summaryrefslogtreecommitdiffstats
path: root/arch/x86_64/kernel/mce.c
blob: 21e70625a4953e26361c060c432f1803393952c5 (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
/*
 * Machine check handler.
 * K8 parts Copyright 2002,2003 Andi Kleen, SuSE Labs.
 * Rest from unknown author(s). 
 * 2004 Andi Kleen. Rewrote most of it. 
 */

#include <linux/init.h>
#include <linux/types.h>
#include <linux/kernel.h>
#include <linux/sched.h>
#include <linux/string.h>
#include <linux/rcupdate.h>
#include <linux/kallsyms.h>
#include <linux/sysdev.h>
#include <linux/miscdevice.h>
#include <linux/fs.h>
#include <asm/processor.h> 
#include <asm/msr.h>
#include <asm/mce.h>
#include <asm/kdebug.h>
#include <asm/uaccess.h>

#define MISC_MCELOG_MINOR 227
#define NR_BANKS 5

static int mce_dont_init;

/* 0: always panic, 1: panic if deadlock possible, 2: try to avoid panic,
   3: never panic or exit (for testing only) */
static int tolerant = 1;
static int banks;
static unsigned long bank[NR_BANKS] = { [0 ... NR_BANKS-1] = ~0UL };
static unsigned long console_logged;
static int notify_user;
static int rip_msr;

/*
 * Lockless MCE logging infrastructure.
 * This avoids deadlocks on printk locks without having to break locks. Also
 * separate MCEs from kernel messages to avoid bogus bug reports.
 */

struct mce_log mcelog = { 
	MCE_LOG_SIGNATURE,
	MCE_LOG_LEN,
}; 

void mce_log(struct mce *mce)
{
	unsigned next, entry;
	mce->finished = 0;
	smp_wmb();
	for (;;) {
		entry = rcu_dereference(mcelog.next);
		/* When the buffer fills up discard new entries. Assume 
		   that the earlier errors are the more interesting. */
		if (entry >= MCE_LOG_LEN) {
			set_bit(MCE_OVERFLOW, &mcelog.flags);
			return;
		}
		/* Old left over entry. Skip. */
		if (mcelog.entry[entry].finished)
			continue;
		smp_rmb();
		next = entry + 1;
		if (cmpxchg(&mcelog.next, entry, next) == entry)
			break;
	}
	memcpy(mcelog.entry + entry, mce, sizeof(struct mce));
	smp_wmb();
	mcelog.entry[entry].finished = 1;
	smp_wmb();

	if (!test_and_set_bit(0, &console_logged))
		notify_user = 1;
}

static void print_mce(struct mce *m)
{
	printk(KERN_EMERG "\n"
	       KERN_EMERG
	       "CPU %d: Machine Check Exception: %16Lx Bank %d: %016Lx\n",
	       m->cpu, m->mcgstatus, m->bank, m->status);
	if (m->rip) {
		printk(KERN_EMERG 
		       "RIP%s %02x:<%016Lx> ",
		       !(m->mcgstatus & MCG_STATUS_EIPV) ? " !INEXACT!" : "",
		       m->cs, m->rip);
		if (m->cs == __KERNEL_CS)
			print_symbol("{%s}", m->rip);
		printk("\n");
	}
	printk(KERN_EMERG "TSC %Lx ", m->tsc); 
	if (m->addr)
		printk("ADDR %Lx ", m->addr);
	if (m->misc)
		printk("MISC %Lx ", m->misc); 	
	printk("\n");
}

static void mce_panic(char *msg, struct mce *backup, unsigned long start)
{ 
	int i;
	oops_begin();
	for (i = 0; i < MCE_LOG_LEN; i++) {
		unsigned long tsc = mcelog.entry[i].tsc;
		if (time_before(tsc, start))
			continue;
		print_mce(&mcelog.entry[i]); 
		if (backup && mcelog.entry[i].tsc == backup->tsc)
			backup = NULL;
	}
	if (backup)
		print_mce(backup);
	if (tolerant >= 3)
		printk("Fake panic: %s\n", msg);
	else
		panic(msg);
} 

static int mce_available(struct cpuinfo_x86 *c)
{
	return test_bit(X86_FEATURE_MCE, &c->x86_capability) &&
	       test_bit(X86_FEATURE_MCA, &c->x86_capability);
}

static inline void mce_get_rip(struct mce *m, struct pt_regs *regs)
{
	if (regs && (m->mcgstatus & MCG_STATUS_RIPV)) {
		m->rip = regs->rip;
		m->cs = regs->cs;
	} else {
		m->rip = 0;
		m->cs = 0;
	}
	if (rip_msr) {
		/* Assume the RIP in the MSR is exact. Is this true? */
		m->mcgstatus |= MCG_STATUS_EIPV;
		rdmsrl(rip_msr, m->rip);
		m->cs = 0;
	}
}

