summaryrefslogtreecommitdiff
path: root/arch/x86/events/amd/ibs.c
blob: 40669eac9d6db51f09227cd5e8283e79be42bfd0 (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
/*
 * Performance events - AMD IBS
 *
 *  Copyright (C) 2011 Advanced Micro Devices, Inc., Robert Richter
 *
 *  For licencing details see kernel-base/COPYING
 */

#include <linux/perf_event.h>
#include <linux/init.h>
#include <linux/export.h>
#include <linux/pci.h>
#include <linux/ptrace.h>
#include <linux/syscore_ops.h>
#include <linux/sched/clock.h>

#include <asm/apic.h>

#include "../perf_event.h"

static u32 ibs_caps;

#if defined(CONFIG_PERF_EVENTS) && defined(CONFIG_CPU_SUP_AMD)

#include <linux/kprobes.h>
#include <linux/hardirq.h>

#include <asm/nmi.h>

#define IBS_FETCH_CONFIG_MASK	(IBS_FETCH_RAND_EN | IBS_FETCH_MAX_CNT)
#define IBS_OP_CONFIG_MASK	IBS_OP_MAX_CNT


/*
 * IBS states:
 *
 * ENABLED; tracks the pmu::add(), pmu::del() state, when set the counter is taken
 * and any further add()s must fail.
 *
 * STARTED/STOPPING/STOPPED; deal with pmu::start(), pmu::stop() state but are
 * complicated by the fact that the IBS hardware can send late NMIs (ie. after
 * we've cleared the EN bit).
 *
 * In order to consume these late NMIs we have the STOPPED state, any NMI that
 * happens after we've cleared the EN state will clear this bit and report the
 * NMI handled (this is fundamentally racy in the face or multiple NMI sources,
 * someone else can consume our BIT and our NMI will go unhandled).
 *
 * And since we cannot set/clear this separate bit together with the EN bit,
 * there are races; if we cleared STARTED early, an NMI could land in
 * between clearing STARTED and clearing the EN bit (in fact multiple NMIs
 * could happen if the period is small enough), and consume our STOPPED bit
 * and trigger streams of unhandled NMIs.
 *
 * If, however, we clear STARTED late, an NMI can hit between clearing the
 * EN bit and clearing STARTED, still see STARTED set and process the event.
 * If this event will have the VALID bit clear, we bail properly, but this
 * is not a given. With VALID set we can end up calling pmu::stop() again
 * (the throttle logic) and trigger the WARNs in there.
 *
 * So what we do is set STOPPING before clearing EN to avoid the pmu::stop()
 * nesting, and clear STARTED late, so that we have a well defined state over
 * the clearing of the EN bit.
 *
 * XXX: we could probably be using !atomic bitops for all this.
 */

enum ibs_states {
	IBS_ENABLED	= 0,
	IBS_STARTED	= 1,
	IBS_STOPPING	= 2,
	IBS_STOPPED	= 3,

	IBS_MAX_STATES,
};

struct cpu_perf_ibs {
	struct perf_event	*event;
	unsigned long		state[BITS_TO_LONGS(IBS_MAX_STATES)];
};

struct perf_ibs {
	struct pmu			pmu;
	unsigned int			msr;
	u64				config_mask;
	u64				cnt_mask;
	u64				enable_mask;
	u64				valid_mask;
	u64				max_period;
	unsigned long			offset_mask[1];
	int				offset_max;
	unsigned int			fetch_count_reset_broken : 1;
	struct cpu_perf_ibs __percpu	*pcpu;

	struct attribute		**format_attrs;
	struct attribute_group		format_group;
	const struct attribute_group	*attr_groups[2];

	u64				(*get_count)(u64 config);
};

struct perf_ibs_data {
	u32		size;
	union {
		u32	data[0];	/* data buffer starts here */
		u32	caps;
	};
	u64		regs[MSR_AMD64_IBS_REG_COUNT_MAX];
};

static int
perf_event_set_period(struct hw_perf_event *hwc, u64 min, u64 max, u64 *hw_period)
{
	s64 left = local64_read(&hwc->period_left);
	s64 period = hwc->sample_period;
	int overflow = 0;

	/*
	 * If we are way outside a reasonable range then just skip forward:
	 */
	if (unlikely(left <= -period)) {
		left = period;
		local64_set(&hwc->period_left, left);
		hwc->last_period = period;
		overflow = 1;
	}

	if (unlikely(left < (s64)min)) {
		left += period;
		local64_set(&hwc->period_left, left);
		hwc->last_period = period;
		overflow = 1;
	}

