You need to sign in or sign up before continuing.
printk.c 64.7 KB
Newer Older
L
Linus Torvalds 已提交
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.
J
Jesper Juhl 已提交
13
 * Fixed SMP synchronization, 08/08/99, Manfred Spraul
14
 *     manfred@colorfullife.com
L
Linus Torvalds 已提交
15
 * Rewrote bits to get rid of console_lock
16
 *	01Mar01 Andrew Morton
L
Linus Torvalds 已提交
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>
R
Randy Dunlap 已提交
25 26
#include <linux/jiffies.h>
#include <linux/nmi.h>
L
Linus Torvalds 已提交
27
#include <linux/module.h>
J
Jan Engelhardt 已提交
28
#include <linux/moduleparam.h>
L
Linus Torvalds 已提交
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>
L
Linus Torvalds 已提交
35
#include <linux/syscalls.h>
36
#include <linux/kexec.h>
37
#include <linux/kdb.h>
38
#include <linux/ratelimit.h>
39
#include <linux/kmsg_dump.h>
40
#include <linux/syslog.h>
41 42
#include <linux/cpu.h>
#include <linux/notifier.h>
43
#include <linux/rculist.h>
44
#include <linux/poll.h>
L
Linus Torvalds 已提交
45 46 47

#include <asm/uaccess.h>

48 49 50
#define CREATE_TRACE_POINTS
#include <trace/events/printk.h>

51 52 53
/*
 * Architectures can override it:
 */
54
void asmlinkage __attribute__((weak)) early_printk(const char *fmt, ...)
55 56 57
{
}

L
Linus Torvalds 已提交
58
/* printk's without a loglevel use this.. */
59
#define DEFAULT_MESSAGE_LOGLEVEL CONFIG_DEFAULT_MESSAGE_LOGLEVEL
L
Linus Torvalds 已提交
60 61 62 63 64 65 66 67 68 69 70 71 72 73 74

/* 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 */

DECLARE_WAIT_QUEUE_HEAD(log_wait);

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 */
};

/*
P
Patrick Pletscher 已提交
75
 * Low level drivers may need that to know if they can schedule in
L
Linus Torvalds 已提交
76 77 78 79 80 81 82 83 84 85
 * 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.
 */
86
static DEFINE_SEMAPHORE(console_sem);
L
Linus Torvalds 已提交
87
struct console *console_drivers;
I
Ingo Molnar 已提交
88 89
EXPORT_SYMBOL_GPL(console_drivers);

L
Linus Torvalds 已提交
90 91 92 93 94 95 96 97
/*
 * 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
 */
98
static int console_locked, console_suspended;
L
Linus Torvalds 已提交
99

100 101 102 103 104
/*
 * If exclusive_console is non-NULL then only this console is to be printed to.
 */
static struct console *exclusive_console;

L
Linus Torvalds 已提交
105 106 107 108 109 110 111 112
/*
 *	Array of consoles built from command line options (console=)
 */
struct console_cmdline
{
	char	name[8];			/* Name of the driver	    */
	int	index;				/* Minor dev. to use	    */
	char	*options;			/* Options for the driver   */
113 114 115
#ifdef CONFIG_A11Y_BRAILLE_CONSOLE
	char	*brl_options;			/* Options for braille driver */
#endif
L
Linus Torvalds 已提交
116 117 118 119 120 121 122
};

#define MAX_CMDLINECONSOLES 8

static struct console_cmdline console_cmdline[MAX_CMDLINECONSOLES];
static int selected_console = -1;
static int preferred_console = -1;
123 124
int console_set_on_cmdline;
EXPORT_SYMBOL(console_set_on_cmdline);
L
Linus Torvalds 已提交
125 126 127 128

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

129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151
/*
 * 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.
 *
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
 * 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.
194 195
 */

196 197 198 199 200
enum log_flags {
	LOG_DEFAULT = 0,
	LOG_NOCONS = 1,		/* already flushed, do not print to console */
};