/* 
 * The actual machine check handler
 */

void do_machine_check(struct pt_regs * regs, long error_code)
{
	struct mce m, panicm;
	int nowayout = (tolerant < 1); 
	int kill_it = 0;
	u64 mcestart = 0;
	int i;
	int panicm_found = 0;

	if (regs)
		notify_die(DIE_NMI, "machine check", regs, error_code, 255, SIGKILL);
	if (!banks)
		return;

	memset(&m, 0, sizeof(struct mce));
	m.cpu = hard_smp_processor_id();
	rdmsrl(MSR_IA32_MCG_STATUS, m.mcgstatus);
	if (!(m.mcgstatus & MCG_STATUS_RIPV))
		kill_it = 1;
	
	rdtscll(mcestart);
	barrier();

	for (i = 0; i < banks; i++) {
		if (!bank[i])
			continue;
		
		m.misc = 0; 
		m.addr = 0;
		m.bank = i;
		m.tsc = 0;

		rdmsrl(MSR_IA32_MC0_STATUS + i*4, m.status);
		if ((m.status & MCI_STATUS_VAL) == 0)
			continue;

		if (m.status & MCI_STATUS_EN) {
			/* In theory _OVER could be a nowayout too, but
			   assume any overflowed errors were no fatal. */
			nowayout |= !!(m.status & MCI_STATUS_PCC);
			kill_it |= !!(m.status & MCI_STATUS_UC);
		}

		if (m.status & MCI_STATUS_MISCV)
			rdmsrl(MSR_IA32_MC0_MISC + i*4, m.misc);
		if (m.status & MCI_STATUS_ADDRV)
			rdmsrl(MSR_IA32_MC0_ADDR + i*4, m.addr);

		mce_get_rip(&m, regs);
		if (error_code != -1)
			rdtscll(m.tsc);
		wrmsrl(MSR_IA32_MC0_STATUS + i*4, 0);
		mce_log(&m);

		/* Did this bank cause the exception? */
		/* Assume that the bank with uncorrectable errors did it,
		   and that there is only a single one. */
		if ((m.status & MCI_STATUS_UC) && (m.status & MCI_STATUS_EN)) {
			panicm = m;
			panicm_found = 1;
		}

		tainted |= TAINT_MACHINE_CHECK;
	}

	/* Never do anything final in the polling timer */
	if (!regs)
		goto out;

	/* If we didn't find an uncorrectable error, pick
	   the last one (shouldn't happen, just being safe). */
	if (!panicm_found)
		panicm = m;
	if (nowayout)
		mce_panic("Machine check", &panicm, mcestart);
	if (kill_it) {
		int user_space = 0;

		if (m.mcgstatus & MCG_STATUS_RIPV)
			user_space = panicm.rip && (panicm.cs & 3);
		
		/* When the machine was in user space and the CPU didn't get
		   confused it's normally not necessary to panic, unless you 
		   are paranoid (tolerant == 0)

		   RED-PEN could be more tolerant for MCEs in idle,
		   but most likely they occur at boot anyways, where
		   it is best to just halt the machine. */
		if ((!user_space && (panic_on_oops || tolerant < 2)) ||
		    (unsigned)current->pid <= 1)
			mce_panic("Uncorrected machine check", &panicm, mcestart);

		/* do_exit takes an awful lot of locks and has as
		   slight risk of deadlocking. If you don't want that
		   don't set tolerant >= 2 */
		if (tolerant < 3)
			do_exit(SIGBUS);
	}

 out:
	/* Last thing done in the machine check exception to clear state. */
	wrmsrl(MSR_IA32_MCG_STATUS, 0);
}

/*
 * Periodic polling timer for "silent" machine check errors.
 */

static int check_interval = 5 * 60; /* 5 minutes */
static void mcheck_timer(void *data);
static DECLARE_WORK(mcheck_work, mcheck_timer, NULL);

static void mcheck_check_cpu(void *info)
{
	if (mce_available(&current_cpu_data))
		do_machine_check(NULL, 0);
}

static void mcheck_timer(void *data)
{
	on_each_cpu(mcheck_check_cpu, NULL, 1, 1);
	schedule_delayed_work(&mcheck_work, check_interval * HZ);

