printk.c 72.3 KB
Newer Older
Linus Torvalds's avatar
Linus Torvalds committed
1
2
3
4
5
6
7
8
9
10
11
12
/*
 *  linux/kernel/printk.c
 *
 *  Copyright (C) 1991, 1992  Linus Torvalds
 *
 * Modified to make sys_syslog() more flexible: added commands to
 * return the last 4k of kernel messages, regardless of whether
 * they've been read or not.  Added option to suppress kernel printk's
 * to the console.  Added hook for sending the console messages
 * elsewhere, in preparation for a serial line console (someday).
 * Ted Ts'o, 2/11/93.
 * Modified for sysctl support, 1/8/97, Chris Horn.
13
 * Fixed SMP synchronization, 08/08/99, Manfred Spraul
14
 *     manfred@colorfullife.com
Linus Torvalds's avatar
Linus Torvalds committed
15
 * Rewrote bits to get rid of console_lock
16
 *	01Mar01 Andrew Morton
Linus Torvalds's avatar
Linus Torvalds committed
17
18
19
20
21
22
23
24
 */

#include <linux/kernel.h>
#include <linux/mm.h>
#include <linux/tty.h>
#include <linux/tty_driver.h>
#include <linux/console.h>
#include <linux/init.h>
Randy Dunlap's avatar
Randy Dunlap committed
25
26
#include <linux/jiffies.h>
#include <linux/nmi.h>
Linus Torvalds's avatar
Linus Torvalds committed
27
#include <linux/module.h>
Jan Engelhardt's avatar
Jan Engelhardt committed
28
#include <linux/moduleparam.h>
Linus Torvalds's avatar
Linus Torvalds committed
29
30
31
32
33
#include <linux/interrupt.h>			/* For in_interrupt() */
#include <linux/delay.h>
#include <linux/smp.h>
#include <linux/security.h>
#include <linux/bootmem.h>
34
#include <linux/memblock.h>
35
#include <linux/aio.h>
Linus Torvalds's avatar
Linus Torvalds committed
36
#include <linux/syscalls.h>
37
#include <linux/kexec.h>
38
#include <linux/kdb.h>
39
#include <linux/ratelimit.h>
40
#include <linux/kmsg_dump.h>
41
#include <linux/syslog.h>
42
43
#include <linux/cpu.h>
#include <linux/notifier.h>
44
#include <linux/rculist.h>
45
#include <linux/poll.h>
46
#include <linux/irq_work.h>
47
#include <linux/utsname.h>
Linus Torvalds's avatar
Linus Torvalds committed
48
49
50

#include <asm/uaccess.h>

51
52
53
#define CREATE_TRACE_POINTS
#include <trace/events/printk.h>

Joe Perches's avatar
Joe Perches committed
54
55
#include "console_cmdline.h"

Linus Torvalds's avatar
Linus Torvalds committed
56
/* printk's without a loglevel use this.. */
57
#define DEFAULT_MESSAGE_LOGLEVEL CONFIG_DEFAULT_MESSAGE_LOGLEVEL
Linus Torvalds's avatar
Linus Torvalds committed
58
59
60
61
62
63
64
65
66
67
68
69
70

/* We show everything that is MORE important than this.. */
#define MINIMUM_CONSOLE_LOGLEVEL 1 /* Minimum loglevel we let people use */
#define DEFAULT_CONSOLE_LOGLEVEL 7 /* anything MORE serious than KERN_DEBUG */

int console_printk[4] = {
	DEFAULT_CONSOLE_LOGLEVEL,	/* console_loglevel */
	DEFAULT_MESSAGE_LOGLEVEL,	/* default_message_loglevel */
	MINIMUM_CONSOLE_LOGLEVEL,	/* minimum_console_loglevel */
	DEFAULT_CONSOLE_LOGLEVEL,	/* default_console_loglevel */
};

/*
71
 * Low level drivers may need that to know if they can schedule in
Linus Torvalds's avatar
Linus Torvalds committed
72
73
74
75
76
77
78
79
80
81
 * their unblank() callback or not. So let's export it.
 */
int oops_in_progress;
EXPORT_SYMBOL(oops_in_progress);

/*
 * console_sem protects the console_drivers list, and also
 * provides serialisation for access to the entire console
 * driver system.
 */
82
static DEFINE_SEMAPHORE(console_sem);
Linus Torvalds's avatar
Linus Torvalds committed
83
struct console *console_drivers;
Ingo Molnar's avatar
Ingo Molnar committed
84
85
EXPORT_SYMBOL_GPL(console_drivers);

86
87
88
89
90
91
#ifdef CONFIG_LOCKDEP
static struct lockdep_map console_lock_dep_map = {
	.name = "console_lock"
};
#endif

Linus Torvalds's avatar
Linus Torvalds committed
92
93
94
95
96
97
98
99
/*
 * This is used for debugging the mess that is the VT code by
 * keeping track if we have the console semaphore held. It's
 * definitely not the perfect debug tool (we don't know if _WE_
 * hold it are racing, but it helps tracking those weird code
 * path in the console code where we end up in places I want
 * locked without the console sempahore held
 */
100
static int console_locked, console_suspended;
Linus Torvalds's avatar
Linus Torvalds committed
101

102
103
104
105
106
/*
 * If exclusive_console is non-NULL then only this console is to be printed to.
 */
static struct console *exclusive_console;

Linus Torvalds's avatar
Linus Torvalds committed
107
108
109
110
111
112
113
/*
 *	Array of consoles built from command line options (console=)
 */

#define MAX_CMDLINECONSOLES 8

static struct console_cmdline console_cmdline[MAX_CMDLINECONSOLES];
Joe Perches's avatar
Joe Perches committed
114

Linus Torvalds's avatar
Linus Torvalds committed
115
116
static int selected_console = -1;
static int preferred_console = -1;
117
118
int console_set_on_cmdline;
EXPORT_SYMBOL(console_set_on_cmdline);
Linus Torvalds's avatar
Linus Torvalds committed
119
120
121
122

/* Flag: console code may call schedule() */
static int console_may_schedule;