	/*
	 * If the hw period that triggers the sw overflow is too short
	 * we might hit the irq handler. This biases the results.
	 * Thus we shorten the next-to-last period and set the last
	 * period to the max period.
	 */
	if (left > max) {
		left -= max;
		if (left > max)
			left = max;
		else if (left < min)
			left = min;
	}

	*hw_period = (u64)left;

	return overflow;
}

static  int
perf_event_try_update(struct perf_event *event, u64 new_raw_count, int width)
{
	struct hw_perf_event *hwc = &event->hw;
	int shift = 64 - width;
	u64 prev_raw_count;
	u64 delta;

	/*
	 * Careful: an NMI might modify the previous event value.
	 *
	 * Our tactic to handle this is to first atomically read and
	 * exchange a new raw count - then add that new-prev delta
	 * count to the generic event atomically:
	 */
	prev_raw_count = local64_read(&hwc->prev_count);
	if (local64_cmpxchg(&hwc->prev_count, prev_raw_count,
					new_raw_count) != prev_raw_count)
		return 0;

	/*
	 * 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;

	local64_add(delta, &event->count);
	local64_sub(delta, &hwc->period_left);

	return 1;
}

static struct perf_ibs perf_ibs_fetch;
static struct perf_ibs perf_ibs_op;

static struct perf_ibs *get_ibs_pmu(int type)
{
	if (perf_ibs_fetch.pmu.type == type)
		return &perf_ibs_fetch;
	if (perf_ibs_op.pmu.type == type)
		return &perf_ibs_op;
	return NULL;
}

/*
 * Use IBS for precise event sampling:
 *
 *  perf record -a -e cpu-cycles:p ...    # use ibs op counting cycle count
 *  perf record -a -e r076:p ...          # same as -e cpu-cycles:p
 *  perf record -a -e r0C1:p ...          # use ibs op counting micro-ops
 *
 * IbsOpCntCtl (bit 19) of IBS Execution Control Register (IbsOpCtl,
 * MSRC001_1033) is used to select either cycle or micro-ops counting
 * mode.
 *
 * The rip of IBS samples has skid 0. Thus, IBS supports precise
 * levels 1 and 2 and the PERF_EFLAGS_EXACT is set. In rare cases the
 * rip is invalid when IBS was not able to record the rip correctly.
 * We clear PERF_EFLAGS_EXACT and take the rip from pt_regs then.
 *
 */
static int perf_ibs_precise_event(struct perf_event *event, u64 *config)
{
	switch (event->attr.precise_ip) {
	case 0:
		return -ENOENT;
	case 1:
	case 2:
		break;
	default:
		return -EOPNOTSUPP;
	}

	switch (event->attr.type) {
	case PERF_TYPE_HARDWARE:
		switch (event->attr.config) {
		case PERF_COUNT_HW_CPU_CYCLES:
			*config = 0;
			return 0;
		}
		break;
	case PERF_TYPE_RAW:
		switch (event->attr.config) {
		case 0x0076:
			*config = 0;
			return 0;
		case 0x00C1:
			*config = IBS_OP_CNT_CTL;
			return 0;
		}
		break;
	default:
		return -ENOENT;
	}

	return -EOPNOTSUPP;
}

static int perf_ibs_init(struct perf_event *event)
{
	struct hw_perf_event *hwc = &event->hw;
	struct perf_ibs *perf_ibs;
	u64 max_cnt, config;
	int ret;

	perf_ibs = get_ibs_pmu(event->attr.type);
	if (perf_ibs) {
		config = event->attr.config;
	} else {
		perf_ibs = &perf_ibs_op;
		ret = perf_ibs_precise_event(event, &config);
		if (ret)
			return ret;
	}

	if (event->pmu != &perf_ibs->pmu)
		return -ENOENT;

	if (config & ~perf_ibs->config_mask)
		return -EINVAL;

	if (hwc->sample_period) {
		if (config & perf_ibs->cnt_mask)
			/* raw max_cnt may not be set */
			return -EINVAL;
		if (!event->attr.sample_freq && hwc->sample_period & 0x0f)
			/*
			 * lower 4 bits can not be set in ibs max cnt,
			 * but allowing it in case we adjust the
			 * sample period to set a frequency.
			 */
			return -EINVAL;
		hwc->sample_period &= ~0x0FULL;
		if (!hwc->sample_period)
			hwc->sample_period = 0x10;
	} else {
		max_cnt = config & perf_ibs->cnt_mask;
		config &= ~perf_ibs->cnt_mask;
		event->attr.sample_period = max_cnt << 4;
		hwc->sample_period = event->attr.sample_period;
	}

	if (!hwc->sample_period)
		return -EINVAL;

