summaryrefslogtreecommitdiffstats
path: root/arch/x86/events/intel/rapl.c
blob: b834a3f55a015b0cd2a45972a0e2bdde24c24dfa (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
/*
 * perf_event_intel_rapl.c: support Intel RAPL energy consumption counters
 * Copyright (C) 2013 Google, Inc., Stephane Eranian
 *
 * Intel RAPL interface is specified in the IA-32 Manual Vol3b
 * section 14.7.1 (September 2013)
 *
 * RAPL provides more controls than just reporting energy consumption
 * however here we only expose the 3 energy consumption free running
 * counters (pp0, pkg, dram).
 *
 * Each of those counters increments in a power unit defined by the
 * RAPL_POWER_UNIT MSR. On SandyBridge, this unit is 1/(2^16) Joules
 * but it can vary.
 *
 * Counter to rapl events mappings:
 *
 *  pp0 counter: consumption of all physical cores (power plane 0)
 * 	  event: rapl_energy_cores
 *    perf code: 0x1
 *
 *  pkg counter: consumption of the whole processor package
 *	  event: rapl_energy_pkg
 *    perf code: 0x2
 *
 * dram counter: consumption of the dram domain (servers only)
 *	  event: rapl_energy_dram
 *    perf code: 0x3
 *
 * dram counter: consumption of the builtin-gpu domain (client only)
 *	  event: rapl_energy_gpu
 *    perf code: 0x4
 *
 * We manage those counters as free running (read-only). They may be
 * use simultaneously by other tools, such as turbostat.
 *
 * The events only support system-wide mode counting. There is no
 * sampling support because it does not make sense and is not
 * supported by the RAPL hardware.
 *
 * Because we want to avoid floating-point operations in the kernel,
 * the events are all reported in fixed point arithmetic (32.32).
 * Tools must adjust the counts to convert them to Watts using
 * the duration of the measurement. Tools may use a function such as
 * ldexp(raw_count, -32);
 */

#define pr_fmt(fmt) "RAPL PMU: " fmt

#include <linux/module.h>
#include <linux/slab.h>
#include <linux/perf_event.h>
#include <asm/cpu_device_id.h>
#include "../perf_event.h"

/*
 * RAPL energy status counters
 */
#define RAPL_IDX_PP0_NRG_STAT	0	/* all cores */
#define INTEL_RAPL_PP0		0x1	/* pseudo-encoding */
#define RAPL_IDX_PKG_NRG_STAT	1	/* entire package */
#define INTEL_RAPL_PKG		0x2	/* pseudo-encoding */
#define RAPL_IDX_RAM_NRG_STAT	2	/* DRAM */
#define INTEL_RAPL_RAM		0x3	/* pseudo-encoding */
#define RAPL_IDX_PP1_NRG_STAT	3	/* gpu */
#define INTEL_RAPL_PP1		0x4	/* pseudo-encoding */

#define NR_RAPL_DOMAINS         0x4
static const char *const rapl_domain_names[NR_RAPL_DOMAINS] __initconst = {
	"pp0-core",
	"package",
	"dram",
	"pp1-gpu",
};

/* Clients have PP0, PKG */
#define RAPL_IDX_CLN	(1<<RAPL_IDX_PP0_NRG_STAT|\
			 1<<RAPL_IDX_PKG_NRG_STAT|\
			 1<<RAPL_IDX_PP1_NRG_STAT)

/* Servers have PP0, PKG, RAM */
#define RAPL_IDX_SRV	(1<<RAPL_IDX_PP0_NRG_STAT|\
			 1<<RAPL_IDX_PKG_NRG_STAT|\
			 1<<RAPL_IDX_RAM_NRG_STAT)

/* Servers have PP0, PKG, RAM, PP1 */
#define RAPL_IDX_HSW	(1<<RAPL_IDX_PP0_NRG_STAT|\
			 1<<RAPL_IDX_PKG_NRG_STAT|\
			 1<<RAPL_IDX_RAM_NRG_STAT|\
			 1<<RAPL_IDX_PP1_NRG_STAT)