123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
/*
 * The printk log buffer consists of a chain of concatenated variable
 * length records. Every record starts with a record header, containing
 * the overall length of the record.
 *
 * The heads to the first and last entry in the buffer, as well as the
 * sequence numbers of these both entries are maintained when messages
 * are stored..
 *
 * If the heads indicate available messages, the length in the header
 * tells the start next message. A length == 0 for the next message
 * indicates a wrap-around to the beginning of the buffer.
 *
 * Every record carries the monotonic timestamp in microseconds, as well as
 * the standard userspace syslog level and syslog facility. The usual
 * kernel messages use LOG_KERN; userspace-injected messages always carry
 * a matching syslog facility, by default LOG_USER. The origin of every
 * message can be reliably determined that way.
 *
 * The human readable log message directly follows the message header. The
 * length of the message text is stored in the header, the stored message
 * is not terminated.
 *
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
 * Optionally, a message can carry a dictionary of properties (key/value pairs),
 * to provide userspace with a machine-readable message context.
 *
 * Examples for well-defined, commonly used property names are:
 *   DEVICE=b12:8               device identifier
 *                                b12:8         block dev_t
 *                                c127:3        char dev_t
 *                                n8            netdev ifindex
 *                                +sound:card0  subsystem:devname
 *   SUBSYSTEM=pci              driver-core subsystem name
 *
 * Valid characters in property names are [a-zA-Z0-9.-_]. The plain text value
 * follows directly after a '=' character. Every property is terminated by
 * a '\0' character. The last property is not terminated.
 *
 * Example of a message structure:
 *   0000  ff 8f 00 00 00 00 00 00      monotonic time in nsec
 *   0008  34 00                        record is 52 bytes long
 *   000a        0b 00                  text is 11 bytes long
 *   000c              1f 00            dictionary is 23 bytes long
 *   000e                    03 00      LOG_KERN (facility) LOG_ERR (level)
 *   0010  69 74 27 73 20 61 20 6c      "it's a l"
 *         69 6e 65                     "ine"
 *   001b           44 45 56 49 43      "DEVIC"
 *         45 3d 62 38 3a 32 00 44      "E=b8:2\0D"
 *         52 49 56 45 52 3d 62 75      "RIVER=bu"
 *         67                           "g"
 *   0032     00 00 00                  padding to next message header
 *
 * The 'struct log' buffer header must never be directly exported to
 * userspace, it is a kernel-private implementation detail that might
 * need to be changed in the future, when the requirements change.
 *
 * /dev/kmsg exports the structured data in the following line format:
 *   "level,sequnum,timestamp;<message text>\n"
 *
 * The optional key/value pairs are attached as continuation lines starting
 * with a space character and terminated by a newline. All possible
 * non-prinatable characters are escaped in the "\xff" notation.
 *
 * Users of the export format should ignore possible additional values
 * separated by ',', and find the message after the ';' character.
188
189
 */

190
enum log_flags {
191
192
193
194
	LOG_NOCONS	= 1,	/* already flushed, do not print to console */
	LOG_NEWLINE	= 2,	/* text ended with a newline */
	LOG_PREFIX	= 4,	/* text started with a prefix */
	LOG_CONT	= 8,	/* text is a fragment of a continuation line */
195
196
};

197
198
199
200
201
struct log {
	u64 ts_nsec;		/* timestamp in nanoseconds */
	u16 len;		/* length of entire record */
	u16 text_len;		/* length of text buffer */
	u16 dict_len;		/* length of dictionary buffer */
202
203
204
	u8 facility;		/* syslog facility */
	u8 flags:5;		/* internal record flags */
	u8 level:3;		/* syslog level */
205
206
207
208
209
210
211
};

/*
 * The logbuf_lock protects kmsg buffer, indices, counters. It is also
 * used in interesting ways to provide interlocking in console_unlock();
 */
static DEFINE_RAW_SPINLOCK(logbuf_lock);
212

213
#ifdef CONFIG_PRINTK
214
DECLARE_WAIT_QUEUE_HEAD(log_wait);
215
216
217
/* the next printk record to read by syslog(READ) or /proc/kmsg */
static u64 syslog_seq;
static u32 syslog_idx;
218
static enum log_flags syslog_prev;
219
static size_t syslog_partial;
220
221
222
223
224
225
226
227
228

/* index and sequence number of the first record stored in the buffer */
static u64 log_first_seq;
static u32 log_first_idx;

/* index and sequence number of the next record to store in the buffer */
static u64 log_next_seq;
static u32 log_next_idx;

229
230
231
232
233
/* the next printk record to write to the console */
static u64 console_seq;
static u32 console_idx;
static enum log_flags console_prev;

234
235
236
237
/* the next printk record to read after the last 'clear' command */
static u64 clear_seq;
static u32 clear_idx;

238
239
#define PREFIX_MAX		32
#define LOG_LINE_MAX		1024 - PREFIX_MAX
240
241

/* record buffer */
242
#if defined(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS)
243
244
#define LOG_ALIGN 4
#else
245
#define LOG_ALIGN __alignof__(struct log)
246
#endif
247
#define __LOG_BUF_LEN (1 << CONFIG_LOG_BUF_SHIFT)
248
static char __log_buf[__LOG_BUF_LEN] __aligned(LOG_ALIGN);
249
250
251
252
253
static char *log_buf = __log_buf;
static u32 log_buf_len = __LOG_BUF_LEN;

/* cpu currently holding logbuf_lock */
static volatile unsigned int logbuf_cpu = UINT_MAX;
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

/* human readable text of the record */
static char *log_text(const struct log *msg)
{
	return (char *)msg + sizeof(struct log);
}

/* optional key/value pair dictionary attached to the record */
static char *log_dict(const struct log *msg)
{
	return (char *)msg + sizeof(struct log) + msg->text_len;
}

/* get record by index; idx must point to valid msg */
static struct log *log_from_idx(u32 idx)
{
	struct log *msg = (struct log *)(log_buf + idx);

	/*
	 * A length == 0 record is the end of buffer marker. Wrap around and
	 * read the message at the start of the buffer.
	 */
	if (!msg->len)
		return (struct log *)log_buf;
	return msg;
}

/* get next record; idx must point to valid msg */
static u32 log_next(u32 idx)
{
	struct log *msg = (struct log *)(log_buf + idx);

	/* length == 0 indicates the end of the buffer; wrap */
	/*
	 * A length == 0 record is the end of buffer marker. Wrap around and
	 * read the message at the start of the buffer as *this* one, and
	 * return the one after that.
	 */
	if (!msg->len) {
		msg = (struct log *)log_buf;
		return msg->len;
	}
	return idx + msg->len;
}

/* insert record into the buffer, discard old ones, update heads */
static void log_store(int facility, int level,
301
		      enum log_flags flags, u64 ts_nsec,
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
		      const char *dict, u16 dict_len,
		      const char *text, u16 text_len)
{
	struct log *msg;
	u32 size, pad_len;