201 202 203 204 205
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 */
206 207 208
	u8 facility;		/* syslog facility */
	u8 flags:5;		/* internal record flags */
	u8 level:3;		/* syslog level */
209 210 211 212 213 214 215
};

/*
 * 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);
M
Matt Mackall 已提交
216

217 218 219
/* the next printk record to read by syslog(READ) or /proc/kmsg */
static u64 syslog_seq;
static u32 syslog_idx;
220 221 222 223 224 225 226

/* 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;
227
#ifdef CONFIG_PRINTK
228 229 230 231 232 233
static u32 log_next_idx;

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

234 235 236
#define LOG_LINE_MAX 1024

/* record buffer */
237
#if defined(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS)
238 239
#define LOG_ALIGN 4
#else
240
#define LOG_ALIGN __alignof__(struct log)
241
#endif
242
#define __LOG_BUF_LEN (1 << CONFIG_LOG_BUF_SHIFT)
243
static char __log_buf[__LOG_BUF_LEN] __aligned(LOG_ALIGN);
244 245 246 247 248
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;
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

/* 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,
296
		      enum log_flags flags, u64 ts_nsec,
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
		      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;
340 341 342 343 344 345 346
	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();
347 348 349 350 351 352 353
	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++;
}
M
Matt Mackall 已提交
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
/* /dev/kmsg - userspace message inject/listen interface */
struct devkmsg_user {
	u64 seq;
	u32 idx;
	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++) {
		if (copy_from_user(line, iv[i].iov_base, iv[i].iov_len))
			goto out;
		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;
422
	u64 ts_usec;
423 424 425 426 427 428 429
	size_t i;
	size_t len;
	ssize_t ret;

	if (!user)
		return -EBADF;

430 431 432
	ret = mutex_lock_interruptible(&user->lock);
	if (ret)
		return ret;
433
	raw_spin_lock_irq(&logbuf_lock);
434 435 436
	while (user->seq == log_next_seq) {
		if (file->f_flags & O_NONBLOCK) {
			ret = -EAGAIN;
437
			raw_spin_unlock_irq(&logbuf_lock);
438 439 440
			goto out;
		}

441
		raw_spin_unlock_irq(&logbuf_lock);
442 443 444 445
		ret = wait_event_interruptible(log_wait,
					       user->seq != log_next_seq);
		if (ret)
			goto out;
446
		raw_spin_lock_irq(&logbuf_lock);
447 448 449 450 451 452 453
	}

	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;
454
		raw_spin_unlock_irq(&logbuf_lock);
455 456 457 458
		goto out;
	}

	msg = log_from_idx(user->idx);
459 460
	ts_usec = msg->ts_nsec;
	do_div(ts_usec, 1000);
461
	len = sprintf(user->buf, "%u,%llu,%llu;",
462
		      (msg->facility << 3) | msg->level, user->seq, ts_usec);
463 464 465

	/* escape non-printable characters */
	for (i = 0; i < msg->text_len; i++) {
466
		unsigned char c = log_text(msg)[i];
467 468 469 470 471 472 473 474 475 476 477 478

		if (c < ' ' || c >= 128)
			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++) {
479
			unsigned char c = log_dict(msg)[i];
480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503

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

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

			if (c < ' ' || c >= 128) {
				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++;
504
	raw_spin_unlock_irq(&logbuf_lock);
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

	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;

531
	raw_spin_lock_irq(&logbuf_lock);
532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554
	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;
	}
555
	raw_spin_unlock_irq(&logbuf_lock);
556 557 558 559 560 561 562 563 564 565 566 567 568
	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);

569
	raw_spin_lock_irq(&logbuf_lock);
570 571 572 573 574 575
	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;
		ret = POLLIN|POLLRDNORM;
	}
576
	raw_spin_unlock_irq(&logbuf_lock);
577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599

	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;

	err = security_syslog(SYSLOG_ACTION_READ_ALL);
	if (err)
		return err;

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

	mutex_init(&user->lock);

600
	raw_spin_lock_irq(&logbuf_lock);
601 602
	user->idx = log_first_idx;
	user->seq = log_first_seq;
603
	raw_spin_unlock_irq(&logbuf_lock);
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

