type == TYPE_RUNNING && end > start && start > 0) {
		c->total_time += (end-start);
		p->total_time += (end-start);
	}

	if (c->start_time == 0 || c->start_time > start)
		c->start_time = start;
	if (p->start_time == 0 || p->start_time > start)
		p->start_time = start;

	if (cpu > numcpus)
		numcpus = cpu;
}

#define MAX_CPUS 4096

static u64 cpus_cstate_start_times[MAX_CPUS];
static int cpus_cstate_state[MAX_CPUS];
static u64 cpus_pstate_start_times[MAX_CPUS];
static u64 cpus_pstate_state[MAX_CPUS];

static int
process_comm_event(event_t *event)
{
	pid_set_comm(event->comm.pid, event->comm.comm);
	return 0;
}
static int
process_fork_event(event_t *event)
{
	pid_fork(event->fork.pid, event->fork.ppid, event->fork.time);
	return 0;
}

static int
process_exit_event(event_t *event)
{
	pid_exit(event->fork.pid, event->fork.time);
	return 0;
}

struct trace_entry {
	u32			size;
	unsigned short		type;
	unsigned char		flags;
	unsigned char		preempt_count;
	int			pid;
	int			tgid;
};

struct power_entry {
	struct trace_entry te;
	s64	type;
	s64	value;
};

#define TASK_COMM_LEN 16
struct wakeup_entry {
	struct trace_entry te;
	char comm[TASK_COMM_LEN];
	int   pid;
	int   prio;
	int   success;
};

/*
 * trace_flag_type is an enumeration that holds different
 * states when a trace occurs. These are:
 *  IRQS_OFF            - interrupts were disabled
 *  IRQS_NOSUPPORT      - arch does not support irqs_disabled_flags
 *  NEED_RESCED         - reschedule is requested
 *  HARDIRQ             - inside an interrupt handler
 *  SOFTIRQ             - inside a softirq handler
 */
enum trace_flag_type {
	TRACE_FLAG_IRQS_OFF		= 0x01,
	TRACE_FLAG_IRQS_NOSUPPORT	= 0x02,
	TRACE_FLAG_NEED_RESCHED		= 0x04,
	TRACE_FLAG_HARDIRQ		= 0x08,
	TRACE_FLAG_SOFTIRQ		= 0x10,
};



struct sched_switch {
	struct trace_entry te;
	char prev_comm[TASK_COMM_LEN];
	int  prev_pid;
	int  prev_prio;
	long prev_state; /* Arjan weeps. */
	char next_comm[TASK_COMM_LEN];
	int  next_pid;
	int  next_prio;
};

static void c_state_start(int cpu, u64 timestamp, int state)
{
	cpus_cstate_start_times[cpu] = timestamp;
	cpus_cstate_state[cpu] = state;
}

static void c_state_end(int cpu, u64 timestamp)
{
	struct power_event *pwr;
	pwr = malloc(sizeof(struct power_event));
	if (!pwr)
		return;
	memset(pwr, 0, sizeof(struct power_event));

	pwr->state = cpus_cstate_state[cpu];
	pwr->start_time = cpus_cstate_start_times[cpu];
	pwr->end_time = timestamp;
	pwr->cpu = cpu;
	pwr->type = CSTATE;
	pwr->next = power_events;

	power_events = pwr;
}

static void p_state_change(int cpu, u64 timestamp, u64 new_freq)
{
	struct power_event *pwr;
	pwr = malloc(sizeof(struct power_event));

	if (new_freq > 8000000) /* detect invalid data */
		return;

	if (!pwr)
		return;
	memset(pwr, 0, sizeof(struct power_event));

	pwr->state = cpus_pstate_state[cpu];
	pwr->start_time = cpus_pstate_start_times[cpu];
	pwr->end_time = timestamp;
	pwr->cpu = cpu;
	pwr->type = PSTATE;
	pwr->next = power_events;

	if (!pwr->start_time)
		pwr->start_time = first_time;

	power_events = pwr;

	cpus_pstate_state[cpu] = new_freq;
	cpus_pstate_start_times[cpu] = timestamp;

	if ((u64)new_freq > max_freq)
		max_freq = new_freq;

	if (new_freq < min_freq || min_freq == 0)
		min_freq = new_freq;