/* Knights Landing has PKG, RAM */
#define RAPL_IDX_KNL	(1<<RAPL_IDX_PKG_NRG_STAT|\
			 1<<RAPL_IDX_RAM_NRG_STAT)

/*
 * event code: LSB 8 bits, passed in attr->config
 * any other bit is reserved
 */
#define RAPL_EVENT_MASK	0xFFULL

#define DEFINE_RAPL_FORMAT_ATTR(_var, _name, _format)		\
static ssize_t __rapl_##_var##_show(struct kobject *kobj,	\
				struct kobj_attribute *attr,	\
				char *page)			\
{								\
	BUILD_BUG_ON(sizeof(_format) >= PAGE_SIZE);		\
	return sprintf(page, _format "\n");			\
}								\
static struct kobj_attribute format_attr_##_var =		\
	__ATTR(_name, 0444, __rapl_##_var##_show, NULL)

#define RAPL_CNTR_WIDTH 32

#define RAPL_EVENT_ATTR_STR(_name, v, str)					\
static struct perf_pmu_events_attr event_attr_##v = {				\
	.attr		= __ATTR(_name, 0444, perf_event_sysfs_show, NULL),	\
	.id		= 0,							\
	.event_str	= str,							\
};

struct rapl_pmu {
	raw_spinlock_t		lock;
	int			n_active;
	int			cpu;
	struct list_head	active_list;
	struct pmu		*pmu;
	ktime_t			timer_interval;
	struct hrtimer		hrtimer;
};

struct rapl_pmus {
	struct pmu		pmu;
	unsigned int		maxpkg;
	struct rapl_pmu		*pmus[];
};

 /* 1/2^hw_unit Joule */
static int rapl_hw_unit[NR_RAPL_DOMAINS] __read_mostly;
static struct rapl_pmus *rapl_pmus;
static cpumask_t rapl_cpu_mask;
static unsigned int rapl_cntr_mask;
static u64 rapl_timer_ms;

static inline struct rapl_pmu *cpu_to_rapl_pmu(unsigned int cpu)
{
	return rapl_pmus->pmus[topology_logical_package_id(cpu)];
}

static inline u64 rapl_read_counter(struct perf_event *event)
{
	u64 raw;
	rdmsrl(event->hw.event_base, raw);
	return raw;
}

static inline u64 rapl_scale(u64 v, int cfg)
{
	if (cfg > NR_RAPL_DOMAINS) {
		pr_warn("Invalid domain %d, failed to scale data\n", cfg);
		return v;
	}
	/*
	 * scale delta to smallest unit (1/2^32)
	 * users must then scale back: count * 1/(1e9*2^32) to get Joules
	 * or use ldexp(count, -32).
	 * Watts = Joules/Time delta
	 */
	return v << (32 - rapl_hw_unit[cfg - 1]);
}

static u64 rapl_event_update(struct perf_event *event)
{
	struct hw_perf_event *hwc = &event->hw;
	u64 prev_raw_count, new_raw_count;
	s64 delta, sdelta;
	int shift = RAPL_CNTR_WIDTH;

again:
	prev_raw_count = local64_read(&hwc->prev_count);
	rdmsrl(event->hw.event_base, new_raw_count);

	if (local64_cmpxchg(&hwc->prev_count, prev_raw_count,
			    new_raw_count) != prev_raw_count) {
		cpu_relax();
		goto again;
	}

