aboutsummaryrefslogtreecommitdiffstats
path: root/arch/x86/kernel/cpu/mcheck/mce_amd_64.c
blob: 752fb16a817d7187ac1732e66c426a12a19ff352 (plain) (blame)
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
/*
 *  (c) 2005, 2006 Advanced Micro Devices, Inc.
 *  Your use of this code is subject to the terms and conditions of the
 *  GNU general public license version 2. See "COPYING" or
 *  http://www.gnu.org/licenses/gpl.html
 *
 *  Written by Jacob Shin - AMD, Inc.
 *
 *  Support : jacob.shin@amd.com
 *
 *  April 2006
 *     - added support for AMD Family 0x10 processors
 *
 *  All MC4_MISCi registers are shared between multi-cores
 */

#include <linux/cpu.h>
#include <linux/errno.h>
#include <linux/init.h>
#include <linux/interrupt.h>
#include <linux/kobject.h>
#include <linux/notifier.h>
#include <linux/sched.h>
#include <linux/smp.h>
#include <linux/sysdev.h>
#include <linux/sysfs.h>
#include <asm/apic.h>
#include <asm/mce.h>
#include <asm/msr.h>
#include <asm/percpu.h>
#include <asm/idle.h>

#define PFX               "mce_threshold: "
#define VERSION           "version 1.1.1"
#define NR_BANKS          6
#define NR_BLOCKS         9
#define THRESHOLD_MAX     0xFFF
#define INT_TYPE_APIC     0x00020000
#define MASK_VALID_HI     0x80000000
#define MASK_CNTP_HI      0x40000000
#define MASK_LOCKED_HI    0x20000000
#define MASK_LVTOFF_HI    0x00F00000
#define MASK_COUNT_EN_HI  0x00080000
#define MASK_INT_TYPE_HI  0x00060000
#define MASK_OVERFLOW_HI  0x00010000
#define MASK_ERR_COUNT_HI 0x00000FFF
#define MASK_BLKPTR_LO    0xFF000000
#define MCG_XBLK_ADDR     0xC0000400

struct threshold_block {
	unsigned int block;
	unsigned int bank;
	unsigned int cpu;
	u32 address;
	u16 interrupt_enable;
	u16 threshold_limit;
	struct kobject kobj;
	struct list_head miscj;
};

/* defaults used early on boot */
static struct threshold_block threshold_defaults = {
	.interrupt_enable = 0,
	.threshold_limit = THRESHOLD_MAX,
};

struct threshold_bank {
	struct kobject kobj;
	struct threshold_block *blocks;
	cpumask_t cpus;
};
static DEFINE_PER_CPU(struct threshold_bank *, threshold_banks[NR_BANKS]);

#ifdef CONFIG_SMP
static unsigned char shared_bank[NR_BANKS] = {
	0, 0, 0, 0, 1
};
#endif

static DEFINE_PER_CPU(unsigned char, bank_map);	/* see which banks are on */

/*
 * CPU Initialization
 */

/* must be called with correct cpu affinity */
static void threshold_restart_bank(struct threshold_block *b,
				   int reset, u16 old_limit)
{
	u32 mci_misc_hi, mci_misc_lo;

	rdmsr(b->address, mci_misc_lo, mci_misc_hi);

	if (b->threshold_limit < (mci_misc_hi & THRESHOLD_MAX))
		reset = 1;	/* limit cannot be lower than err count */

	if (reset) {		/* reset err count and overflow bit */
		mci_misc_hi =
		    (mci_misc_hi & ~(MASK_ERR_COUNT_HI | MASK_OVERFLOW_HI)) |
		    (THRESHOLD_MAX - b->threshold_limit);
	} else if (old_limit) {	/* change limit w/o reset */
		int new_count = (mci_misc_hi & THRESHOLD_MAX) +
		    (old_limit - b->threshold_limit);
		mci_misc_hi = (mci_misc_hi & ~MASK_ERR_COUNT_HI) |
		    (new_count & THRESHOLD_MAX);
	}

	b->interrupt_enable ?
	    (mci_misc_hi = (mci_misc_hi & ~MASK_INT_TYPE_HI) | INT_TYPE_APIC) :
	    (mci_misc_hi &= ~MASK_INT_TYPE_HI);