	/*
	 * If we modify hwc->sample_period, we also need to update
	 * hwc->last_period and hwc->period_left.
	 */
	hwc->last_period = hwc->sample_period;
	local64_set(&hwc->period_left, hwc->sample_period);

	hwc->config_base = perf_ibs->msr;
	hwc->config = config;

	return 0;
}

static int perf_ibs_set_period(struct perf_ibs *perf_ibs,
			       struct hw_perf_event *hwc, u64 *period)
{
	int overflow;

	/* ignore lower 4 bits in min count: */
	overflow = perf_event_set_period(hwc, 1<<4, perf_ibs->max_period, period);
	local64_set(&hwc->prev_count, 0);

	return overflow;
}

static u64 get_ibs_fetch_count(u64 config)
{
	return (config & IBS_FETCH_CNT) >> 12;
}

static u64 get_ibs_op_count(u64 config)
{
	u64 count = 0;

	/*
	 * If the internal 27-bit counter rolled over, the count is MaxCnt
	 * and the lower 7 bits of CurCnt are randomized.
	 * Otherwise CurCnt has the full 27-bit current counter value.
	 */
	if (config & IBS_OP_VAL) {
		count = (config & IBS_OP_MAX_CNT) << 4;
		if (ibs_caps & IBS_CAPS_OPCNTEXT)
			count += config & IBS_OP_MAX_CNT_EXT_MASK;
	} else if (ibs_caps & IBS_CAPS_RDWROPCNT) {
		count = (config & IBS_OP_CUR_CNT) >> 32;
	}

	return count;
}

static void
perf_ibs_event_update(struct perf_ibs *perf_ibs, struct perf_event *event,
		      u64 *config)
{
	u64 count = perf_ibs->get_count(*config);

	/*
	 * Set width to 64 since we do not overflow on max width but
	 * instead on max count. In perf_ibs_set_period() we clear
	 * prev count manually on overflow.
	 */
	while (!perf_event_try_update(event, count, 64)) {
		rdmsrl(event->hw.config_base, *config);
		count = perf_ibs->get_count(*config);
	}
}

static inline void perf_ibs_enable_event(struct perf_ibs *perf_ibs,
					 struct hw_perf_event *hwc, u64 config)
{
	u64 tmp = hwc->config | config;

	if (perf_ibs->fetch_count_reset_broken)
		wrmsrl(hwc->config_base, tmp & ~perf_ibs->enable_mask);

	wrmsrl(hwc->config_base, tmp | perf_ibs->enable_mask);
}

/*
 * Erratum #420 Instruction-Based Sampling Engine May Generate
 * Interrupt that Cannot Be Cleared:
 *
 * Must clear counter mask first, then clear the enable bit. See
 * Revision Guide for AMD Family 10h Processors, Publication #41322.
 */
static inline void perf_ibs_disable_event(struct perf_ibs *perf_ibs,
					  struct hw_perf_event *hwc, u64 config)
{
	config &= ~perf_ibs->cnt_mask;
	if (boot_cpu_data.x86 == 0x10)
		wrmsrl(hwc->config_base, config);
	config &= ~perf_ibs->enable_mask;
	wrmsrl(hwc->config_base, config);
}

/*
 * We cannot restore the ibs pmu state, so we always needs to update
 * the event while stopping it and then reset the state when starting
 * again. Thus, ignoring PERF_EF_RELOAD and PERF_EF_UPDATE flags in
 * perf_ibs_start()/perf_ibs_stop() and instead always do it.
 */
static void perf_ibs_start(struct perf_event *event, int flags)
{
	struct hw_perf_event *hwc = &event->hw;
	struct perf_ibs *perf_ibs = container_of(event->pmu, struct perf_ibs, pmu);
	struct cpu_perf_ibs *pcpu = this_cpu_ptr(perf_ibs->pcpu);
	u64 period, config = 0;

	if (WARN_ON_ONCE(!(hwc->state & PERF_HES_STOPPED)))
		return;