	if (new_freq == max_freq - 1000)
			turbo_frequency = max_freq;
}

static void
sched_wakeup(int cpu, u64 timestamp, int pid, struct trace_entry *te)
{
	struct wake_event *we;
	struct per_pid *p;
	struct wakeup_entry *wake = (void *)te;

	we = malloc(sizeof(struct wake_event));
	if (!we)
		return;

	memset(we, 0, sizeof(struct wake_event));
	we->time = timestamp;
	we->waker = pid;

	if ((te->flags & TRACE_FLAG_HARDIRQ) || (te->flags & TRACE_FLAG_SOFTIRQ))
		we->waker = -1;

	we->wakee = wake->pid;
	we->next = wake_events;
	wake_events = we;
	p = find_create_pid(we->wakee);

	if (p && p->current && p->current->state == TYPE_NONE) {
		p->current->state_since = timestamp;
		p->current->state = TYPE_WAITING;
	}
	if (p && p->current && p->current->state == TYPE_BLOCKED) {
		pid_put_sample(p->pid, p->current->state, cpu, p->current->state_since, timestamp);
		p->current->state_since = timestamp;
		p->current->state = TYPE_WAITING;
	}
}

static void sched_switch(int cpu, u64 timestamp, struct trace_entry *te)
{
	struct per_pid *p = NULL, *prev_p;
	struct sched_switch *sw = (void *)te;


	prev_p = find_create_pid(sw->prev_pid);

	p = find_create_pid(sw->next_pid);

	if (prev_p->current && prev_p->current->state != TYPE_NONE)
		pid_put_sample(sw->prev_pid, TYPE_RUNNING, cpu, prev_p->current->state_since, timestamp);
	if (p && p->current) {
		if (p->current->state != TYPE_NONE)
			pid_put_sample(sw->next_pid, p->current->state, cpu, p->current->state_since, timestamp);

			p->current->state_since = timestamp;
			p->current->state = TYPE_RUNNING;
	}

	if (prev_p->current) {
		prev_p->current->state = TYPE_NONE;
		prev_p->current->state_since = timestamp;
		if (sw->prev_state & 2)
			prev_p->current->state = TYPE_BLOCKED;
		if (sw->prev_state == 0)
			prev_p->current->state = TYPE_WAITING;
	}
}


static int
process_sample_event(event_t *event)
{
	int cursor = 0;
	u64 addr = 0;
	u64 stamp = 0;
	u32 cpu = 0;
	u32 pid = 0;
	struct trace_entry *te;

	if (sample_type & PERF_SAMPLE_IP)
		cursor++;

	if (sample_type & PERF_SAMPLE_TID) {
		pid = event->sample.array[cursor]>>32;
		cursor++;
	}
	if (sample_type & PERF_SAMPLE_TIME) {
		stamp = event->sample.array[cursor++];

		if (!first_time || first_time > stamp)
			first_time = stamp;
		if (last_time < stamp)
			last_time = stamp;

	}
	if (sample_type & PERF_SAMPLE_ADDR)
		addr = event->sample.array[cursor++];
	if (sample_type & PERF_SAMPLE_ID)
		cursor++;
	if (sample_type & PERF_SAMPLE_STREAM_ID)
		cursor++;
	if (sample_type & PERF_SAMPLE_CPU)
		cpu = event->sample.array[cursor++] & 0xFFFFFFFF;
	if (sample_type & PERF_SAMPLE_PERIOD)
		cursor++;

	te = (void *)&event->sample.array[cursor];

	if (sample_type & PERF_SAMPLE_RAW && te->size > 0) {
		char *event_str;
		struct power_entry *pe;

		pe = (void *)te;

		event_str = perf_header__find_event(te->type);

		if (!event_str)
			return 0;

		if (strcmp(event_str, "power:power_start") == 0)
			c_state_start(cpu, stamp, pe->value);

		if (strcmp(event_str, "power:power_end") == 0)
			c_state_end(cpu, stamp);

		if (strcmp(event_str, "power:power_frequency") == 0)
			p_state_change(cpu, stamp, pe->value);

		if (strcmp(event_str, "sched:sched_wakeup") == 0)
			sched_wakeup(cpu, stamp, pid, te);

		if (strcmp(event_str, "sched:sched_switch") == 0)
			sched_switch(cpu, stamp, te);
	}
	return 0;
}