	mci_misc_hi |= MASK_COUNT_EN_HI;
	wrmsr(b->address, mci_misc_lo, mci_misc_hi);
}

/* cpu init entry point, called from mce.c with preempt off */
void __cpuinit mce_amd_feature_init(struct cpuinfo_x86 *c)
{
	unsigned int bank, block;
	unsigned int cpu = smp_processor_id();
	u32 low = 0, high = 0, address = 0;

	for (bank = 0; bank < NR_BANKS; ++bank) {
		for (block = 0; block < NR_BLOCKS; ++block) {
			if (block == 0)
				address = MSR_IA32_MC0_MISC + bank * 4;
			else if (block == 1) {
				address = (low & MASK_BLKPTR_LO) >> 21;
				if (!address)
					break;
				address += MCG_XBLK_ADDR;
			}
			else
				++address;

			if (rdmsr_safe(address, &low, &high))
				break;

			if (!(high & MASK_VALID_HI)) {
				if (block)
					continue;
				else
					break;
			}

			if (!(high & MASK_CNTP_HI)  ||
			     (high & MASK_LOCKED_HI))
				continue;

			if (!block)
				per_cpu(bank_map, cpu) |= (1 << bank);
#ifdef CONFIG_SMP
			if (shared_bank[bank] && c->cpu_core_id)
				break;
#endif
			high &= ~MASK_LVTOFF_HI;
			high |= K8_APIC_EXT_LVT_ENTRY_THRESHOLD << 20;
			wrmsr(address, low, high);

			setup_APIC_extended_lvt(K8_APIC_EXT_LVT_ENTRY_THRESHOLD,
						THRESHOLD_APIC_VECTOR,
						K8_APIC_EXT_INT_MSG_FIX, 0);

			threshold_defaults.address = address;
			threshold_restart_bank(&threshold_defaults, 0, 0);
		}
	}
}

/*
 * APIC Interrupt Handler
 */

/*
 * threshold interrupt handler will service THRESHOLD_APIC_VECTOR.
 * the interrupt goes off when error_count reaches threshold_limit.
 * the handler will simply log mcelog w/ software defined bank number.
 */
asmlinkage void mce_threshold_interrupt(void)
{
	unsigned int bank, block;
	struct mce m;
	u32 low = 0, high = 0, address = 0;

	ack_APIC_irq();
	exit_idle();
	irq_enter();

	memset(&m, 0, sizeof(m));
	rdtscll(m.tsc);
	m.cpu = smp_processor_id();

	/* assume first bank caused it */
	for (bank = 0; bank < NR_BANKS; ++bank) {
		if (!(per_cpu(bank_map, m.cpu) & (1 << bank)))
			continue;
		for (block = 0; block < NR_BLOCKS; ++block) {
			if (block == 0)
				address = MSR_IA32_MC0_MISC + bank * 4;
			else if (block == 1) {
				address = (low & MASK_BLKPTR_LO) >> 21;
				if (!address)
					break;
				address += MCG_XBLK_ADDR;
			}
			else
				++address;

			if (rdmsr_safe(address, &low, &high))
				break;

			if (!(high & MASK_VALID_HI)) {
				if (block)
					continue;
				else
					break;
			}

			if (!(high & MASK_CNTP_HI)  ||
			     (high & MASK_LOCKED_HI))
				continue;

			/* Log the machine check that caused the threshold
			   event. */
			do_machine_check(NULL, 0);

			if (high & MASK_OVERFLOW_HI) {
				rdmsrl(address, m.misc);
				rdmsrl(MSR_IA32_MC0_STATUS + bank * 4,
				       m.status);
				m.bank = K8_MCE_THRESHOLD_BASE
				       + bank * NR_BLOCKS
				       + block;
				mce_log(&m);
				goto out;
			}
		}
	}
out:
	add_pda(irq_threshold_count, 1);
	irq_exit();
}