	/* number of '\0' padding bytes to next message */
	size = sizeof(struct log) + text_len + dict_len;
	pad_len = (-size) & (LOG_ALIGN - 1);
	size += pad_len;

	while (log_first_seq < log_next_seq) {
		u32 free;

		if (log_next_idx > log_first_idx)
			free = max(log_buf_len - log_next_idx, log_first_idx);
		else
			free = log_first_idx - log_next_idx;

		if (free > size + sizeof(struct log))
			break;

		/* drop old messages until we have enough contiuous space */
		log_first_idx = log_next(log_first_idx);
		log_first_seq++;
	}

	if (log_next_idx + size + sizeof(struct log) >= log_buf_len) {
		/*
		 * This message + an additional empty header does not fit
		 * at the end of the buffer. Add an empty header with len == 0
		 * to signify a wrap around.
		 */
		memset(log_buf + log_next_idx, 0, sizeof(struct log));
		log_next_idx = 0;
	}

	/* fill message */
	msg = (struct log *)(log_buf + log_next_idx);
	memcpy(log_text(msg), text, text_len);
	msg->text_len = text_len;
	memcpy(log_dict(msg), dict, dict_len);
	msg->dict_len = dict_len;
345
346
347
348
349
350
351
	msg->facility = facility;
	msg->level = level & 7;
	msg->flags = flags & 0x1f;
	if (ts_nsec > 0)
		msg->ts_nsec = ts_nsec;
	else
		msg->ts_nsec = local_clock();
352
353
354
355
356
357
358
	memset(log_dict(msg) + dict_len, 0, pad_len);
	msg->len = sizeof(struct log) + text_len + dict_len + pad_len;

	/* insert message */
	log_next_idx += msg->len;
	log_next_seq++;
}
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
#ifdef CONFIG_SECURITY_DMESG_RESTRICT
int dmesg_restrict = 1;
#else
int dmesg_restrict;
#endif

static int syslog_action_restricted(int type)
{
	if (dmesg_restrict)
		return 1;
	/*
	 * Unless restricted, we allow "read all" and "get buffer size"
	 * for everybody.
	 */
	return type != SYSLOG_ACTION_READ_ALL &&
	       type != SYSLOG_ACTION_SIZE_BUFFER;
}

static int check_syslog_permissions(int type, bool from_file)
{
	/*
	 * If this is from /proc/kmsg and we've already opened it, then we've
	 * already done the capabilities checks at open time.
	 */
	if (from_file && type != SYSLOG_ACTION_OPEN)
		return 0;

	if (syslog_action_restricted(type)) {
		if (capable(CAP_SYSLOG))
			return 0;
		/*
		 * For historical reasons, accept CAP_SYS_ADMIN too, with
		 * a warning.
		 */
		if (capable(CAP_SYS_ADMIN)) {
			pr_warn_once("%s (%d): Attempt to access syslog with "
				     "CAP_SYS_ADMIN but no CAP_SYSLOG "
				     "(deprecated).\n",
				 current->comm, task_pid_nr(current));
			return 0;
		}
		return -EPERM;
	}
	return security_syslog(type);
}


407
408
409
410
/* /dev/kmsg - userspace message inject/listen interface */
struct devkmsg_user {
	u64 seq;
	u32 idx;
411
	enum log_flags prev;
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
	struct mutex lock;
	char buf[8192];
};

static ssize_t devkmsg_writev(struct kiocb *iocb, const struct iovec *iv,
			      unsigned long count, loff_t pos)
{
	char *buf, *line;
	int i;
	int level = default_message_loglevel;
	int facility = 1;	/* LOG_USER */
	size_t len = iov_length(iv, count);
	ssize_t ret = len;

	if (len > LOG_LINE_MAX)
		return -EINVAL;
	buf = kmalloc(len+1, GFP_KERNEL);
	if (buf == NULL)
		return -ENOMEM;

	line = buf;
	for (i = 0; i < count; i++) {
434
435
		if (copy_from_user(line, iv[i].iov_base, iv[i].iov_len)) {
			ret = -EFAULT;
436
			goto out;
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
		line += iv[i].iov_len;
	}

	/*
	 * Extract and skip the syslog prefix <[0-9]*>. Coming from userspace
	 * the decimal value represents 32bit, the lower 3 bit are the log
	 * level, the rest are the log facility.
	 *
	 * If no prefix or no userspace facility is specified, we
	 * enforce LOG_USER, to be able to reliably distinguish
	 * kernel-generated messages from userspace-injected ones.
	 */
	line = buf;
	if (line[0] == '<') {
		char *endp = NULL;

		i = simple_strtoul(line+1, &endp, 10);
		if (endp && endp[0] == '>') {
			level = i & 7;
			if (i >> 3)
				facility = i >> 3;
			endp++;
			len -= endp - line;
			line = endp;
		}
	}
	line[len] = '\0';

	printk_emit(facility, level, NULL, 0, "%s", line);
out:
	kfree(buf);
	return ret;
}

static ssize_t devkmsg_read(struct file *file, char __user *buf,
			    size_t count, loff_t *ppos)
{
	struct devkmsg_user *user = file->private_data;
	struct log *msg;
477
	u64 ts_usec;
478
	size_t i;
479
	char cont = '-';
480
481
482
483
484
485
	size_t len;
	ssize_t ret;

	if (!user)
		return -EBADF;

486
487
488
	ret = mutex_lock_interruptible(&user->lock);
	if (ret)
		return ret;
489
	raw_spin_lock_irq(&logbuf_lock);
490
491
492
	while (user->seq == log_next_seq) {
		if (file->f_flags & O_NONBLOCK) {
			ret = -EAGAIN;
493
			raw_spin_unlock_irq(&logbuf_lock);
494
495
496
			goto out;
		}

497
		raw_spin_unlock_irq(&logbuf_lock);
498
499
500
501
		ret = wait_event_interruptible(log_wait,
					       user->seq != log_next_seq);
		if (ret)
			goto out;
502
		raw_spin_lock_irq(&logbuf_lock);
503
504
505
506
507
508
509
	}

	if (user->seq < log_first_seq) {
		/* our last seen message is gone, return error and reset */
		user->idx = log_first_idx;
		user->seq = log_first_seq;
		ret = -EPIPE;
510
		raw_spin_unlock_irq(&logbuf_lock);
511
512
513
514
		goto out;
	}

	msg = log_from_idx(user->idx);
515
516
	ts_usec = msg->ts_nsec;
	do_div(ts_usec, 1000);
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535