	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,
};

630 631 632 633 634 635 636 637 638 639 640 641 642
#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);
643 644
	VMCOREINFO_SYMBOL(log_first_idx);
	VMCOREINFO_SYMBOL(log_next_idx);
645 646 647
}
#endif

648 649 650 651
/* 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 */
L
Linus Torvalds 已提交
652 653
static int __init log_buf_len_setup(char *str)
{
654
	unsigned size = memparse(str, &str);
L
Linus Torvalds 已提交
655 656 657

	if (size)
		size = roundup_pow_of_two(size);
658 659 660 661
	if (size > log_buf_len)
		new_log_buf_len = size;

	return 0;
L
Linus Torvalds 已提交
662
}
663 664 665 666 667 668 669 670 671 672
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;
L
Linus Torvalds 已提交
673

674 675 676 677
	if (early) {
		unsigned long mem;

		mem = memblock_alloc(new_log_buf_len, PAGE_SIZE);
T
Tejun Heo 已提交
678
		if (!mem)
679 680 681 682 683 684 685 686 687 688 689 690
			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;
	}

691
	raw_spin_lock_irqsave(&logbuf_lock, flags);
692 693 694
	log_buf_len = new_log_buf_len;
	log_buf = new_log_buf;
	new_log_buf_len = 0;
695 696
	free = __LOG_BUF_LEN - log_next_idx;
	memcpy(log_buf, __log_buf, __LOG_BUF_LEN);
697
	raw_spin_unlock_irqrestore(&logbuf_lock, flags);
698 699 700 701 702

	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);
}
L
Linus Torvalds 已提交
703

R
Randy Dunlap 已提交
704 705
#ifdef CONFIG_BOOT_PRINTK_DELAY

706
static int boot_delay; /* msecs delay after each printk during bootup */
707
static unsigned long long loops_per_msec;	/* based on boot_delay */
R
Randy Dunlap 已提交
708 709 710 711 712 713 714 715 716 717 718 719

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;

720 721 722
	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);
R
Randy Dunlap 已提交
723 724 725 726 727 728 729 730 731 732 733 734
	return 1;
}
__setup("boot_delay=", boot_delay_setup);

static void boot_delay_msec(void)
{
	unsigned long long k;
	unsigned long timeout;

	if (boot_delay == 0 || system_state != SYSTEM_BOOTING)
		return;

735
	k = (unsigned long long)loops_per_msec * boot_delay;
R
Randy Dunlap 已提交
736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756

	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
static inline void boot_delay_msec(void)
{
}
#endif

757 758 759 760 761 762
#ifdef CONFIG_SECURITY_DMESG_RESTRICT
int dmesg_restrict = 1;
#else
int dmesg_restrict;
#endif

763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784
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)) {
785 786 787 788
			printk_once(KERN_WARNING "%s (%d): "
				 "Attempt to access syslog with CAP_SYS_ADMIN "
				 "but no CAP_SYSLOG (deprecated).\n",
				 current->comm, task_pid_nr(current));
789 790 791 792 793 794 795
			return 0;
		}
		return -EPERM;
	}
	return 0;
}

796 797 798 799 800 801 802
#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);

803 804 805 806 807 808 809 810 811 812 813 814 815 816 817
static size_t print_time(u64 ts, char *buf)
{
	unsigned long rem_nsec;

	if (!printk_time)
		return 0;

	if (!buf)
		return 15;

	rem_nsec = do_div(ts, 1000000000);
	return sprintf(buf, "[%5lu.%06lu] ",
		       (unsigned long)ts, rem_nsec / 1000);
}

818
static size_t print_prefix(const struct log *msg, bool syslog, char *buf)
819
{
820
	size_t len = 0;
821

822 823 824 825 826 827 828 829 830 831 832
	if (syslog) {
		if (buf) {
			len += sprintf(buf, "<%u>", msg->level);
		} else {
			len += 3;
			if (msg->level > 9)
				len++;
			if (msg->level > 99)
				len++;
		}
	}
833