	/*
	 * Now we have the new raw value and have updated the prev
	 * timestamp already. We can now calculate the elapsed delta
	 * (event-)time and add that to the generic event.
	 *
	 * Careful, not all hw sign-extends above the physical width
	 * of the count.
	 */
	delta = (new_raw_count << shift) - (prev_raw_count << shift);
	delta >>= shift;

	sdelta = rapl_scale(delta, event->hw.config);

	local64_add(sdelta, &event->count);

	return new_raw_count;
}

static void rapl_start_hrtimer(struct rapl_pmu *pmu)
{
       hrtimer_start(&pmu->hrtimer, pmu->timer_interval,
		     HRTIMER_MODE_REL_PINNED);
}

static enum hrtimer_restart rapl_hrtimer_handle(struct hrtimer *hrtimer)
{
	struct rapl_pmu *pmu = container_of(hrtimer, struct rapl_pmu, hrtimer);
	struct perf_event *event;
	unsigned long flags;

	if (!pmu->n_active)
		return HRTIMER_NORESTART;

	raw_spin_lock_irqsave(&pmu->lock, flags);

	list_for_each_entry(event, &pmu->active_list, active_entry)
		rapl_event_update(event);

	raw_spin_unlock_irqrestore(&pmu->lock, flags);

	hrtimer_forward_now(hrtimer, pmu->timer_interval);

	return HRTIMER_RESTART;
}

static void rapl_hrtimer_init(struct rapl_pmu *pmu)
{
	struct hrtimer *hr = &pmu->hrtimer;

	hrtimer_init(hr, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
	hr->function = rapl_hrtimer_handle;
}

static void __rapl_pmu_event_start(struct rapl_pmu *pmu,
				   struct perf_event *event)
{
	if (WARN_ON_ONCE(!(event->hw.state & PERF_HES_STOPPED)))
		return;

	event->hw.state = 0;

	list_add_tail(&event->active_entry, &pmu->active_list);

	local64_set(&event->hw.prev_count, rapl_read_counter(event));

	pmu->n_active++;
	if (pmu->n_active == 1)
		rapl_start_hrtimer(pmu);
}

static void rapl_pmu_event_start(struct perf_event *event, int mode)
{
	struct rapl_pmu *pmu = event->pmu_private;
	unsigned long flags;

	raw_spin_lock_irqsave(&pmu->lock, flags);
	__rapl_pmu_event_start(pmu, event);
	raw_spin_unlock_irqrestore(&pmu->lock, flags);
}

static void rapl_pmu_event_stop(struct perf_event *event, int mode)
{
	struct rapl_pmu *pmu = event->pmu_private;
	struct hw_perf_event *hwc = &event->hw;
	unsigned long flags;

	raw_spin_lock_irqsave(&pmu->lock, flags);

	/* mark event as deactivated and stopped */
	if (!(hwc->state & PERF_HES_STOPPED)) {
		WARN_ON_ONCE(pmu->n_active <= 0);
		pmu->n_active--;
		if (pmu->n_active == 0)
			hrtimer_cancel(&pmu->hrtimer);

		list_del(&event->active_entry);

		WARN_ON_ONCE(hwc->state & PERF_HES_STOPPED);
		hwc->state |= PERF_HES_STOPPED;
	}

	/* check if update of sw counter is necessary */
	if ((mode & PERF_EF_UPDATE) && !(hwc->state & PERF_HES_UPTODATE)) {
		/*
		 * Drain the remaining delta count out of a event
		 * that we are disabling:
		 */
		rapl_event_update(event);
		hwc->state |= PERF_HES_UPTODATE;
	}

	raw_spin_unlock_irqrestore(&pmu->lock, flags);
}

static int rapl_pmu_event_add(struct perf_event *event, int mode)
{
	struct rapl_pmu *pmu = event->pmu_private;
	struct hw_perf_event *hwc = &event->hw;
	unsigned long flags;

	raw_spin_lock_irqsave(&pmu->lock, flags);

	hwc->state = PERF_HES_UPTODATE | PERF_HES_STOPPED;

	if (mode & PERF_EF_START)
		__rapl_pmu_event_start(pmu, event);

	raw_spin_unlock_irqrestore(&pmu->lock, flags);

	return 0;
}

static void rapl_pmu_event_del(struct perf_event *event, int flags)
{
	rapl_pmu_event_stop(event, PERF_EF_UPDATE);
}

static int rapl_pmu_event_init(struct perf_event *event)
{
	u64 cfg = event->attr.config & RAPL_EVENT_MASK;
	int bit, msr, ret = 0;
	struct rapl_pmu *pmu;