	/*
	 * If we couldn't merge continuation line fragments during the print,
	 * export the stored flags to allow an optional external merge of the
	 * records. Merging the records isn't always neccessarily correct, like
	 * when we hit a race during printing. In most cases though, it produces
	 * better readable output. 'c' in the record flags mark the first
	 * fragment of a line, '+' the following.
	 */
	if (msg->flags & LOG_CONT && !(user->prev & LOG_CONT))
		cont = 'c';
	else if ((msg->flags & LOG_CONT) ||
		 ((user->prev & LOG_CONT) && !(msg->flags & LOG_PREFIX)))
		cont = '+';

	len = sprintf(user->buf, "%u,%llu,%llu,%c;",
		      (msg->facility << 3) | msg->level,
		      user->seq, ts_usec, cont);
	user->prev = msg->flags;
536
537
538

	/* escape non-printable characters */
	for (i = 0; i < msg->text_len; i++) {
539
		unsigned char c = log_text(msg)[i];
540

541
		if (c < ' ' || c >= 127 || c == '\\')
542
543
544
545
546
547
548
549
550
551
			len += sprintf(user->buf + len, "\\x%02x", c);
		else
			user->buf[len++] = c;
	}
	user->buf[len++] = '\n';

	if (msg->dict_len) {
		bool line = true;

		for (i = 0; i < msg->dict_len; i++) {
552
			unsigned char c = log_dict(msg)[i];
553
554
555
556
557
558
559
560
561
562
563
564

			if (line) {
				user->buf[len++] = ' ';
				line = false;
			}

			if (c == '\0') {
				user->buf[len++] = '\n';
				line = true;
				continue;
			}

565
			if (c < ' ' || c >= 127 || c == '\\') {
566
567
568
569
570
571
572
573
574
575
576
				len += sprintf(user->buf + len, "\\x%02x", c);
				continue;
			}

			user->buf[len++] = c;
		}
		user->buf[len++] = '\n';
	}

	user->idx = log_next(user->idx);
	user->seq++;
577
	raw_spin_unlock_irq(&logbuf_lock);
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

	if (len > count) {
		ret = -EINVAL;
		goto out;
	}

	if (copy_to_user(buf, user->buf, len)) {
		ret = -EFAULT;
		goto out;
	}
	ret = len;
out:
	mutex_unlock(&user->lock);
	return ret;
}

static loff_t devkmsg_llseek(struct file *file, loff_t offset, int whence)
{
	struct devkmsg_user *user = file->private_data;
	loff_t ret = 0;

	if (!user)
		return -EBADF;
	if (offset)
		return -ESPIPE;

604
	raw_spin_lock_irq(&logbuf_lock);
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
	switch (whence) {
	case SEEK_SET:
		/* the first record */
		user->idx = log_first_idx;
		user->seq = log_first_seq;
		break;
	case SEEK_DATA:
		/*
		 * The first record after the last SYSLOG_ACTION_CLEAR,
		 * like issued by 'dmesg -c'. Reading /dev/kmsg itself
		 * changes no global state, and does not clear anything.
		 */
		user->idx = clear_idx;
		user->seq = clear_seq;
		break;
	case SEEK_END:
		/* after the last record */
		user->idx = log_next_idx;
		user->seq = log_next_seq;
		break;
	default:
		ret = -EINVAL;
	}
628
	raw_spin_unlock_irq(&logbuf_lock);
629
630
631
632
633
634
635
636
637
638
639
640
641
	return ret;
}

static unsigned int devkmsg_poll(struct file *file, poll_table *wait)
{
	struct devkmsg_user *user = file->private_data;
	int ret = 0;

	if (!user)
		return POLLERR|POLLNVAL;

	poll_wait(file, &log_wait, wait);

642
	raw_spin_lock_irq(&logbuf_lock);
643
644
645
646
	if (user->seq < log_next_seq) {
		/* return error when data has vanished underneath us */
		if (user->seq < log_first_seq)
			ret = POLLIN|POLLRDNORM|POLLERR|POLLPRI;
647
648
		else
			ret = POLLIN|POLLRDNORM;
649
	}
650
	raw_spin_unlock_irq(&logbuf_lock);
651
652
653
654
655
656
657
658
659
660
661
662
663

	return ret;
}

static int devkmsg_open(struct inode *inode, struct file *file)
{
	struct devkmsg_user *user;
	int err;

	/* write-only does not need any file context */
	if ((file->f_flags & O_ACCMODE) == O_WRONLY)
		return 0;

664
665
	err = check_syslog_permissions(SYSLOG_ACTION_READ_ALL,
				       SYSLOG_FROM_READER);
666
667
668
669
670
671
672
673
674
	if (err)
		return err;

	user = kmalloc(sizeof(struct devkmsg_user), GFP_KERNEL);
	if (!user)
		return -ENOMEM;

	mutex_init(&user->lock);

675
	raw_spin_lock_irq(&logbuf_lock);
676
677
	user->idx = log_first_idx;
	user->seq = log_first_seq;
678
	raw_spin_unlock_irq(&logbuf_lock);
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

	file->private_data = user;
	return 0;
}

static int devkmsg_release(struct inode *inode, struct file *file)
{
	struct devkmsg_user *user = file->private_data;

	if (!user)
		return 0;

	mutex_destroy(&user->lock);
	kfree(user);
	return 0;
}

const struct file_operations kmsg_fops = {
	.open = devkmsg_open,
	.read = devkmsg_read,
	.aio_write = devkmsg_writev,
	.llseek = devkmsg_llseek,
	.poll = devkmsg_poll,
	.release = devkmsg_release,
};

705
706
707
708
709
710
711
712
713
714
715
716
717
#ifdef CONFIG_KEXEC
/*
 * This appends the listed symbols to /proc/vmcoreinfo
 *
 * /proc/vmcoreinfo is used by various utiilties, like crash and makedumpfile to
 * obtain access to symbols that are otherwise very difficult to locate.  These
 * symbols are specifically used so that utilities can access and extract the
 * dmesg log from a vmcore file after a crash.
 */
void log_buf_kexec_setup(void)
{
	VMCOREINFO_SYMBOL(log_buf);
	VMCOREINFO_SYMBOL(log_buf_len);
718
719
	VMCOREINFO_SYMBOL(log_first_idx);
	VMCOREINFO_SYMBOL(log_next_idx);
720
721
722
723
724
725
726
727
728
	/*
	 * Export struct log size and field offsets. User space tools can
	 * parse it and detect any changes to structure down the line.
	 */
	VMCOREINFO_STRUCT_SIZE(log);
	VMCOREINFO_OFFSET(log, ts_nsec);
	VMCOREINFO_OFFSET(log, len);
	VMCOREINFO_OFFSET(log, text_len);
	VMCOREINFO_OFFSET(log, dict_len);
729
730
731
}
#endif