	WARN_ON_ONCE(!(hwc->state & PERF_HES_UPTODATE));
	hwc->state = 0;

	perf_ibs_set_period(perf_ibs, hwc, &period);
	if (perf_ibs == &perf_ibs_op && (ibs_caps & IBS_CAPS_OPCNTEXT)) {
		config |= period & IBS_OP_MAX_CNT_EXT_MASK;
		period &= ~IBS_OP_MAX_CNT_EXT_MASK;
	}
	config |= period >> 4;

	/*
	 * Set STARTED before enabling the hardware, such that a subsequent NMI
	 * must observe it.
	 */
	set_bit(IBS_STARTED,    pcpu->state);
	clear_bit(IBS_STOPPING, pcpu->state);
	perf_ibs_enable_event(perf_ibs, hwc, config);

	perf_event_update_userpage(event);
}

static void perf_ibs_stop(struct perf_event *event, int flags)
{
	struct hw_perf_event *hwc = &event->hw;
	struct perf_ibs *perf_ibs = container_of(event->pmu, struct perf_ibs, pmu);
	struct cpu_perf_ibs *pcpu = this_cpu_ptr(perf_ibs->pcpu);
	u64 config;
	int stopping;

	if (test_and_set_bit(IBS_STOPPING, pcpu->state))
		return;

	stopping = test_bit(IBS_STARTED, pcpu->state);

	if (!stopping && (hwc->state & PERF_HES_UPTODATE))
		return;

	rdmsrl(hwc->config_base, config);

	if (stopping) {
		/*
		 * Set STOPPED before disabling the hardware, such that it
		 * must be visible to NMIs the moment we clear the EN bit,
		 * at which point we can generate an !VALID sample which
		 * we need to consume.
		 */
		set_bit(IBS_STOPPED, pcpu->state);
		perf_ibs_disable_event(perf_ibs, hwc, config);
		/*
		 * Clear STARTED after disabling the hardware; if it were
		 * cleared before an NMI hitting after the clear but before
		 * clearing the EN bit might think it a spurious NMI and not
		 * handle it.
		 *
		 * Clearing it after, however, creates the problem of the NMI
		 * handler seeing STARTED but not having a valid sample.
		 */
		clear_bit(IBS_STARTED, pcpu->state);
		WARN_ON_ONCE(hwc->state & PERF_HES_STOPPED);
		hwc->state |= PERF_HES_STOPPED;
	}

	if (hwc->state & PERF_HES_UPTODATE)
		return;

	/*
	 * Clear valid bit to not count rollovers on update, rollovers
	 * are only updated in the irq handler.
	 */
	config &= ~perf_ibs->valid_mask;

	perf_ibs_event_update(perf_ibs, event, &config);
	hwc->state |= PERF_HES_UPTODATE;
}

static int perf_ibs_add(struct perf_event *event, int flags)
{
	struct perf_ibs *perf_ibs = container_of(event->pmu, struct perf_ibs, pmu);
	struct cpu_perf_ibs *pcpu = this_cpu_ptr(perf_ibs->pcpu);

	if (test_and_set_bit(IBS_ENABLED, pcpu->state))
		return -ENOSPC;

	event->hw.state = PERF_HES_UPTODATE | PERF_HES_STOPPED;

	pcpu->event = event;

	if (flags & PERF_EF_START)
		perf_ibs_start(event, PERF_EF_RELOAD);

	return 0;
}

static void perf_ibs_del(struct perf_event *event, int flags)
{
	struct perf_ibs *perf_ibs = container_of(event->pmu, struct perf_ibs, pmu);
	struct cpu_perf_ibs *pcpu = this_cpu_ptr(perf_ibs->pcpu);

	if (!test_and_clear_bit(IBS_ENABLED, pcpu->state))
		return;

	perf_ibs_stop(event, PERF_EF_UPDATE);

	pcpu->event = NULL;

	perf_event_update_userpage(event);
}

static void perf_ibs_read(struct perf_event *event) { }

PMU_FORMAT_ATTR(rand_en,	"config:57");
PMU_FORMAT_ATTR(cnt_ctl,	"config:19");

static struct attribute *ibs_fetch_format_attrs[] = {
	&format_attr_rand_en.attr,
	NULL,
};

static struct attribute *ibs_op_format_attrs[] = {
	NULL,	/* &format_attr_cnt_ctl.attr if IBS_CAPS_OPCNT */
	NULL,
};

static struct perf_ibs perf_ibs_fetch = {
	.pmu = {
		.task_ctx_nr	= perf_invalid_context,