	/* only look at RAPL events */
	if (event->attr.type != rapl_pmus->pmu.type)
		return -ENOENT;

	/* check only supported bits are set */
	if (event->attr.config & ~RAPL_EVENT_MASK)
		return -EINVAL;

	if (event->cpu < 0)
		return -EINVAL;

	/*
	 * check event is known (determines counter)
	 */
	switch (cfg) {
	case INTEL_RAPL_PP0:
		bit = RAPL_IDX_PP0_NRG_STAT;
		msr = MSR_PP0_ENERGY_STATUS;
		break;
	case INTEL_RAPL_PKG:
		bit = RAPL_IDX_PKG_NRG_STAT;
		msr = MSR_PKG_ENERGY_STATUS;
		break;
	case INTEL_RAPL_RAM:
		bit = RAPL_IDX_RAM_NRG_STAT;
		msr = MSR_DRAM_ENERGY_STATUS;
		break;
	case INTEL_RAPL_PP1:
		bit = RAPL_IDX_PP1_NRG_STAT;
		msr = MSR_PP1_ENERGY_STATUS;
		break;
	default:
		return -EINVAL;
	}
	/* check event supported */
	if (!(rapl_cntr_mask & (1 << bit)))
		return -EINVAL;

	/* unsupported modes and filters */
	if (event->attr.exclude_user   ||
	    event->attr.exclude_kernel ||
	    event->attr.exclude_hv     ||
	    event->attr.exclude_idle   ||
	    event->attr.exclude_host   ||
	    event->attr.exclude_guest  ||
	    event->attr.sample_period) /* no sampling */
		return -EINVAL;

	/* must be done before validate_group */
	pmu = cpu_to_rapl_pmu(event->cpu);
	event->cpu = pmu->cpu;
	event->pmu_private = pmu;
	event->hw.event_base = msr;
	event->hw.config = cfg;
	event->hw.idx = bit;

	return ret;
}

static void rapl_pmu_event_read(struct perf_event *event)
{
	rapl_event_update(event);
}

static ssize_t rapl_get_attr_cpumask(struct device *dev,
				struct device_attribute *attr, char *buf)
{
	return cpumap_print_to_pagebuf(true, buf, &rapl_cpu_mask);
}

static DEVICE_ATTR(cpumask, S_IRUGO, rapl_get_attr_cpumask, NULL);

static struct attribute *rapl_pmu_attrs[] = {
	&dev_attr_cpumask.attr,
	NULL,
};

static struct attribute_group rapl_pmu_attr_group = {
	.attrs = rapl_pmu_attrs,
};

RAPL_EVENT_ATTR_STR(energy-cores, rapl_cores, "event=0x01");
RAPL_EVENT_ATTR_STR(energy-pkg  ,   rapl_pkg, "event=0x02");
RAPL_EVENT_ATTR_STR(energy-ram  ,   rapl_ram, "event=0x03");
RAPL_EVENT_ATTR_STR(energy-gpu  ,   rapl_gpu, "event=0x04");

RAPL_EVENT_ATTR_STR(energy-cores.unit, rapl_cores_unit, "Joules");
RAPL_EVENT_ATTR_STR(energy-pkg.unit  ,   rapl_pkg_unit, "Joules");
RAPL_EVENT_ATTR_STR(energy-ram.unit  ,   rapl_ram_unit, "Joules");
RAPL_EVENT_ATTR_STR(energy-gpu.unit  ,   rapl_gpu_unit, "Joules");