732
733
734
735
/* requested log_buf_len from kernel cmdline */
static unsigned long __initdata new_log_buf_len;

/* save requested log_buf_len since it's too early to process it */
Linus Torvalds's avatar
Linus Torvalds committed
736
737
static int __init log_buf_len_setup(char *str)
{
738
	unsigned size = memparse(str, &str);
Linus Torvalds's avatar
Linus Torvalds committed
739
740
741

	if (size)
		size = roundup_pow_of_two(size);
742
743
744
745
	if (size > log_buf_len)
		new_log_buf_len = size;

	return 0;
Linus Torvalds's avatar
Linus Torvalds committed
746
}
747
748
749
750
751
752
753
754
755
756
early_param("log_buf_len", log_buf_len_setup);

void __init setup_log_buf(int early)
{
	unsigned long flags;
	char *new_log_buf;
	int free;

	if (!new_log_buf_len)
		return;
Linus Torvalds's avatar
Linus Torvalds committed
757

758
759
760
761
	if (early) {
		unsigned long mem;

		mem = memblock_alloc(new_log_buf_len, PAGE_SIZE);
Tejun Heo's avatar
Tejun Heo committed
762
		if (!mem)
763
764
765
766
767
768
769
770
771
772
773
774
			return;
		new_log_buf = __va(mem);
	} else {
		new_log_buf = alloc_bootmem_nopanic(new_log_buf_len);
	}

	if (unlikely(!new_log_buf)) {
		pr_err("log_buf_len: %ld bytes not available\n",
			new_log_buf_len);
		return;
	}

775
	raw_spin_lock_irqsave(&logbuf_lock, flags);
776
777
778
	log_buf_len = new_log_buf_len;
	log_buf = new_log_buf;
	new_log_buf_len = 0;
779
780
	free = __LOG_BUF_LEN - log_next_idx;
	memcpy(log_buf, __log_buf, __LOG_BUF_LEN);
781
	raw_spin_unlock_irqrestore(&logbuf_lock, flags);
782
783
784
785
786

	pr_info("log_buf_len: %d\n", log_buf_len);
	pr_info("early log buf free: %d(%d%%)\n",
		free, (free * 100) / __LOG_BUF_LEN);
}
Linus Torvalds's avatar
Linus Torvalds committed
787

788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
static bool __read_mostly ignore_loglevel;

static int __init ignore_loglevel_setup(char *str)
{
	ignore_loglevel = 1;
	printk(KERN_INFO "debug: ignoring loglevel setting.\n");

	return 0;
}

early_param("ignore_loglevel", ignore_loglevel_setup);
module_param(ignore_loglevel, bool, S_IRUGO | S_IWUSR);
MODULE_PARM_DESC(ignore_loglevel, "ignore loglevel setting, to"
	"print all kernel messages to the console.");

Randy Dunlap's avatar
Randy Dunlap committed
803
804
#ifdef CONFIG_BOOT_PRINTK_DELAY

805
static int boot_delay; /* msecs delay after each printk during bootup */
806
static unsigned long long loops_per_msec;	/* based on boot_delay */
Randy Dunlap's avatar
Randy Dunlap committed
807
808
809
810
811
812
813
814
815
816
817
818

static int __init boot_delay_setup(char *str)
{
	unsigned long lpj;

	lpj = preset_lpj ? preset_lpj : 1000000;	/* some guess */
	loops_per_msec = (unsigned long long)lpj / 1000 * HZ;

	get_option(&str, &boot_delay);
	if (boot_delay > 10 * 1000)
		boot_delay = 0;

819
820
821
	pr_debug("boot_delay: %u, preset_lpj: %ld, lpj: %lu, "
		"HZ: %d, loops_per_msec: %llu\n",
		boot_delay, preset_lpj, lpj, HZ, loops_per_msec);
Randy Dunlap's avatar
Randy Dunlap committed
822
823
824
825
	return 1;
}
__setup("boot_delay=", boot_delay_setup);

826
static void boot_delay_msec(int level)
Randy Dunlap's avatar
Randy Dunlap committed
827
828
829
830
{
	unsigned long long k;
	unsigned long timeout;

831
832
	if ((boot_delay == 0 || system_state != SYSTEM_BOOTING)
		|| (level >= console_loglevel && !ignore_loglevel)) {
Randy Dunlap's avatar
Randy Dunlap committed
833
		return;
834
	}
Randy Dunlap's avatar
Randy Dunlap committed
835

836
	k = (unsigned long long)loops_per_msec * boot_delay;
Randy Dunlap's avatar
Randy Dunlap committed
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852

	timeout = jiffies + msecs_to_jiffies(boot_delay);
	while (k) {
		k--;
		cpu_relax();
		/*
		 * use (volatile) jiffies to prevent
		 * compiler reduction; loop termination via jiffies
		 * is secondary and may or may not happen.
		 */
		if (time_after(jiffies, timeout))
			break;
		touch_nmi_watchdog();
	}
}
#else
853
static inline void boot_delay_msec(int level)
Randy Dunlap's avatar
Randy Dunlap committed
854
855
856
857
{
}
#endif

858
859
860
861
862
863
864
#if defined(CONFIG_PRINTK_TIME)
static bool printk_time = 1;
#else
static bool printk_time;
#endif
module_param_named(time, printk_time, bool, S_IRUGO | S_IWUSR);

865
866
867
868
869
870
871
static size_t print_time(u64 ts, char *buf)
{
	unsigned long rem_nsec;

	if (!printk_time)
		return 0;

872
873
	rem_nsec = do_div(ts, 1000000000);

874
	if (!buf)
875
		return snprintf(NULL, 0, "[%5lu.000000] ", (unsigned long)ts);
876
877
878
879
880

	return sprintf(buf, "[%5lu.%06lu] ",
		       (unsigned long)ts, rem_nsec / 1000);
}

881
static size_t print_prefix(const struct log *msg, bool syslog, char *buf)
882
{
883
	size_t len = 0;
884
	unsigned int prefix = (msg->facility << 3) | msg->level;
885

886
887
	if (syslog) {
		if (buf) {
888
			len += sprintf(buf, "<%u>", prefix);
889
890
		} else {
			len += 3;
891
892
893
894
895
			if (prefix > 999)
				len += 3;
			else if (prefix > 99)
				len += 2;
			else if (prefix > 9)
896
897
898
				len++;
		}
	}
899