	/*
	 * It's ok to read stale data here for notify_user and
	 * console_logged as we'll simply get the updated versions
	 * on the next mcheck_timer execution and atomic operations
	 * on console_logged act as synchronization for notify_user
	 * writes.
	 */
	if (notify_user && console_logged) {
		notify_user = 0;
		clear_bit(0, &console_logged);
		printk(KERN_INFO "Machine check events logged\n");
	}
}


static __init int periodic_mcheck_init(void)
{ 
	if (check_interval)
		schedule_delayed_work(&mcheck_work, check_interval*HZ);
	return 0;
} 
__initcall(periodic_mcheck_init);


/* 
 * Initialize Machine Checks for a CPU.
 */
static void mce_init(void *dummy)
{
	u64 cap;
	int i;

	rdmsrl(MSR_IA32_MCG_CAP, cap);
	banks = cap & 0xff;
	if (banks > NR_BANKS) { 
		printk(KERN_INFO "MCE: warning: using only %d banks\n", banks);
		banks = NR_BANKS; 
	}
	/* Use accurate RIP reporting if available. */
	if ((cap & (1<<9)) && ((cap >> 16) & 0xff) >= 9)
		rip_msr = MSR_IA32_MCG_EIP;

	/* Log the machine checks left over from the previous reset.
	   This also clears all registers */
	do_machine_check(NULL, -1);

	set_in_cr4(X86_CR4_MCE);

	if (cap & MCG_CTL_P)
		wrmsr(MSR_IA32_MCG_CTL, 0xffffffff, 0xffffffff);

	for (i = 0; i < banks; i++) {
		wrmsrl(MSR_IA32_MC0_CTL+4*i, bank[i]);
		wrmsrl(MSR_IA32_MC0_STATUS+4*i, 0);
	}	
}

/* Add per CPU specific workarounds here */
static void __cpuinit mce_cpu_quirks(struct cpuinfo_x86 *c)
{ 
	/* This should be disabled by the BIOS, but isn't always */
	if (c->x86_vendor == X86_VENDOR_AMD && c->x86 == 15) {
		/* disable GART TBL walk error reporting, which trips off 
		   incorrectly with the IOMMU & 3ware & Cerberus. */
		clear_bit(10, &bank[4]);
	}
}			

static void __cpuinit mce_cpu_features(struct cpuinfo_x86 *c)
{
	switch (c->x86_vendor) {
	case X86_VENDOR_INTEL:
		mce_intel_feature_init(c);
		break;
	default:
		break;
	}
}

/* 
 * Called for each booted CPU to set up machine checks.
 * Must be called with preempt off. 
 */
void __cpuinit mcheck_init(struct cpuinfo_x86 *c)
{
	static cpumask_t mce_cpus __initdata = CPU_MASK_NONE;

	mce_cpu_quirks(c); 

	if (mce_dont_init ||
	    cpu_test_and_set(smp_processor_id(), mce_cpus) ||
	    !mce_available(c))
		return;

	mce_init(NULL);
	mce_cpu_features(c);
}

/*
 * Character device to read and clear the MCE log.
 */

static void collect_tscs(void *data) 
{ 
	unsigned long *cpu_tsc = (unsigned long *)data;
	rdtscll(cpu_tsc[smp_processor_id()]);
} 

static ssize_t mce_read(struct file *filp, char __user *ubuf, size_t usize, loff_t *off)
{
	unsigned long *cpu_tsc;
	static DECLARE_MUTEX(mce_read_sem);
	unsigned next;
	char __user *buf = ubuf;
	int i, err;

	cpu_tsc = kmalloc(NR_CPUS * sizeof(long), GFP_KERNEL);
	if (!cpu_tsc)
		return -ENOMEM;

	down(&mce_read_sem); 
	next = rcu_dereference(mcelog.next);

	/* Only supports full reads right now */
	if (*off != 0 || usize < MCE_LOG_LEN*sizeof(struct mce)) { 
		up(&mce_read_sem);
		kfree(cpu_tsc);
		return -EINVAL;
	}

	err = 0;
	for (i = 0; i < next; i++) {
		if (!mcelog.entry[i].finished)
			continue;
		smp_rmb();
		err |= copy_to_user(buf, mcelog.entry + i, sizeof(struct mce));
		buf += sizeof(struct mce); 
	} 

	memset(mcelog.entry, 0, next * sizeof(struct mce));
	mcelog.next = 0;

	synchronize_sched();