/*
 * After the last sample we need to wrap up the current C/P state
 * and close out each CPU for these.
 */
static void end_sample_processing(void)
{
	u64 cpu;
	struct power_event *pwr;

	for (cpu = 0; cpu <= numcpus; cpu++) {
		pwr = malloc(sizeof(struct power_event));
		if (!pwr)
			return;
		memset(pwr, 0, sizeof(struct power_event));

		/* C state */
#if 0
		pwr->state = cpus_cstate_state[cpu];
		pwr->start_time = cpus_cstate_start_times[cpu];
		pwr->end_time = last_time;
		pwr->cpu = cpu;
		pwr->type = CSTATE;
		pwr->next = power_events;

		power_events = pwr;
#endif
		/* P state */

		pwr = malloc(sizeof(struct power_event));
		if (!pwr)
			return;
		memset(pwr, 0, sizeof(struct power_event));

		pwr->state = cpus_pstate_state[cpu];
		pwr->start_time = cpus_pstate_start_times[cpu];
		pwr->end_time = last_time;
		pwr->cpu = cpu;
		pwr->type = PSTATE;
		pwr->next = power_events;

		if (!pwr->start_time)
			pwr->start_time = first_time;
		if (!pwr->state)
			pwr->state = min_freq;
		power_events = pwr;
	}
}

static u64 sample_time(event_t *event)
{
	int cursor;

	cursor = 0;
	if (sample_type & PERF_SAMPLE_IP)
		cursor++;
	if (sample_type & PERF_SAMPLE_TID)
		cursor++;
	if (sample_type & PERF_SAMPLE_TIME)
		return event->sample.array[cursor];
	return 0;
}


/*
 * We first queue all events, sorted backwards by insertion.
 * The order will get flipped later.
 */
static int
queue_sample_event(event_t *event)
{
	struct sample_wrapper *copy, *prev;
	int size;

	size = event->sample.header.size + sizeof(struct sample_wrapper) + 8;

	copy = malloc(size);
	if (!copy)
		return 1;

	memset(copy, 0, size);

	copy->next = NULL;
	copy->timestamp = sample_time(event);

	memcpy(&copy->data, event, event->sample.header.size);

	/* insert in the right place in the list */

	if (!all_samples) {
		/* first sample ever */
		all_samples = copy;
		return 0;
	}

	if (all_samples->timestamp < copy->timestamp) {
		/* insert at the head of the list */
		copy->next = all_samples;
		all_samples = copy;
		return 0;
	}

	prev = all_samples;
	while (prev->next) {
		if (prev->next->timestamp < copy->timestamp) {
			copy->next = prev->next;
			prev->next = copy;
			return 0;
		}
		prev = prev->next;
	}
	/* insert at the end of the list */
	prev->next = copy;

	return 0;
}

static void sort_queued_samples(void)
{
	struct sample_wrapper *cursor, *next;

	cursor = all_samples;
	all_samples = NULL;

	while (cursor) {
		next = cursor->next;
		cursor->next = all_samples;
		all_samples = cursor;
		cursor = next;
	}
}

/*
 * Sort the pid datastructure
 */
static void sort_pids(void)
{
	struct per_pid *new_list, *p, *cursor, *prev;
	/* sort by ppid first, then by pid, lowest to highest */

	new_list = NULL;

	while (all_data) {
		p = all_data;
		all_data = p->next;
		p->next = NULL;

		if (new_list == NULL) {
			new_list = p;
			p->next = NULL;
			continue;
		}
		prev = NULL;
		cursor = new_list;
		while (cursor) {
			if (cursor->ppid > p->ppid ||
				(cursor->ppid == p->ppid && cursor->pid > p->pid)) {
				/* must insert before */
				if (prev) {
					p->next = prev->next;
					prev->next = p;
					cursor = NULL;
					continue;
				} else {
					p->next = new_list;
					new_list = p;
					cursor = NULL;
					continue;
				}
			}

			prev = cursor;
			cursor = cursor->next;
			if (!cursor)
				prev->next = p;
		}
	}
	all_data = new_list;
}


static void draw_c_p_states(void)
{
	struct power_event *pwr;
	pwr = power_events;