		.event_init	= perf_ibs_init,
		.add		= perf_ibs_add,
		.del		= perf_ibs_del,
		.start		= perf_ibs_start,
		.stop		= perf_ibs_stop,
		.read		= perf_ibs_read,
		.capabilities	= PERF_PMU_CAP_NO_EXCLUDE,
	},
	.msr			= MSR_AMD64_IBSFETCHCTL,
	.config_mask		= IBS_FETCH_CONFIG_MASK,
	.cnt_mask		= IBS_FETCH_MAX_CNT,
	.enable_mask		= IBS_FETCH_ENABLE,
	.valid_mask		= IBS_FETCH_VAL,
	.max_period		= IBS_FETCH_MAX_CNT << 4,
	.offset_mask		= { MSR_AMD64_IBSFETCH_REG_MASK },
	.offset_max		= MSR_AMD64_IBSFETCH_REG_COUNT,
	.format_attrs		= ibs_fetch_format_attrs,

	.get_count		= get_ibs_fetch_count,
};

static struct perf_ibs perf_ibs_op = {
	.pmu = {
		.task_ctx_nr	= perf_invalid_context,

		.event_init	= perf_ibs_init,
		.add		= perf_ibs_add,
		.del		= perf_ibs_del,
		.start		= perf_ibs_start,
		.stop		= perf_ibs_stop,
		.read		= perf_ibs_read,
	},
	.msr			= MSR_AMD64_IBSOPCTL,
	.config_mask		= IBS_OP_CONFIG_MASK,
	.cnt_mask		= IBS_OP_MAX_CNT | IBS_OP_CUR_CNT |
				  IBS_OP_CUR_CNT_RAND,
	.enable_mask		= IBS_OP_ENABLE,
	.valid_mask		= IBS_OP_VAL,
	.max_period		= IBS_OP_MAX_CNT << 4,
	.offset_mask		= { MSR_AMD64_IBSOP_REG_MASK },
	.offset_max		= MSR_AMD64_IBSOP_REG_COUNT,
	.format_attrs		= ibs_op_format_attrs,

	.get_count		= get_ibs_op_count,
};

static int perf_ibs_handle_irq(struct perf_ibs *perf_ibs, struct pt_regs *iregs)
{
	struct cpu_perf_ibs *pcpu = this_cpu_ptr(perf_ibs->pcpu);
	struct perf_event *event = pcpu->event;
	struct hw_perf_event *hwc;
	struct perf_sample_data data;
	struct perf_raw_record raw;
	struct pt_regs regs;
	struct perf_ibs_data ibs_data;
	int offset, size, check_rip, offset_max, throttle = 0;
	unsigned int msr;
	u64 *buf, *config, period, new_config = 0;

	if (!test_bit(IBS_STARTED, pcpu->state)) {
fail:
		/*
		 * Catch spurious interrupts after stopping IBS: After
		 * disabling IBS there could be still incoming NMIs
		 * with samples that even have the valid bit cleared.
		 * Mark all this NMIs as handled.
		 */
		if (test_and_clear_bit(IBS_STOPPED, pcpu->state))
			return 1;

		return 0;
	}

	if (WARN_ON_ONCE(!event))
		goto fail;

	hwc = &event->hw;
	msr = hwc->config_base;
	buf = ibs_data.regs;
	rdmsrl(msr, *buf);
	if (!(*buf++ & perf_ibs->valid_mask))
		goto fail;

	config = &ibs_data.regs[0];
	perf_ibs_event_update(perf_ibs, event, config);
	perf_sample_data_init(&data, 0, hwc->last_period);
	if (!perf_ibs_set_period(perf_ibs, hwc, &period))
		goto out;	/* no sw counter overflow */