	/* Collect entries that were still getting written before the synchronize. */

	on_each_cpu(collect_tscs, cpu_tsc, 1, 1);
	for (i = next; i < MCE_LOG_LEN; i++) { 
		if (mcelog.entry[i].finished && 
		    mcelog.entry[i].tsc < cpu_tsc[mcelog.entry[i].cpu]) {  
			err |= copy_to_user(buf, mcelog.entry+i, sizeof(struct mce));
			smp_rmb();
			buf += sizeof(struct mce);
			memset(&mcelog.entry[i], 0, sizeof(struct mce));
		}
	} 	
	up(&mce_read_sem);
	kfree(cpu_tsc);
	return err ? -EFAULT : buf - ubuf; 
}

static int mce_ioctl(struct inode *i, struct file *f,unsigned int cmd, unsigned long arg)
{
	int __user *p = (int __user *)arg;
	if (!capable(CAP_SYS_ADMIN))
		return -EPERM; 
	switch (cmd) {
	case MCE_GET_RECORD_LEN: 
		return put_user(sizeof(struct mce), p);
	case MCE_GET_LOG_LEN:
		return put_user(MCE_LOG_LEN, p);		
	case MCE_GETCLEAR_FLAGS: {
		unsigned flags;
		do { 
			flags = mcelog.flags;
		} while (cmpxchg(&mcelog.flags, flags, 0) != flags); 
		return put_user(flags, p); 
	}
	default:
		return -ENOTTY; 
	} 
}

static struct file_operations mce_chrdev_ops = {
	.read = mce_read,
	.ioctl = mce_ioctl,
};

static struct miscdevice mce_log_device = {
	MISC_MCELOG_MINOR,
	"mcelog",
	&mce_chrdev_ops,
};

/* 
 * Old style boot options parsing. Only for compatibility. 
 */

static int __init mcheck_disable(char *str)
{
	mce_dont_init = 1;
	return 0;
}

/* mce=off disables machine check. Note you can reenable it later
   using sysfs */
static int __init mcheck_enable(char *str)
{
	if (!strcmp(str, "off"))
		mce_dont_init = 1;
	else
		printk("mce= argument %s ignored. Please use /sys", str); 
	return 0;
}

__setup("nomce", mcheck_disable);
__setup("mce", mcheck_enable);

/* 
 * Sysfs support
 */ 

/* On resume clear all MCE state. Don't want to see leftovers from the BIOS. */
static int mce_resume(struct sys_device *dev)
{
	on_each_cpu(mce_init, NULL, 1, 1);
	return 0;
}

/* Reinit MCEs after user configuration changes */
static void mce_restart(void) 
{ 
	if (check_interval)
		cancel_delayed_work(&mcheck_work);
	/* Timer race is harmless here */
	on_each_cpu(mce_init, NULL, 1, 1);       
	if (check_interval)
		schedule_delayed_work(&mcheck_work, check_interval*HZ);
}

static struct sysdev_class mce_sysclass = {
	.resume = mce_resume,
	set_kset_name("machinecheck"),
};

static struct sys_device device_mce = {
	.id	= 0,
	.cls	= &mce_sysclass,
};

/* Why are there no generic functions for this? */
#define ACCESSOR(name, var, start) \
	static ssize_t show_ ## name(struct sys_device *s, char *buf) { 	   	   \
		return sprintf(buf, "%lx\n", (unsigned long)var);		   \
	} 									   \
	static ssize_t set_ ## name(struct sys_device *s,const char *buf,size_t siz) { \
		char *end; 							   \
		unsigned long new = simple_strtoul(buf, &end, 0); 		   \
		if (end == buf) return -EINVAL;					   \
		var = new;							   \
		start; 								   \
		return end-buf;		     					   \
	}									   \
	static SYSDEV_ATTR(name, 0644, show_ ## name, set_ ## name);

ACCESSOR(bank0ctl,bank[0],mce_restart())
ACCESSOR(bank1ctl,bank[1],mce_restart())
ACCESSOR(bank2ctl,bank[2],mce_restart())
ACCESSOR(bank3ctl,bank[3],mce_restart())
ACCESSOR(bank4ctl,bank[4],mce_restart())
ACCESSOR(tolerant,tolerant,)
ACCESSOR(check_interval,check_interval,mce_restart())

static __cpuinit int mce_init_device(void)
{
	int err;
	if (!mce_available(&boot_cpu_data))
		return -EIO;
	err = sysdev_class_register(&mce_sysclass);
	if (!err)
		err = sysdev_register(&device_mce);
	if (!err) { 
		/* could create per CPU objects, but it is not worth it. */
		sysdev_create_file(&device_mce, &attr_bank0ctl); 
		sysdev_create_file(&device_mce, &attr_bank1ctl); 
		sysdev_create_file(&device_mce, &attr_bank2ctl); 
		sysdev_create_file(&device_mce, &attr_bank3ctl); 
		sysdev_create_file(&device_mce, &attr_bank4ctl); 
		sysdev_create_file(&device_mce, &attr_tolerant); 
		sysdev_create_file(&device_mce, &attr_check_interval);
	} 
	
	misc_register(&mce_log_device);
	return err;

}
device_initcall(mce_init_device);
OpenPOWER on IntegriCloud