	/*
	 * two pass drawing so that the P state bars are on top of the C state blocks
	 */
	while (pwr) {
		if (pwr->type == CSTATE)
			svg_cstate(pwr->cpu, pwr->start_time, pwr->end_time, pwr->state);
		pwr = pwr->next;
	}

	pwr = power_events;
	while (pwr) {
		if (pwr->type == PSTATE) {
			if (!pwr->state)
				pwr->state = min_freq;
			svg_pstate(pwr->cpu, pwr->start_time, pwr->end_time, pwr->state);
		}
		pwr = pwr->next;
	}
}

static void draw_wakeups(void)
{
	struct wake_event *we;
	struct per_pid *p;
	struct per_pidcomm *c;

	we = wake_events;
	while (we) {
		int from = 0, to = 0;
		char *task_from = NULL, *task_to = NULL;

		/* locate the column of the waker and wakee */
		p = all_data;
		while (p) {
			if (p->pid == we->waker || p->pid == we->wakee) {
				c = p->all;
				while (c) {
					if (c->Y && c->start_time <= we->time && c->end_time >= we->time) {
						if (p->pid == we->waker && !from) {
							from = c->Y;
							task_from = strdup(c->comm);
						}
						if (p->pid == we->wakee && !to) {
							to = c->Y;
							task_to = strdup(c->comm);
						}
					}
					c = c->next;
				}
				c = p->all;
				while (c) {
					if (p->pid == we->waker && !from) {
						from = c->Y;
						task_from = strdup(c->comm);
					}
					if (p->pid == we->wakee && !to) {
						to = c->Y;
						task_to = strdup(c->comm);
					}
					c = c->next;
				}
			}
			p = p->next;
		}

		if (!task_from) {
			task_from = malloc(40);
			sprintf(task_from, "[%i]", we->waker);
		}
		if (!task_to) {
			task_to = malloc(40);
			sprintf(task_to, "[%i]", we->wakee);
		}

		if (we->waker == -1)
			svg_interrupt(we->time, to);
		else if (from && to && abs(from - to) == 1)
			svg_wakeline(we->time, from, to);
		else
			svg_partial_wakeline(we->time, from, task_from, to, task_to);
		we = we->next;

		free(task_from);
		free(task_to);
	}
}

static void draw_cpu_usage(void)
{
	struct per_pid *p;
	struct per_pidcomm *c;
	struct cpu_sample *sample;
	p = all_data;
	while (p) {
		c = p->all;
		while (c) {
			sample = c->samples;
			while (sample) {
				if (sample->type == TYPE_RUNNING)
					svg_process(sample->cpu, sample->start_time, sample->end_time, "sample", c->comm);

				sample = sample->next;
			}
			c = c->next;
		}
		p = p->next;
	}
}

static void draw_process_bars(void)
{
	struct per_pid *p;
	struct per_pidcomm *c;
	struct cpu_sample *sample;
	int Y = 0;

	Y = 2 * numcpus + 2;

	p = all_data;
	while (p) {
		c = p->all;
		while (c) {
			if (!c->display) {
				c->Y = 0;
				c = c->next;
				continue;
			}

			svg_box(Y, c->start_time, c->end_time, "process");
			sample = c->samples;
			while (sample) {
				if (sample->type == TYPE_RUNNING)
					svg_sample(Y, sample->cpu, sample->start_time, sample->end_time);
				if (sample->type == TYPE_BLOCKED)
					svg_box(Y, sample->start_time, sample->end_time, "blocked");
				if (sample->type == TYPE_WAITING)
					svg_waiting(Y, sample->start_time, sample->end_time);
				sample = sample->next;
			}

			if (c->comm) {
				char comm[256];
				if (c->total_time > 5000000000) /* 5 seconds */
					sprintf(comm, "%s:%i (%2.2fs)", c->comm, p->pid, c->total_time / 1000000000.0);
				else
					sprintf(comm, "%s:%i (%3.1fms)", c->comm, p->pid, c->total_time / 1000000.0);

				svg_text(Y, c->start_time, comm);
			}
			c->Y = Y;
			Y++;
			c = c->next;
		}
		p = p->next;
	}
}

static void add_process_filter(const char *string)
{
	struct process_filter *filt;
	int pid;

	pid = strtoull(string, NULL, 10);
	filt = malloc(sizeof(struct process_filter));
	if (!filt)
		return;