/*
 * we compute in 0.23 nJ increments regardless of MSR
 */
RAPL_EVENT_ATTR_STR(energy-cores.scale, rapl_cores_scale, "2.3283064365386962890625e-10");
RAPL_EVENT_ATTR_STR(energy-pkg.scale,     rapl_pkg_scale, "2.3283064365386962890625e-10");
RAPL_EVENT_ATTR_STR(energy-ram.scale,     rapl_ram_scale, "2.3283064365386962890625e-10");
RAPL_EVENT_ATTR_STR(energy-gpu.scale,     rapl_gpu_scale, "2.3283064365386962890625e-10");

static struct attribute *rapl_events_srv_attr[] = {
	EVENT_PTR(rapl_cores),
	EVENT_PTR(rapl_pkg),
	EVENT_PTR(rapl_ram),

	EVENT_PTR(rapl_cores_unit),
	EVENT_PTR(rapl_pkg_unit),
	EVENT_PTR(rapl_ram_unit),

	EVENT_PTR(rapl_cores_scale),
	EVENT_PTR(rapl_pkg_scale),
	EVENT_PTR(rapl_ram_scale),
	NULL,
};

static struct attribute *rapl_events_cln_attr[] = {
	EVENT_PTR(rapl_cores),
	EVENT_PTR(rapl_pkg),
	EVENT_PTR(rapl_gpu),

	EVENT_PTR(rapl_cores_unit),
	EVENT_PTR(rapl_pkg_unit),
	EVENT_PTR(rapl_gpu_unit),

	EVENT_PTR(rapl_cores_scale),
	EVENT_PTR(rapl_pkg_scale),
	EVENT_PTR(rapl_gpu_scale),
	NULL,
};

static struct attribute *rapl_events_hsw_attr[] = {
	EVENT_PTR(rapl_cores),
	EVENT_PTR(rapl_pkg),
	EVENT_PTR(rapl_gpu),
	EVENT_PTR(rapl_ram),

	EVENT_PTR(rapl_cores_unit),
	EVENT_PTR(rapl_pkg_unit),
	EVENT_PTR(rapl_gpu_unit),
	EVENT_PTR(rapl_ram_unit),

	EVENT_PTR(rapl_cores_scale),
	EVENT_PTR(rapl_pkg_scale),
	EVENT_PTR(rapl_gpu_scale),
	EVENT_PTR(rapl_ram_scale),
	NULL,
};

static struct attribute *rapl_events_knl_attr[] = {
	EVENT_PTR(rapl_pkg),
	EVENT_PTR(rapl_ram),

	EVENT_PTR(rapl_pkg_unit),
	EVENT_PTR(rapl_ram_unit),

	EVENT_PTR(rapl_pkg_scale),
	EVENT_PTR(rapl_ram_scale),
	NULL,
};

static struct attribute_group rapl_pmu_events_group = {
	.name = "events",
	.attrs = NULL, /* patched at runtime */
};

DEFINE_RAPL_FORMAT_ATTR(event, event, "config:0-7");
static struct attribute *rapl_formats_attr[] = {
	&format_attr_event.attr,
	NULL,
};

static struct attribute_group rapl_pmu_format_group = {
	.name = "format",
	.attrs = rapl_formats_attr,
};

const struct attribute_group *rapl_attr_groups[] = {
	&rapl_pmu_attr_group,
	&rapl_pmu_format_group,
	&rapl_pmu_events_group,
	NULL,
};

static void rapl_cpu_exit(int cpu)
{
	struct rapl_pmu *pmu = cpu_to_rapl_pmu(cpu);
	int target;

	/* Check if exiting cpu is used for collecting rapl events */
	if (!cpumask_test_and_clear_cpu(cpu, &rapl_cpu_mask))
		return;

	pmu->cpu = -1;
	/* Find a new cpu to collect rapl events */
	target = cpumask_any_but(topology_core_cpumask(cpu), cpu);

	/* Migrate rapl events to the new target */
	if (target < nr_cpu_ids) {
		cpumask_set_cpu(target, &rapl_cpu_mask);
		pmu->cpu = target;
		perf_pmu_migrate_context(pmu->pmu, cpu, target);
	}
}

static void rapl_cpu_init(int cpu)
{
	struct rapl_pmu *pmu = cpu_to_rapl_pmu(cpu);
	int target;