	ibs_data.caps = ibs_caps;
	size = 1;
	offset = 1;
	check_rip = (perf_ibs == &perf_ibs_op && (ibs_caps & IBS_CAPS_RIPINVALIDCHK));
	if (event->attr.sample_type & PERF_SAMPLE_RAW)
		offset_max = perf_ibs->offset_max;
	else if (check_rip)
		offset_max = 3;
	else
		offset_max = 1;
	do {
		rdmsrl(msr + offset, *buf++);
		size++;
		offset = find_next_bit(perf_ibs->offset_mask,
				       perf_ibs->offset_max,
				       offset + 1);
	} while (offset < offset_max);
	/*
	 * Read IbsBrTarget, IbsOpData4, and IbsExtdCtl separately
	 * depending on their availability.
	 * Can't add to offset_max as they are staggered
	 */
	if (event->attr.sample_type & PERF_SAMPLE_RAW) {
		if (perf_ibs == &perf_ibs_op) {
			if (ibs_caps & IBS_CAPS_BRNTRGT) {
				rdmsrl(MSR_AMD64_IBSBRTARGET, *buf++);
				size++;
			}
			if (ibs_caps & IBS_CAPS_OPDATA4) {
				rdmsrl(MSR_AMD64_IBSOPDATA4, *buf++);
				size++;
			}
		}
		if (perf_ibs == &perf_ibs_fetch && (ibs_caps & IBS_CAPS_FETCHCTLEXTD)) {
			rdmsrl(MSR_AMD64_ICIBSEXTDCTL, *buf++);
			size++;
		}
	}
	ibs_data.size = sizeof(u64) * size;

	regs = *iregs;
	if (check_rip && (ibs_data.regs[2] & IBS_RIP_INVALID)) {
		regs.flags &= ~PERF_EFLAGS_EXACT;
	} else {
		set_linear_ip(&regs, ibs_data.regs[1]);
		regs.flags |= PERF_EFLAGS_EXACT;
	}

	if (event->attr.sample_type & PERF_SAMPLE_RAW) {
		raw = (struct perf_raw_record){
			.frag = {
				.size = sizeof(u32) + ibs_data.size,
				.data = ibs_data.data,
			},
		};
		data.raw = &raw;
	}

	throttle = perf_event_overflow(event, &data, &regs);
out:
	if (throttle) {
		perf_ibs_stop(event, 0);
	} else {
		if (perf_ibs == &perf_ibs_op) {
			if (ibs_caps & IBS_CAPS_OPCNTEXT) {
				new_config = period & IBS_OP_MAX_CNT_EXT_MASK;
				period &= ~IBS_OP_MAX_CNT_EXT_MASK;
			}
			if ((ibs_caps & IBS_CAPS_RDWROPCNT) && (*config & IBS_OP_CNT_CTL))
				new_config |= *config & IBS_OP_CUR_CNT_RAND;
		}
		new_config |= period >> 4;

		perf_ibs_enable_event(perf_ibs, hwc, new_config);
	}

	perf_event_update_userpage(event);

	return 1;
}

static int
perf_ibs_nmi_handler(unsigned int cmd, struct pt_regs *regs)
{
	u64 stamp = sched_clock();
	int handled = 0;

	handled += perf_ibs_handle_irq(&perf_ibs_fetch, regs);
	handled += perf_ibs_handle_irq(&perf_ibs_op, regs);

	if (handled)
		inc_irq_stat(apic_perf_irqs);

	perf_sample_event_took(sched_clock() - stamp);

	return handled;
}
NOKPROBE_SYMBOL(perf_ibs_nmi_handler);

static __init int perf_ibs_pmu_init(struct perf_ibs *perf_ibs, char *name)
{
	struct cpu_perf_ibs __percpu *pcpu;
	int ret;

	pcpu = alloc_percpu(struct cpu_perf_ibs);
	if (!pcpu)
		return -ENOMEM;

	perf_ibs->pcpu = pcpu;

	/* register attributes */
	if (perf_ibs->format_attrs[0]) {
		memset(&perf_ibs->format_group, 0, sizeof(perf_ibs->format_group));
		perf_ibs->format_group.name	= "format";
		perf_ibs->format_group.attrs	= perf_ibs->format_attrs;

		memset(&perf_ibs->attr_groups, 0, sizeof(perf_ibs->attr_groups));
		perf_ibs->attr_groups[0]	= &perf_ibs->format_group;
		perf_ibs->pmu.attr_groups	= perf_ibs->attr_groups;
	}

	ret = perf_pmu_register(&perf_ibs->pmu, name, -1);
	if (ret) {
		perf_ibs->pcpu = NULL;
		free_percpu(pcpu);
	}

	return ret;
}

static __init void perf_event_ibs_init(void)
{
	struct attribute **attr = ibs_op_format_attrs;