/*
 * Sysfs Interface
 */

struct threshold_attr {
	struct attribute attr;
	ssize_t(*show) (struct threshold_block *, char *);
	ssize_t(*store) (struct threshold_block *, const char *, size_t count);
};

static cpumask_t affinity_set(unsigned int cpu)
{
	cpumask_t oldmask = current->cpus_allowed;
	cpumask_t newmask = CPU_MASK_NONE;
	cpu_set(cpu, newmask);
	set_cpus_allowed(current, newmask);
	return oldmask;
}

static void affinity_restore(cpumask_t oldmask)
{
	set_cpus_allowed(current, oldmask);
}

#define SHOW_FIELDS(name)                                           \
static ssize_t show_ ## name(struct threshold_block * b, char *buf) \
{                                                                   \
        return sprintf(buf, "%lx\n", (unsigned long) b->name);      \
}
SHOW_FIELDS(interrupt_enable)
SHOW_FIELDS(threshold_limit)

static ssize_t store_interrupt_enable(struct threshold_block *b,
				      const char *buf, size_t count)
{
	char *end;
	cpumask_t oldmask;
	unsigned long new = simple_strtoul(buf, &end, 0);
	if (end == buf)
		return -EINVAL;
	b->interrupt_enable = !!new;

	oldmask = affinity_set(b->cpu);
	threshold_restart_bank(b, 0, 0);
	affinity_restore(oldmask);

	return end - buf;
}

static ssize_t store_threshold_limit(struct threshold_block *b,
				     const char *buf, size_t count)
{
	char *end;
	cpumask_t oldmask;
	u16 old;
	unsigned long new = simple_strtoul(buf, &end, 0);
	if (end == buf)
		return -EINVAL;
	if (new > THRESHOLD_MAX)
		new = THRESHOLD_MAX;
	if (new < 1)
		new = 1;
	old = b->threshold_limit;
	b->threshold_limit = new;

	oldmask = affinity_set(b->cpu);
	threshold_restart_bank(b, 0, old);
	affinity_restore(oldmask);

	return end - buf;
}

static ssize_t show_error_count(struct threshold_block *b, char *buf)
{
	u32 high, low;
	cpumask_t oldmask;
	oldmask = affinity_set(b->cpu);
	rdmsr(b->address, low, high);
	affinity_restore(oldmask);
	return sprintf(buf, "%x\n",
		       (high & 0xFFF) - (THRESHOLD_MAX - b->threshold_limit));
}

static ssize_t store_error_count(struct threshold_block *b,
				 const char *buf, size_t count)
{
	cpumask_t oldmask;
	oldmask = affinity_set(b->cpu);
	threshold_restart_bank(b, 1, 0);
	affinity_restore(oldmask);
	return 1;
}

#define THRESHOLD_ATTR(_name,_mode,_show,_store) {            \
        .attr = {.name = __stringify(_name), .mode = _mode }, \
        .show = _show,                                        \
        .store = _store,                                      \
};

#define RW_ATTR(name)                                           \
static struct threshold_attr name =                             \
        THRESHOLD_ATTR(name, 0644, show_## name, store_## name)

RW_ATTR(interrupt_enable);
RW_ATTR(threshold_limit);
RW_ATTR(error_count);

static struct attribute *default_attrs[] = {
	&interrupt_enable.attr,
	&threshold_limit.attr,
	&error_count.attr,
	NULL
};

#define to_block(k) container_of(k, struct threshold_block, kobj)
#define to_attr(a) container_of(a, struct threshold_attr, attr)

static ssize_t show(struct kobject *kobj, struct attribute *attr, char *buf)
{
	struct threshold_block *b = to_block(kobj);
	struct threshold_attr *a = to_attr(attr);
	ssize_t ret;
	ret = a->show ? a->show(b, buf) : -EIO;
	return ret;
}

static ssize_t store(struct kobject *kobj, struct attribute *attr,
		     const char *buf, size_t count)
{
	struct threshold_block *b = to_block(kobj);
	struct threshold_attr *a = to_attr(attr);
	ssize_t ret;
	ret = a->store ? a->store(b, buf, count) : -EIO;
	return ret;
}

static struct sysfs_ops threshold_ops = {
	.show = show,
	.store = store,
};

static struct kobj_type threshold_ktype = {
	.sysfs_ops = &threshold_ops,
	.default_attrs = default_attrs,
};