	filt->name = strdup(string);
	filt->pid  = pid;
	filt->next = process_filter;

	process_filter = filt;
}

static int passes_filter(struct per_pid *p, struct per_pidcomm *c)
{
	struct process_filter *filt;
	if (!process_filter)
		return 1;

	filt = process_filter;
	while (filt) {
		if (filt->pid && p->pid == filt->pid)
			return 1;
		if (strcmp(filt->name, c->comm) == 0)
			return 1;
		filt = filt->next;
	}
	return 0;
}

static int determine_display_tasks_filtered(void)
{
	struct per_pid *p;
	struct per_pidcomm *c;
	int count = 0;

	p = all_data;
	while (p) {
		p->display = 0;
		if (p->start_time == 1)
			p->start_time = first_time;

		/* no exit marker, task kept running to the end */
		if (p->end_time == 0)
			p->end_time = last_time;

		c = p->all;

		while (c) {
			c->display = 0;

			if (c->start_time == 1)
				c->start_time = first_time;

			if (passes_filter(p, c)) {
				c->display = 1;
				p->display = 1;
				count++;
			}

			if (c->end_time == 0)
				c->end_time = last_time;

			c = c->next;
		}
		p = p->next;
	}
	return count;
}

static int determine_display_tasks(u64 threshold)
{
	struct per_pid *p;
	struct per_pidcomm *c;
	int count = 0;

	if (process_filter)
		return determine_display_tasks_filtered();

	p = all_data;
	while (p) {
		p->display = 0;
		if (p->start_time == 1)
			p->start_time = first_time;

		/* no exit marker, task kept running to the end */
		if (p->end_time == 0)
			p->end_time = last_time;
		if (p->total_time >= threshold && !power_only)
			p->display = 1;

		c = p->all;

		while (c) {
			c->display = 0;

			if (c->start_time == 1)
				c->start_time = first_time;

			if (c->total_time >= threshold && !power_only) {
				c->display = 1;
				count++;
			}

			if (c->end_time == 0)
				c->end_time = last_time;

			c = c->next;
		}
		p = p->next;
	}
	return count;
}



#define TIME_THRESH 10000000

static void write_svg_file(const char *filename)
{
	u64 i;
	int count;

	numcpus++;


	count = determine_display_tasks(TIME_THRESH);

	/* We'd like to show at least 15 tasks; be less picky if we have fewer */
	if (count < 15)
		count = determine_display_tasks(TIME_THRESH / 10);

	open_svg(filename, numcpus, count, first_time, last_time);

	svg_time_grid();
	svg_legenda();

	for (i = 0; i < numcpus; i++)
		svg_cpu_box(i, max_freq, turbo_frequency);

	draw_cpu_usage();
	draw_process_bars();
	draw_c_p_states();
	draw_wakeups();

	svg_close();
}

static int
process_event(event_t *event)
{

	switch (event->header.type) {

	case PERF_RECORD_COMM:
		return process_comm_event(event);
	case PERF_RECORD_FORK:
		return process_fork_event(event);
	case PERF_RECORD_EXIT:
		return process_exit_event(event);
	case PERF_RECORD_SAMPLE:
		return queue_sample_event(event);

	/*
	 * We dont process them right now but they are fine:
	 */
	case PERF_RECORD_MMAP:
	case PERF_RECORD_THROTTLE:
	case PERF_RECORD_UNTHROTTLE:
		return 0;

	default:
		return -1;
	}

	return 0;
}

static void process_samples(void)
{
	struct sample_wrapper *cursor;
	event_t *event;

	sort_queued_samples();

	cursor = all_samples;
	while (cursor) {
		event = (void *)&cursor->data;
		cursor = cursor->next;
		process_sample_event(event);
	}
}


static int __cmd_timechart(void)
{
	int err, rc = EXIT_FAILURE;
	unsigned long offset = 0;
	unsigned long head, shift;
	struct stat statbuf;
	event_t *event;
	uint32_t size;
	char *buf;
	int input;

	input = open(input_name, O_RDONLY);
	if (input < 0) {
		fprintf(stderr, " failed to open file: %s", input_name);
		if (!strcmp(input_name, "perf.data"))
			fprintf(stderr, "  (try 'perf record' first)");
		fprintf(stderr, "\n");
		exit(-1);
	}