900
	len += print_time(msg->ts_nsec, buf ? buf + len : NULL);
901
	return len;
902
903
}

904
905
static size_t msg_print_text(const struct log *msg, enum log_flags prev,
			     bool syslog, char *buf, size_t size)
906
{
907
908
	const char *text = log_text(msg);
	size_t text_size = msg->text_len;
909
910
	bool prefix = true;
	bool newline = true;
911
912
	size_t len = 0;

913
914
915
916
917
918
919
920
921
922
923
	if ((prev & LOG_CONT) && !(msg->flags & LOG_PREFIX))
		prefix = false;

	if (msg->flags & LOG_CONT) {
		if ((prev & LOG_CONT) && !(prev & LOG_NEWLINE))
			prefix = false;

		if (!(msg->flags & LOG_NEWLINE))
			newline = false;
	}

924
925
926
927
928
929
930
931
932
933
934
	do {
		const char *next = memchr(text, '\n', text_size);
		size_t text_len;

		if (next) {
			text_len = next - text;
			next++;
			text_size -= next - text;
		} else {
			text_len = text_size;
		}
935

936
937
		if (buf) {
			if (print_prefix(msg, syslog, NULL) +
938
			    text_len + 1 >= size - len)
939
				break;
940

941
942
			if (prefix)
				len += print_prefix(msg, syslog, buf + len);
943
944
			memcpy(buf + len, text, text_len);
			len += text_len;
945
946
			if (next || newline)
				buf[len++] = '\n';
947
948
		} else {
			/* SYSLOG_ACTION_* buffer size only calculation */
949
950
951
952
953
			if (prefix)
				len += print_prefix(msg, syslog, NULL);
			len += text_len;
			if (next || newline)
				len++;
954
		}
955

956
		prefix = true;
957
958
		text = next;
	} while (text);
959
960
961
962
963
964
965

	return len;
}

static int syslog_print(char __user *buf, int size)
{
	char *text;
966
	struct log *msg;
967
	int len = 0;
968

969
	text = kmalloc(LOG_LINE_MAX + PREFIX_MAX, GFP_KERNEL);
970
971
972
	if (!text)
		return -ENOMEM;

973
974
	while (size > 0) {
		size_t n;
975
		size_t skip;
976
977
978
979
980
981

		raw_spin_lock_irq(&logbuf_lock);
		if (syslog_seq < log_first_seq) {
			/* messages are gone, move to first one */
			syslog_seq = log_first_seq;
			syslog_idx = log_first_idx;
982
			syslog_prev = 0;
983
			syslog_partial = 0;
984
985
986
987
988
		}
		if (syslog_seq == log_next_seq) {
			raw_spin_unlock_irq(&logbuf_lock);
			break;
		}
989
990

		skip = syslog_partial;
991
		msg = log_from_idx(syslog_idx);
992
993
		n = msg_print_text(msg, syslog_prev, true, text,
				   LOG_LINE_MAX + PREFIX_MAX);
994
995
		if (n - syslog_partial <= size) {
			/* message fits into buffer, move forward */
996
997
			syslog_idx = log_next(syslog_idx);
			syslog_seq++;
998
			syslog_prev = msg->flags;
999
1000
1001
1002
1003
1004
			n -= syslog_partial;
			syslog_partial = 0;
		} else if (!len){
			/* partial read(), remember position */
			n = size;
			syslog_partial += n;
1005
1006
1007
1008
1009
1010
1011
		} else
			n = 0;
		raw_spin_unlock_irq(&logbuf_lock);

		if (!n)
			break;

1012
		if (copy_to_user(buf, text + skip, n)) {
1013
1014
1015
1016
			if (!len)
				len = -EFAULT;
			break;
		}
1017
1018
1019
1020

		len += n;
		size -= n;
		buf += n;
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
	}

	kfree(text);
	return len;
}

static int syslog_print_all(char __user *buf, int size, bool clear)
{
	char *text;
	int len = 0;

1032
	text = kmalloc(LOG_LINE_MAX + PREFIX_MAX, GFP_KERNEL);
1033
1034
1035
1036
1037
1038
1039
1040
	if (!text)
		return -ENOMEM;

	raw_spin_lock_irq(&logbuf_lock);
	if (buf) {
		u64 next_seq;
		u64 seq;
		u32 idx;
1041
		enum log_flags prev;
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051

		if (clear_seq < log_first_seq) {
			/* messages are gone, move to first available one */
			clear_seq = log_first_seq;
			clear_idx = log_first_idx;
		}

		/*
		 * Find first record that fits, including all following records,
		 * into the user-provided buffer for this dump.
1052
		 */
1053
1054
		seq = clear_seq;
		idx = clear_idx;
1055
		prev = 0;
1056
		while (seq < log_next_seq) {
1057
1058
			struct log *msg = log_from_idx(idx);

1059
			len += msg_print_text(msg, prev, true, NULL, 0);
1060
			prev = msg->flags;
1061
1062
1063
			idx = log_next(idx);
			seq++;
		}
1064
1065

		/* move first record forward until length fits into the buffer */
1066
1067
		seq = clear_seq;
		idx = clear_idx;
1068
		prev = 0;
1069
		while (len > size && seq < log_next_seq) {
1070
1071
			struct log *msg = log_from_idx(idx);

1072
			len -= msg_print_text(msg, prev, true, NULL, 0);
1073
			prev = msg->flags;
1074
1075
1076
1077
			idx = log_next(idx);
			seq++;
		}

1078
		/* last message fitting into this dump */
1079
1080
1081
		next_seq = log_next_seq;

		len = 0;
1082
		prev = 0;
1083
		while (len >= 0 && seq < next_seq) {
1084
			struct log *msg = log_from_idx(idx);
1085
1086
			int textlen;

1087
1088
			textlen = msg_print_text(msg, prev, true, text,
						 LOG_LINE_MAX + PREFIX_MAX);
1089
1090
1091
1092
1093
1094
			if (textlen < 0) {
				len = textlen;
				break;
			}
			idx = log_next(idx);
			seq++;
1095
			prev = msg->flags;
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107

			raw_spin_unlock_irq(&logbuf_lock);
			if (copy_to_user(buf + len, text, textlen))
				len = -EFAULT;
			else
				len += textlen;
			raw_spin_lock_irq(&logbuf_lock);

			if (seq < log_first_seq) {
				/* messages are gone, move to next one */
				seq = log_first_seq;
				idx = log_first_idx;
1108
				prev = 0;
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
			}
		}
	}

	if (clear) {
		clear_seq = log_next_seq;
		clear_idx = log_next_idx;
	}
	raw_spin_unlock_irq(&logbuf_lock);

	kfree(text);
	return len;
}

1123
int do_syslog(int type, char __user *buf, int len, bool from_file)
Linus Torvalds's avatar
Linus Torvalds committed
1124
{
1125
1126
	bool clear = false;
	static int saved_console_loglevel = -1;
1127
	int error;
Linus Torvalds's avatar
Linus Torvalds committed
1128