834
	len += print_time(msg->ts_nsec, buf ? buf + len : NULL);
835
	return len;
836 837
}

838 839
static size_t msg_print_text(const struct log *msg, bool syslog,
			     char *buf, size_t size)
840
{
841 842 843 844 845 846 847 848 849 850 851 852 853 854 855
	const char *text = log_text(msg);
	size_t text_size = msg->text_len;
	size_t len = 0;

	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;
		}
856

857 858 859 860
		if (buf) {
			if (print_prefix(msg, syslog, NULL) +
			    text_len + 1>= size - len)
				break;
861

862 863 864 865 866 867 868 869 870
			len += print_prefix(msg, syslog, buf + len);
			memcpy(buf + len, text, text_len);
			len += text_len;
			buf[len++] = '\n';
		} else {
			/* SYSLOG_ACTION_* buffer size only calculation */
			len += print_prefix(msg, syslog, NULL);
			len += text_len + 1;
		}
871

872 873
		text = next;
	} while (text);
874 875 876 877 878 879 880

	return len;
}

static int syslog_print(char __user *buf, int size)
{
	char *text;
881
	struct log *msg;
882
	int len = 0;
883 884 885 886 887

	text = kmalloc(LOG_LINE_MAX, GFP_KERNEL);
	if (!text)
		return -ENOMEM;

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
	while (size > 0) {
		size_t n;

		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;
		}
		if (syslog_seq == log_next_seq) {
			raw_spin_unlock_irq(&logbuf_lock);
			break;
		}
		msg = log_from_idx(syslog_idx);
		n = msg_print_text(msg, true, text, LOG_LINE_MAX);
		if (n <= size) {
			syslog_idx = log_next(syslog_idx);
			syslog_seq++;
		} else
			n = 0;
		raw_spin_unlock_irq(&logbuf_lock);

		if (!n)
			break;

		len += n;
		size -= n;
		buf += n;
		n = copy_to_user(buf - n, text, n);

		if (n) {
			len -= n;
			if (!len)
				len = -EFAULT;
			break;
		}
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
	}

	kfree(text);
	return len;
}

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

	text = kmalloc(LOG_LINE_MAX, GFP_KERNEL);
	if (!text)
		return -ENOMEM;

	raw_spin_lock_irq(&logbuf_lock);
	if (buf) {
		u64 next_seq;
		u64 seq;
		u32 idx;

		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.
954
		 */
955 956 957
		seq = clear_seq;
		idx = clear_idx;
		while (seq < log_next_seq) {
958 959 960
			struct log *msg = log_from_idx(idx);

			len += msg_print_text(msg, true, NULL, 0);
961 962 963
			idx = log_next(idx);
			seq++;
		}
964 965

		/* move first record forward until length fits into the buffer */
966 967 968
		seq = clear_seq;
		idx = clear_idx;
		while (len > size && seq < log_next_seq) {
969 970 971
			struct log *msg = log_from_idx(idx);

			len -= msg_print_text(msg, true, NULL, 0);
972 973 974 975
			idx = log_next(idx);
			seq++;
		}

976
		/* last message fitting into this dump */
977 978 979 980
		next_seq = log_next_seq;

		len = 0;
		while (len >= 0 && seq < next_seq) {
981
			struct log *msg = log_from_idx(idx);
982 983
			int textlen;

984
			textlen = msg_print_text(msg, true, text, LOG_LINE_MAX);
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
			if (textlen < 0) {
				len = textlen;
				break;
			}
			idx = log_next(idx);
			seq++;

			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;
			}
		}
	}

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

	kfree(text);
	return len;
}

1017
int do_syslog(int type, char __user *buf, int len, bool from_file)
L
Linus Torvalds 已提交
1018
{
1019 1020
	bool clear = false;
	static int saved_console_loglevel = -1;
1021
	static DEFINE_MUTEX(syslog_mutex);
1022
	int error;
L
Linus Torvalds 已提交
1023