static __cpuinit int allocate_threshold_blocks(unsigned int cpu,
					       unsigned int bank,
					       unsigned int block,
					       u32 address)
{
	int err;
	u32 low, high;
	struct threshold_block *b = NULL;

	if ((bank >= NR_BANKS) || (block >= NR_BLOCKS))
		return 0;

	if (rdmsr_safe(address, &low, &high))
		return 0;

	if (!(high & MASK_VALID_HI)) {
		if (block)
			goto recurse;
		else
			return 0;
	}

	if (!(high & MASK_CNTP_HI)  ||
	     (high & MASK_LOCKED_HI))
		goto recurse;

	b = kzalloc(sizeof(struct threshold_block), GFP_KERNEL);
	if (!b)
		return -ENOMEM;

	b->block = block;
	b->bank = bank;
	b->cpu = cpu;
	b->address = address;
	b->interrupt_enable = 0;
	b->threshold_limit = THRESHOLD_MAX;

	INIT_LIST_HEAD(&b->miscj);

	if (per_cpu(threshold_banks, cpu)[bank]->blocks)
		list_add(&b->miscj,
			 &per_cpu(threshold_banks, cpu)[bank]->blocks->miscj);
	else
		per_cpu(threshold_banks, cpu)[bank]->blocks = b;

	kobject_set_name(&b->kobj, "misc%i", block);
	b->kobj.parent = &per_cpu(threshold_banks, cpu)[bank]->kobj;
	b->kobj.ktype = &threshold_ktype;
	err = kobject_register(&b->kobj);
	if (err)
		goto out_free;
recurse:
	if (!block) {
		address = (low & MASK_BLKPTR_LO) >> 21;
		if (!address)
			return 0;
		address += MCG_XBLK_ADDR;
	} else
		++address;

	err = allocate_threshold_blocks(cpu, bank, ++block, address);
	if (err)
		goto out_free;

	return err;

out_free:
	if (b) {
		kobject_unregister(&b->kobj);
		kfree(b);
	}
	return err;
}

/* symlinks sibling shared banks to first core.  first core owns dir/files. */
static __cpuinit int threshold_create_bank(unsigned int cpu, unsigned int bank)
{
	int i, err = 0;
	struct threshold_bank *b = NULL;
	cpumask_t oldmask = CPU_MASK_NONE;
	char name[32];

	sprintf(name, "threshold_bank%i", bank);

#ifdef CONFIG_SMP
	if (cpu_data(cpu).cpu_core_id && shared_bank[bank]) {	/* symlink */
		i = first_cpu(per_cpu(cpu_core_map, cpu));

		/* first core not up yet */
		if (cpu_data(i).cpu_core_id)
			goto out;

		/* already linked */
		if (per_cpu(threshold_banks, cpu)[bank])
			goto out;

		b = per_cpu(threshold_banks, i)[bank];

		if (!b)
			goto out;

		err = sysfs_create_link(&per_cpu(device_mce, cpu).kobj,
					&b->kobj, name);
		if (err)
			goto out;

		b->cpus = per_cpu(cpu_core_map, cpu);
		per_cpu(threshold_banks, cpu)[bank] = b;
		goto out;
	}
#endif

	b = kzalloc(sizeof(struct threshold_bank), GFP_KERNEL);
	if (!b) {
		err = -ENOMEM;
		goto out;
	}

	kobject_set_name(&b->kobj, "threshold_bank%i", bank);
	b->kobj.parent = &per_cpu(device_mce, cpu).kobj;
#ifndef CONFIG_SMP
	b->cpus = CPU_MASK_ALL;
#else
	b->cpus = per_cpu(cpu_core_map, cpu);
#endif
	err = kobject_register(&b->kobj);
	if (err)
		goto out_free;

	per_cpu(threshold_banks, cpu)[bank] = b;

	oldmask = affinity_set(cpu);
	err = allocate_threshold_blocks(cpu, bank, 0,
					MSR_IA32_MC0_MISC + bank * 4);
	affinity_restore(oldmask);

	if (err)
		goto out_free;

	for_each_cpu_mask(i, b->cpus) {
		if (i == cpu)
			continue;

		err = sysfs_create_link(&per_cpu(device_mce, i).kobj,
					&b->kobj, name);
		if (err)
			goto out;

		per_cpu(threshold_banks, i)[bank] = b;
	}

	goto out;

out_free:
	per_cpu(threshold_banks, cpu)[bank] = NULL;
	kfree(b);
out:
	return err;
}