1129
1130
1131
	error = check_syslog_permissions(type, from_file);
	if (error)
		goto out;
1132
1133

	error = security_syslog(type);
Linus Torvalds's avatar
Linus Torvalds committed
1134
1135
1136
1137
	if (error)
		return error;

	switch (type) {
1138
	case SYSLOG_ACTION_CLOSE:	/* Close log */
Linus Torvalds's avatar
Linus Torvalds committed
1139
		break;
1140
	case SYSLOG_ACTION_OPEN:	/* Open log */
Linus Torvalds's avatar
Linus Torvalds committed
1141
		break;
1142
	case SYSLOG_ACTION_READ:	/* Read from log */
Linus Torvalds's avatar
Linus Torvalds committed
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
		error = -EINVAL;
		if (!buf || len < 0)
			goto out;
		error = 0;
		if (!len)
			goto out;
		if (!access_ok(VERIFY_WRITE, buf, len)) {
			error = -EFAULT;
			goto out;
		}
1153
		error = wait_event_interruptible(log_wait,
1154
						 syslog_seq != log_next_seq);
1155
		if (error)
Linus Torvalds's avatar
Linus Torvalds committed
1156
			goto out;
1157
		error = syslog_print(buf, len);
Linus Torvalds's avatar
Linus Torvalds committed
1158
		break;
1159
1160
	/* Read/clear last kernel messages */
	case SYSLOG_ACTION_READ_CLEAR:
1161
		clear = true;
Linus Torvalds's avatar
Linus Torvalds committed
1162
		/* FALL THRU */
1163
1164
	/* Read last kernel messages */
	case SYSLOG_ACTION_READ_ALL:
Linus Torvalds's avatar
Linus Torvalds committed
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
		error = -EINVAL;
		if (!buf || len < 0)
			goto out;
		error = 0;
		if (!len)
			goto out;
		if (!access_ok(VERIFY_WRITE, buf, len)) {
			error = -EFAULT;
			goto out;
		}
1175
		error = syslog_print_all(buf, len, clear);
Linus Torvalds's avatar
Linus Torvalds committed
1176
		break;
1177
1178
	/* Clear ring buffer */
	case SYSLOG_ACTION_CLEAR:
1179
		syslog_print_all(NULL, 0, true);
1180
		break;
1181
1182
	/* Disable logging to console */
	case SYSLOG_ACTION_CONSOLE_OFF:
1183
1184
		if (saved_console_loglevel == -1)
			saved_console_loglevel = console_loglevel;
Linus Torvalds's avatar
Linus Torvalds committed
1185
1186
		console_loglevel = minimum_console_loglevel;
		break;
1187
1188
	/* Enable logging to console */
	case SYSLOG_ACTION_CONSOLE_ON:
1189
1190
1191
1192
		if (saved_console_loglevel != -1) {
			console_loglevel = saved_console_loglevel;
			saved_console_loglevel = -1;
		}
Linus Torvalds's avatar
Linus Torvalds committed
1193
		break;
1194
1195
	/* Set level of messages printed to console */
	case SYSLOG_ACTION_CONSOLE_LEVEL:
Linus Torvalds's avatar
Linus Torvalds committed
1196
1197
1198
1199
1200
1201
		error = -EINVAL;
		if (len < 1 || len > 8)
			goto out;
		if (len < minimum_console_loglevel)
			len = minimum_console_loglevel;
		console_loglevel = len;
1202
1203
		/* Implicitly re-enable logging to console */
		saved_console_loglevel = -1;
Linus Torvalds's avatar
Linus Torvalds committed
1204
1205
		error = 0;
		break;
1206
1207
	/* Number of chars in the log buffer */
	case SYSLOG_ACTION_SIZE_UNREAD:
1208
1209
1210
1211
1212
		raw_spin_lock_irq(&logbuf_lock);
		if (syslog_seq < log_first_seq) {
			/* messages are gone, move to first one */
			syslog_seq = log_first_seq;
			syslog_idx = log_first_idx;
1213
			syslog_prev = 0;
1214
			syslog_partial = 0;
1215
1216
1217
1218
1219
1220
1221
1222
1223
		}
		if (from_file) {
			/*
			 * Short-cut for poll(/"proc/kmsg") which simply checks
			 * for pending data, not the size; return the count of
			 * records, not the length.
			 */
			error = log_next_idx - syslog_idx;
		} else {
1224
1225
1226
			u64 seq = syslog_seq;
			u32 idx = syslog_idx;
			enum log_flags prev = syslog_prev;
1227
1228
1229

			error = 0;
			while (seq < log_next_seq) {
1230
1231
				struct log *msg = log_from_idx(idx);

1232
				error += msg_print_text(msg, prev, true, NULL, 0);
1233
1234
				idx = log_next(idx);
				seq++;
1235
				prev = msg->flags;
1236
			}
1237
			error -= syslog_partial;
1238
1239
		}
		raw_spin_unlock_irq(&logbuf_lock);
Linus Torvalds's avatar
Linus Torvalds committed
1240
		break;
1241
1242
	/* Size of the log buffer */
	case SYSLOG_ACTION_SIZE_BUFFER:
Linus Torvalds's avatar
Linus Torvalds committed
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
		error = log_buf_len;
		break;
	default:
		error = -EINVAL;
		break;
	}
out:
	return error;
}

1253
SYSCALL_DEFINE3(syslog, int, type, char __user *, buf, int, len)
Linus Torvalds's avatar
Linus Torvalds committed
1254
{
1255
	return do_syslog(type, buf, len, SYSLOG_FROM_READER);
Linus Torvalds's avatar
Linus Torvalds committed
1256
1257
1258
1259
1260
}

/*
 * Call the console drivers, asking them to write out
 * log_buf[start] to log_buf[end - 1].
1261
 * The console_lock must be held.
Linus Torvalds's avatar
Linus Torvalds committed
1262
 */
1263
static void call_console_drivers(int level, const char *text, size_t len)
Linus Torvalds's avatar
Linus Torvalds committed
1264
{
1265
	struct console *con;
Linus Torvalds's avatar
Linus Torvalds committed
1266