	/*
	 * Some chips fail to reset the fetch count when it is written; instead
	 * they need a 0-1 transition of IbsFetchEn.
	 */
	if (boot_cpu_data.x86 >= 0x16 && boot_cpu_data.x86 <= 0x18)
		perf_ibs_fetch.fetch_count_reset_broken = 1;

	perf_ibs_pmu_init(&perf_ibs_fetch, "ibs_fetch");

	if (ibs_caps & IBS_CAPS_OPCNT) {
		perf_ibs_op.config_mask |= IBS_OP_CNT_CTL;
		*attr++ = &format_attr_cnt_ctl.attr;
	}

	if (ibs_caps & IBS_CAPS_OPCNTEXT) {
		perf_ibs_op.max_period  |= IBS_OP_MAX_CNT_EXT_MASK;
		perf_ibs_op.config_mask	|= IBS_OP_MAX_CNT_EXT_MASK;
		perf_ibs_op.cnt_mask    |= IBS_OP_MAX_CNT_EXT_MASK;
	}

	perf_ibs_pmu_init(&perf_ibs_op, "ibs_op");

	register_nmi_handler(NMI_LOCAL, perf_ibs_nmi_handler, 0, "perf_ibs");
	pr_info("perf: AMD IBS detected (0x%08x)\n", ibs_caps);
}

#else /* defined(CONFIG_PERF_EVENTS) && defined(CONFIG_CPU_SUP_AMD) */

static __init void perf_event_ibs_init(void) { }

#endif

/* IBS - apic initialization, for perf and oprofile */

static __init u32 __get_ibs_caps(void)
{
	u32 caps;
	unsigned int max_level;

	if (!boot_cpu_has(X86_FEATURE_IBS))
		return 0;

	/* check IBS cpuid feature flags */
	max_level = cpuid_eax(0x80000000);
	if (max_level < IBS_CPUID_FEATURES)
		return IBS_CAPS_DEFAULT;

	caps = cpuid_eax(IBS_CPUID_FEATURES);
	if (!(caps & IBS_CAPS_AVAIL))
		/* cpuid flags not valid */
		return IBS_CAPS_DEFAULT;

	return caps;
}

u32 get_ibs_caps(void)
{
	return ibs_caps;
}

EXPORT_SYMBOL(get_ibs_caps);

static inline int get_eilvt(int offset)
{
	return !setup_APIC_eilvt(offset, 0, APIC_EILVT_MSG_NMI, 1);
}

static inline int put_eilvt(int offset)
{
	return !setup_APIC_eilvt(offset, 0, 0, 1);
}

/*
 * Check and reserve APIC extended interrupt LVT offset for IBS if available.
 */
static inline int ibs_eilvt_valid(void)
{
	int offset;
	u64 val;
	int valid = 0;

	preempt_disable();

	rdmsrl(MSR_AMD64_IBSCTL, val);
	offset = val & IBSCTL_LVT_OFFSET_MASK;

	if (!(val & IBSCTL_LVT_OFFSET_VALID)) {
		pr_err(FW_BUG "cpu %d, invalid IBS interrupt offset %d (MSR%08X=0x%016llx)\n",
		       smp_processor_id(), offset, MSR_AMD64_IBSCTL, val);
		goto out;
	}

	if (!get_eilvt(offset)) {
		pr_err(FW_BUG "cpu %d, IBS interrupt offset %d not available (MSR%08X=0x%016llx)\n",
		       smp_processor_id(), offset, MSR_AMD64_IBSCTL, val);
		goto out;
	}

	valid = 1;
out:
	preempt_enable();

	return valid;
}

static int setup_ibs_ctl(int ibs_eilvt_off)
{
	struct pci_dev *cpu_cfg;
	int nodes;
	u32 value = 0;

	nodes = 0;
	cpu_cfg = NULL;
	do {
		cpu_cfg = pci_get_device(PCI_VENDOR_ID_AMD,
					 PCI_DEVICE_ID_AMD_10H_NB_MISC,
					 cpu_cfg);
		if (!cpu_cfg)
			break;
		++nodes;
		pci_write_config_dword(cpu_cfg, IBSCTL, ibs_eilvt_off
				       | IBSCTL_LVT_OFFSET_VALID);
		pci_read_config_dword(cpu_cfg, IBSCTL, &value);
		if (value != (ibs_eilvt_off | IBSCTL_LVT_OFFSET_VALID)) {
			pci_dev_put(cpu_cfg);
			pr_debug("Failed to setup IBS LVT offset, IBSCTL = 0x%08x\n",
				 value);
			return -EINVAL;
		}
	} while (1);