	/*
	 * Check if there is an online cpu in the package which collects rapl
	 * events already.
	 */
	target = cpumask_any_and(&rapl_cpu_mask, topology_core_cpumask(cpu));
	if (target < nr_cpu_ids)
		return;

	cpumask_set_cpu(cpu, &rapl_cpu_mask);
	pmu->cpu = cpu;
}

static int rapl_cpu_prepare(int cpu)
{
	struct rapl_pmu *pmu = cpu_to_rapl_pmu(cpu);

	if (pmu)
		return 0;

	pmu = kzalloc_node(sizeof(*pmu), GFP_KERNEL, cpu_to_node(cpu));
	if (!pmu)
		return -ENOMEM;

	raw_spin_lock_init(&pmu->lock);
	INIT_LIST_HEAD(&pmu->active_list);
	pmu->pmu = &rapl_pmus->pmu;
	pmu->timer_interval = ms_to_ktime(rapl_timer_ms);
	pmu->cpu = -1;
	rapl_hrtimer_init(pmu);
	rapl_pmus->pmus[topology_logical_package_id(cpu)] = pmu;
	return 0;
}

static int rapl_cpu_notifier(struct notifier_block *self,
			     unsigned long action, void *hcpu)
{
	unsigned int cpu = (long)hcpu;

	switch (action & ~CPU_TASKS_FROZEN) {
	case CPU_UP_PREPARE:
		rapl_cpu_prepare(cpu);
		break;

	case CPU_DOWN_FAILED:
	case CPU_ONLINE:
		rapl_cpu_init(cpu);
		break;

	case CPU_DOWN_PREPARE:
		rapl_cpu_exit(cpu);
		break;
	}
	return NOTIFY_OK;
}

static int rapl_check_hw_unit(bool apply_quirk)
{
	u64 msr_rapl_power_unit_bits;
	int i;

	/* protect rdmsrl() to handle virtualization */
	if (rdmsrl_safe(MSR_RAPL_POWER_UNIT, &msr_rapl_power_unit_bits))
		return -1;
	for (i = 0; i < NR_RAPL_DOMAINS; i++)
		rapl_hw_unit[i] = (msr_rapl_power_unit_bits >> 8) & 0x1FULL;

	/*
	 * DRAM domain on HSW server and KNL has fixed energy unit which can be
	 * different than the unit from power unit MSR. See
	 * "Intel Xeon Processor E5-1600 and E5-2600 v3 Product Families, V2
	 * of 2. Datasheet, September 2014, Reference Number: 330784-001 "
	 */
	if (apply_quirk)
		rapl_hw_unit[RAPL_IDX_RAM_NRG_STAT] = 16;

	/*
	 * Calculate the timer rate:
	 * Use reference of 200W for scaling the timeout to avoid counter
	 * overflows. 200W = 200 Joules/sec
	 * Divide interval by 2 to avoid lockstep (2 * 100)
	 * if hw unit is 32, then we use 2 ms 1/200/2
	 */
	rapl_timer_ms = 2;
	if (rapl_hw_unit[0] < 32) {
		rapl_timer_ms = (1000 / (2 * 100));
		rapl_timer_ms *= (1ULL << (32 - rapl_hw_unit[0] - 1));
	}
	return 0;
}

static void __init rapl_advertise(void)
{
	int i;

	pr_info("API unit is 2^-32 Joules, %d fixed counters, %llu ms ovfl timer\n",
		hweight32(rapl_cntr_mask), rapl_timer_ms);

	for (i = 0; i < NR_RAPL_DOMAINS; i++) {
		if (rapl_cntr_mask & (1 << i)) {
			pr_info("hw unit of domain %s 2^-%d Joules\n",
				rapl_domain_names[i], rapl_hw_unit[i]);
		}
	}
}

static int __init rapl_prepare_cpus(void)
{
	unsigned int cpu, pkg;
	int ret;