1267
	trace_console(text, len);
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285

	if (level >= console_loglevel && !ignore_loglevel)
		return;
	if (!console_drivers)
		return;

	for_each_console(con) {
		if (exclusive_console && con != exclusive_console)
			continue;
		if (!(con->flags & CON_ENABLED))
			continue;
		if (!con->write)
			continue;
		if (!cpu_online(smp_processor_id()) &&
		    !(con->flags & CON_ANYTIME))
			continue;
		con->write(con, text, len);
	}
Linus Torvalds's avatar
Linus Torvalds committed
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
}

/*
 * Zap console related locks when oopsing. Only zap at most once
 * every 10 seconds, to leave time for slow consoles to print a
 * full oops.
 */
static void zap_locks(void)
{
	static unsigned long oops_timestamp;

	if (time_after_eq(jiffies, oops_timestamp) &&
1298
			!time_after(jiffies, oops_timestamp + 30 * HZ))
Linus Torvalds's avatar
Linus Torvalds committed
1299
1300
1301
1302
		return;

	oops_timestamp = jiffies;

1303
	debug_locks_off();
Linus Torvalds's avatar
Linus Torvalds committed
1304
	/* If a crash is occurring, make sure we can't deadlock */
1305
	raw_spin_lock_init(&logbuf_lock);
Linus Torvalds's avatar
Linus Torvalds committed
1306
	/* And make sure that we print immediately */
1307
	sema_init(&console_sem, 1);
Linus Torvalds's avatar
Linus Torvalds committed
1308
1309
}

1310
1311
1312
1313
1314
/* Check if we have any console registered that can be called early in boot. */
static int have_callable_console(void)
{
	struct console *con;

1315
	for_each_console(con)
1316
1317
1318
1319
1320
1321
		if (con->flags & CON_ANYTIME)
			return 1;

	return 0;
}

1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
/*
 * Can we actually use the console at this time on this cpu?
 *
 * Console drivers may assume that per-cpu resources have
 * been allocated. So unless they're explicitly marked as
 * being able to cope (CON_ANYTIME) don't call them until
 * this CPU is officially up.
 */
static inline int can_use_console(unsigned int cpu)
{
	return cpu_online(cpu) || have_callable_console();
}

/*
 * Try to get console ownership to actually show the kernel
 * messages from a 'printk'. Return true (and with the
1338
 * console_lock held, and 'console_locked' set) if it
1339
1340
1341
1342
1343
1344
 * is successful, false otherwise.
 *
 * This gets called with the 'logbuf_lock' spinlock held and
 * interrupts disabled. It should return with 'lockbuf_lock'
 * released but interrupts still disabled.
 */
1345
static int console_trylock_for_printk(unsigned int cpu)
1346
	__releases(&logbuf_lock)
1347
{
1348
	int retval = 0, wake = 0;
1349

1350
	if (console_trylock()) {
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
		retval = 1;

		/*
		 * If we can't use the console, we need to release
		 * the console semaphore by hand to avoid flushing
		 * the buffer. We need to hold the console semaphore
		 * in order to do this test safely.
		 */
		if (!can_use_console(cpu)) {
			console_locked = 0;
1361
			wake = 1;
1362
1363
1364
			retval = 0;
		}
	}
1365
	logbuf_cpu = UINT_MAX;
1366
	raw_spin_unlock(&logbuf_lock);
1367
1368
	if (wake)
		up(&console_sem);
1369
1370
	return retval;
}
1371

1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
int printk_delay_msec __read_mostly;

static inline void printk_delay(void)
{
	if (unlikely(printk_delay_msec)) {
		int m = printk_delay_msec;

		while (m--) {
			mdelay(1);
			touch_nmi_watchdog();
		}
	}
}

1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
/*
 * Continuation lines are buffered, and not committed to the record buffer
 * until the line is complete, or a race forces it. The line fragments
 * though, are printed immediately to the consoles to ensure everything has
 * reached the console in case of a kernel crash.
 */
static struct cont {
	char buf[LOG_LINE_MAX];
	size_t len;			/* length == 0 means unused buffer */
	size_t cons;			/* bytes written to console */
	struct task_struct *owner;	/* task of first print*/
	u64 ts_nsec;			/* time of first print */
	u8 level;			/* log level of first message */
	u8 facility;			/* log level of first message */
1400
	enum log_flags flags;		/* prefix, newline flags */
1401
1402
1403
	bool flushed:1;			/* buffer sealed and committed */
} cont;

1404
static void cont_flush(enum log_flags flags)
1405
1406
1407
1408
1409
1410
{
	if (cont.flushed)
		return;
	if (cont.len == 0)
		return;

1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
	if (cont.cons) {
		/*
		 * If a fragment of this line was directly flushed to the
		 * console; wait for the console to pick up the rest of the
		 * line. LOG_NOCONS suppresses a duplicated output.
		 */
		log_store(cont.facility, cont.level, flags | LOG_NOCONS,
			  cont.ts_nsec, NULL, 0, cont.buf, cont.len);
		cont.flags = flags;
		cont.flushed = true;
	} else {
		/*
		 * If no fragment of this line ever reached the console,
		 * just submit it to the store and free the buffer.
		 */
		log_store(cont.facility, cont.level, flags, 0,
			  NULL, 0, cont.buf, cont.len);
		cont.len = 0;
	}
1430
1431
1432
1433
1434
1435
1436
1437
}

static bool cont_add(int facility, int level, const char *text, size_t len)
{
	if (cont.len && cont.flushed)
		return false;

	if (cont.len + len > sizeof(cont.buf)) {
1438
1439
		/* the line gets too long, split it up in separate records */
		cont_flush(LOG_CONT);
1440
1441
1442
1443
1444
1445
1446
1447
		return false;
	}

	if (!cont.len) {
		cont.facility = facility;
		cont.level = level;
		cont.owner = current;
		cont.ts_nsec = local_clock();
1448
		cont.flags = 0;
1449
1450
1451
1452
1453
1454
		cont.cons = 0;
		cont.flushed = false;
	}

	memcpy(cont.buf + cont.len, text, len);
	cont.len += len;
1455
1456
1457
1458

	if (cont.len > (sizeof(cont.buf) * 80) / 100)
		cont_flush(LOG_CONT);

1459
1460
1461
1462
1463
1464
1465
1466
	return true;
}

static size_t cont_print_text(char *text, size_t size)
{
	size_t textlen = 0;
	size_t len;

1467
	if