	err = fstat(input, &statbuf);
	if (err < 0) {
		perror("failed to stat file");
		exit(-1);
	}

	if (!statbuf.st_size) {
		fprintf(stderr, "zero-sized file, nothing to do!\n");
		exit(0);
	}

	header = perf_header__new();
	if (header == NULL)
		return -ENOMEM;

	err = perf_header__read(header, input);
	if (err < 0) {
		perf_header__delete(header);
		return err;
	}

	head = header->data_offset;

	sample_type = perf_header__sample_type(header);

	shift = page_size * (head / page_size);
	offset += shift;
	head -= shift;

remap:
	buf = (char *)mmap(NULL, page_size * mmap_window, PROT_READ,
			   MAP_SHARED, input, offset);
	if (buf == MAP_FAILED) {
		perror("failed to mmap file");
		exit(-1);
	}

more:
	event = (event_t *)(buf + head);

	size = event->header.size;
	if (!size)
		size = 8;

	if (head + event->header.size >= page_size * mmap_window) {
		int ret2;

		shift = page_size * (head / page_size);

		ret2 = munmap(buf, page_size * mmap_window);
		assert(ret2 == 0);

		offset += shift;
		head -= shift;
		goto remap;
	}

	size = event->header.size;

	if (!size || process_event(event) < 0) {
		pr_warning("%p [%p]: skipping unknown header type: %d\n",
			   (void *)(offset + head),
			   (void *)(long)(event->header.size),
			   event->header.type);
		/*
		 * assume we lost track of the stream, check alignment, and
		 * increment a single u64 in the hope to catch on again 'soon'.
		 */

		if (unlikely(head & 7))
			head &= ~7ULL;

		size = 8;
	}

	head += size;

	if (offset + head >= header->data_offset + header->data_size)
		goto done;

	if (offset + head < (unsigned long)statbuf.st_size)
		goto more;

done:
	rc = EXIT_SUCCESS;
	close(input);


	process_samples();

	end_sample_processing();

	sort_pids();

	write_svg_file(output_name);

	pr_info("Written %2.1f seconds of trace to %s.\n",
		(last_time - first_time) / 1000000000.0, output_name);

	return rc;
}

static const char * const timechart_usage[] = {
	"perf timechart [<options>] {record}",
	NULL
};

static const char *record_args[] = {
	"record",
	"-a",
	"-R",
	"-M",
	"-f",
	"-c", "1",
	"-e", "power:power_start",
	"-e", "power:power_end",
	"-e", "power:power_frequency",
	"-e", "sched:sched_wakeup",
	"-e", "sched:sched_switch",
};

static int __cmd_record(int argc, const char **argv)
{
	unsigned int rec_argc, i, j;
	const char **rec_argv;

	rec_argc = ARRAY_SIZE(record_args) + argc - 1;
	rec_argv = calloc(rec_argc + 1, sizeof(char *));

	for (i = 0; i < ARRAY_SIZE(record_args); i++)
		rec_argv[i] = strdup(record_args[i]);

	for (j = 1; j < (unsigned int)argc; j++, i++)
		rec_argv[i] = argv[j];

	return cmd_record(i, rec_argv, NULL);
}

static int
parse_process(const struct option *opt __used, const char *arg, int __used unset)
{
	if (arg)
		add_process_filter(arg);
	return 0;
}

static const struct option options[] = {
	OPT_STRING('i', "input", &input_name, "file",
		    "input file name"),
	OPT_STRING('o', "output", &output_name, "file",
		    "output file name"),
	OPT_INTEGER('w', "width", &svg_page_width,
		    "page width"),
	OPT_BOOLEAN('P', "power-only", &power_only,
		    "output power data only"),
	OPT_CALLBACK('p', "process", NULL, "process",
		      "process selector. Pass a pid or process name.",
		       parse_process),
	OPT_END()
};


int cmd_timechart(int argc, const char **argv, const char *prefix __used)
{
	symbol__init(0);

	page_size = getpagesize();

	argc = parse_options(argc, argv, options, timechart_usage,
			PARSE_OPT_STOP_AT_NON_OPTION);

	if (argc && !strncmp(argv[0], "rec", 3))
		return __cmd_record(argc, argv);
	else if (argc)
		usage_with_options(timechart_usage, options);

	setup_pager();

	return __cmd_timechart();
}