	if (!nodes) {
		pr_debug("No CPU node configured for IBS\n");
		return -ENODEV;
	}

	return 0;
}

/*
 * This runs only on the current cpu. We try to find an LVT offset and
 * setup the local APIC. For this we must disable preemption. On
 * success we initialize all nodes with this offset. This updates then
 * the offset in the IBS_CTL per-node msr. The per-core APIC setup of
 * the IBS interrupt vector is handled by perf_ibs_cpu_notifier that
 * is using the new offset.
 */
static void force_ibs_eilvt_setup(void)
{
	int offset;
	int ret;

	preempt_disable();
	/* find the next free available EILVT entry, skip offset 0 */
	for (offset = 1; offset < APIC_EILVT_NR_MAX; offset++) {
		if (get_eilvt(offset))
			break;
	}
	preempt_enable();

	if (offset == APIC_EILVT_NR_MAX) {
		pr_debug("No EILVT entry available\n");
		return;
	}

	ret = setup_ibs_ctl(offset);
	if (ret)
		goto out;

	if (!ibs_eilvt_valid())
		goto out;

	pr_info("LVT offset %d assigned\n", offset);

	return;
out:
	preempt_disable();
	put_eilvt(offset);
	preempt_enable();
	return;
}

static void ibs_eilvt_setup(void)
{
	/*
	 * Force LVT offset assignment for family 10h: The offsets are
	 * not assigned by the BIOS for this family, so the OS is
	 * responsible for doing it. If the OS assignment fails, fall
	 * back to BIOS settings and try to setup this.
	 */
	if (boot_cpu_data.x86 == 0x10)
		force_ibs_eilvt_setup();
}

static inline int get_ibs_lvt_offset(void)
{
	u64 val;

	rdmsrl(MSR_AMD64_IBSCTL, val);
	if (!(val & IBSCTL_LVT_OFFSET_VALID))
		return -EINVAL;

	return val & IBSCTL_LVT_OFFSET_MASK;
}

static void setup_APIC_ibs(void)
{
	int offset;

	offset = get_ibs_lvt_offset();
	if (offset < 0)
		goto failed;

	if (!setup_APIC_eilvt(offset, 0, APIC_EILVT_MSG_NMI, 0))
		return;
failed:
	pr_warn("perf: IBS APIC setup failed on cpu #%d\n",
		smp_processor_id());
}

static void clear_APIC_ibs(void)
{
	int offset;

	offset = get_ibs_lvt_offset();
	if (offset >= 0)
		setup_APIC_eilvt(offset, 0, APIC_EILVT_MSG_FIX, 1);
}

static int x86_pmu_amd_ibs_starting_cpu(unsigned int cpu)
{
	setup_APIC_ibs();
	return 0;
}

#ifdef CONFIG_PM

static int perf_ibs_suspend(void)
{
	clear_APIC_ibs();
	return 0;
}

static void perf_ibs_resume(void)
{
	ibs_eilvt_setup();
	setup_APIC_ibs();
}

static struct syscore_ops perf_ibs_syscore_ops = {
	.resume		= perf_ibs_resume,
	.suspend	= perf_ibs_suspend,
};

static void perf_ibs_pm_init(void)
{
	register_syscore_ops(&perf_ibs_syscore_ops);
}

#else

static inline void perf_ibs_pm_init(void) { }

#endif

static int x86_pmu_amd_ibs_dying_cpu(unsigned int cpu)
{
	clear_APIC_ibs();
	return 0;
}

static __init int amd_ibs_init(void)
{
	u32 caps;

	caps = __get_ibs_caps();
	if (!caps)
		return -ENODEV;	/* ibs not supported by the cpu */

	ibs_eilvt_setup();

	if (!ibs_eilvt_valid())
		return -EINVAL;

	perf_ibs_pm_init();

	ibs_caps = caps;
	/* make ibs_caps visible to other cpus: */
	smp_mb();
	/*
	 * x86_pmu_amd_ibs_starting_cpu will be called from core on
	 * all online cpus.
	 */
	cpuhp_setup_state(CPUHP_AP_PERF_X86_AMD_IBS_STARTING,
			  "perf/x86/amd/ibs:starting",
			  x86_pmu_amd_ibs_starting_cpu,
			  x86_pmu_amd_ibs_dying_cpu);

	perf_event_ibs_init();

	return 0;
}

/* Since we need the pci subsystem to init ibs we can't do this earlier: */
device_initcall(amd_ibs_init);