	for_each_online_cpu(cpu) {
		pkg = topology_logical_package_id(cpu);
		if (rapl_pmus->pmus[pkg])
			continue;

		ret = rapl_cpu_prepare(cpu);
		if (ret)
			return ret;
		rapl_cpu_init(cpu);
	}
	return 0;
}

static void __init cleanup_rapl_pmus(void)
{
	int i;

	for (i = 0; i < rapl_pmus->maxpkg; i++)
		kfree(rapl_pmus->pmus + i);
	kfree(rapl_pmus);
}

static int __init init_rapl_pmus(void)
{
	int maxpkg = topology_max_packages();
	size_t size;

	size = sizeof(*rapl_pmus) + maxpkg * sizeof(struct rapl_pmu *);
	rapl_pmus = kzalloc(size, GFP_KERNEL);
	if (!rapl_pmus)
		return -ENOMEM;

	rapl_pmus->maxpkg		= maxpkg;
	rapl_pmus->pmu.attr_groups	= rapl_attr_groups;
	rapl_pmus->pmu.task_ctx_nr	= perf_invalid_context;
	rapl_pmus->pmu.event_init	= rapl_pmu_event_init;
	rapl_pmus->pmu.add		= rapl_pmu_event_add;
	rapl_pmus->pmu.del		= rapl_pmu_event_del;
	rapl_pmus->pmu.start		= rapl_pmu_event_start;
	rapl_pmus->pmu.stop		= rapl_pmu_event_stop;
	rapl_pmus->pmu.read		= rapl_pmu_event_read;
	return 0;
}

static const struct x86_cpu_id rapl_cpu_match[] __initconst = {
	[0] = { .vendor = X86_VENDOR_INTEL, .family = 6 },
	[1] = {},
};

static int __init rapl_pmu_init(void)
{
	bool apply_quirk = false;
	int ret;

	if (!x86_match_cpu(rapl_cpu_match))
		return -ENODEV;

	switch (boot_cpu_data.x86_model) {
	case 42: /* Sandy Bridge */
	case 58: /* Ivy Bridge */
		rapl_cntr_mask = RAPL_IDX_CLN;
		rapl_pmu_events_group.attrs = rapl_events_cln_attr;
		break;
	case 63: /* Haswell-Server */
		apply_quirk = true;
		rapl_cntr_mask = RAPL_IDX_SRV;
		rapl_pmu_events_group.attrs = rapl_events_srv_attr;
		break;
	case 60: /* Haswell */
	case 69: /* Haswell-Celeron */
	case 61: /* Broadwell */
		rapl_cntr_mask = RAPL_IDX_HSW;
		rapl_pmu_events_group.attrs = rapl_events_hsw_attr;
		break;
	case 45: /* Sandy Bridge-EP */
	case 62: /* IvyTown */
		rapl_cntr_mask = RAPL_IDX_SRV;
		rapl_pmu_events_group.attrs = rapl_events_srv_attr;
		break;
	case 87: /* Knights Landing */
		apply_quirk = true;
		rapl_cntr_mask = RAPL_IDX_KNL;
		rapl_pmu_events_group.attrs = rapl_events_knl_attr;
		break;
	default:
		return -ENODEV;
	}

	ret = rapl_check_hw_unit(apply_quirk);
	if (ret)
		return ret;

	ret = init_rapl_pmus();
	if (ret)
		return ret;

	cpu_notifier_register_begin();

	ret = rapl_prepare_cpus();
	if (ret)
		goto out;

	ret = perf_pmu_register(&rapl_pmus->pmu, "power", -1);
	if (ret)
		goto out;

	__perf_cpu_notifier(rapl_cpu_notifier);
	cpu_notifier_register_done();
	rapl_advertise();
	return 0;

out:
	pr_warn("Initialization failed (%d), disabled\n", ret);
	cleanup_rapl_pmus();
	cpu_notifier_register_done();
	return ret;
}
device_initcall(rapl_pmu_init);