/* create dir/files for all valid threshold banks */
static __cpuinit int threshold_create_device(unsigned int cpu)
{
	unsigned int bank;
	int err = 0;

	for (bank = 0; bank < NR_BANKS; ++bank) {
		if (!(per_cpu(bank_map, cpu) & 1 << bank))
			continue;
		err = threshold_create_bank(cpu, bank);
		if (err)
			goto out;
	}
out:
	return err;
}

/*
 * let's be hotplug friendly.
 * in case of multiple core processors, the first core always takes ownership
 *   of shared sysfs dir/files, and rest of the cores will be symlinked to it.
 */

static void deallocate_threshold_block(unsigned int cpu,
						 unsigned int bank)
{
	struct threshold_block *pos = NULL;
	struct threshold_block *tmp = NULL;
	struct threshold_bank *head = per_cpu(threshold_banks, cpu)[bank];

	if (!head)
		return;

	list_for_each_entry_safe(pos, tmp, &head->blocks->miscj, miscj) {
		kobject_unregister(&pos->kobj);
		list_del(&pos->miscj);
		kfree(pos);
	}

	kfree(per_cpu(threshold_banks, cpu)[bank]->blocks);
	per_cpu(threshold_banks, cpu)[bank]->blocks = NULL;
}

static void threshold_remove_bank(unsigned int cpu, int bank)
{
	int i = 0;
	struct threshold_bank *b;
	char name[32];

	b = per_cpu(threshold_banks, cpu)[bank];

	if (!b)
		return;

	if (!b->blocks)
		goto free_out;

	sprintf(name, "threshold_bank%i", bank);

#ifdef CONFIG_SMP
	/* sibling symlink */
	if (shared_bank[bank] && b->blocks->cpu != cpu) {
		sysfs_remove_link(&per_cpu(device_mce, cpu).kobj, name);
		per_cpu(threshold_banks, cpu)[bank] = NULL;
		return;
	}
#endif

	/* remove all sibling symlinks before unregistering */
	for_each_cpu_mask(i, b->cpus) {
		if (i == cpu)
			continue;

		sysfs_remove_link(&per_cpu(device_mce, i).kobj, name);
		per_cpu(threshold_banks, i)[bank] = NULL;
	}

	deallocate_threshold_block(cpu, bank);

free_out:
	kobject_unregister(&b->kobj);
	kfree(b);
	per_cpu(threshold_banks, cpu)[bank] = NULL;
}

static void threshold_remove_device(unsigned int cpu)
{
	unsigned int bank;

	for (bank = 0; bank < NR_BANKS; ++bank) {
		if (!(per_cpu(bank_map, cpu) & 1 << bank))
			continue;
		threshold_remove_bank(cpu, bank);
	}
}

/* get notified when a cpu comes on/off */
static int threshold_cpu_callback(struct notifier_block *nfb,
					    unsigned long action, void *hcpu)
{
	/* cpu was unsigned int to begin with */
	unsigned int cpu = (unsigned long)hcpu;

	if (cpu >= NR_CPUS)
		goto out;

	switch (action) {
	case CPU_ONLINE:
	case CPU_ONLINE_FROZEN:
		threshold_create_device(cpu);
		break;
	case CPU_DEAD:
	case CPU_DEAD_FROZEN:
		threshold_remove_device(cpu);
		break;
	default:
		break;
	}
      out:
	return NOTIFY_OK;
}

static struct notifier_block threshold_cpu_notifier = {
	.notifier_call = threshold_cpu_callback,
};

static __init int threshold_init_device(void)
{
	unsigned lcpu = 0;

	/* to hit CPUs online before the notifier is up */
	for_each_online_cpu(lcpu) {
		int err = threshold_create_device(lcpu);
		if (err)
			return err;
	}
	register_hotcpu_notifier(&threshold_cpu_notifier);
	return 0;
}

device_initcall(threshold_init_device);