1024 1025 1026
	error = check_syslog_permissions(type, from_file);
	if (error)
		goto out;
1027 1028

	error = security_syslog(type);
L
Linus Torvalds 已提交
1029 1030 1031 1032
	if (error)
		return error;

	switch (type) {
1033
	case SYSLOG_ACTION_CLOSE:	/* Close log */
L
Linus Torvalds 已提交
1034
		break;
1035
	case SYSLOG_ACTION_OPEN:	/* Open log */
L
Linus Torvalds 已提交
1036
		break;
1037
	case SYSLOG_ACTION_READ:	/* Read from log */
L
Linus Torvalds 已提交
1038 1039 1040 1041 1042 1043 1044 1045 1046 1047
		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;
		}
1048 1049 1050
		error = mutex_lock_interruptible(&syslog_mutex);
		if (error)
			goto out;
J
Jesper Juhl 已提交
1051
		error = wait_event_interruptible(log_wait,
1052
						 syslog_seq != log_next_seq);
1053 1054
		if (error) {
			mutex_unlock(&syslog_mutex);
L
Linus Torvalds 已提交
1055
			goto out;
1056
		}
1057
		error = syslog_print(buf, len);
1058
		mutex_unlock(&syslog_mutex);
L
Linus Torvalds 已提交
1059
		break;
1060 1061
	/* Read/clear last kernel messages */
	case SYSLOG_ACTION_READ_CLEAR:
1062
		clear = true;
L
Linus Torvalds 已提交
1063
		/* FALL THRU */
1064 1065
	/* Read last kernel messages */
	case SYSLOG_ACTION_READ_ALL:
L
Linus Torvalds 已提交
1066 1067 1068 1069 1070 1071 1072 1073 1074 1075
		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;
		}
1076
		error = syslog_print_all(buf, len, clear);
L
Linus Torvalds 已提交
1077
		break;
1078 1079
	/* Clear ring buffer */
	case SYSLOG_ACTION_CLEAR:
1080
		syslog_print_all(NULL, 0, true);
1081
		break;
1082 1083
	/* Disable logging to console */
	case SYSLOG_ACTION_CONSOLE_OFF:
1084 1085
		if (saved_console_loglevel == -1)
			saved_console_loglevel = console_loglevel;
L
Linus Torvalds 已提交
1086 1087
		console_loglevel = minimum_console_loglevel;
		break;
1088 1089
	/* Enable logging to console */
	case SYSLOG_ACTION_CONSOLE_ON:
1090 1091 1092 1093
		if (saved_console_loglevel != -1) {
			console_loglevel = saved_console_loglevel;
			saved_console_loglevel = -1;
		}
L
Linus Torvalds 已提交
1094
		break;
1095 1096
	/* Set level of messages printed to console */
	case SYSLOG_ACTION_CONSOLE_LEVEL:
L
Linus Torvalds 已提交
1097 1098 1099 1100 1101 1102
		error = -EINVAL;
		if (len < 1 || len > 8)
			goto out;
		if (len < minimum_console_loglevel)
			len = minimum_console_loglevel;
		console_loglevel = len;
1103 1104
		/* Implicitly re-enable logging to console */
		saved_console_loglevel = -1;
L
Linus Torvalds 已提交
1105 1106
		error = 0;
		break;
1107 1108
	/* Number of chars in the log buffer */
	case SYSLOG_ACTION_SIZE_UNREAD:
1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129
		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;
		}
		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 {
			u64 seq;
			u32 idx;

			error = 0;
			seq = syslog_seq;
			idx = syslog_idx;
			while (seq < log_next_seq) {
1130 1131 1132
				struct log *msg = log_from_idx(idx);

				error += msg_print_text(msg, true, NULL, 0);
1133 1134 1135 1136 1137
				idx = log_next(idx);
				seq++;
			}
		}
		raw_spin_unlock_irq(&logbuf_lock);
L
Linus Torvalds 已提交
1138
		break;
1139 1140
	/* Size of the log buffer */
	case SYSLOG_ACTION_SIZE_BUFFER:
L
Linus Torvalds 已提交
1141 1142 1143 1144 1145 1146 1147 1148 1149 1150
		error = log_buf_len;
		break;
	default:
		error = -EINVAL;
		break;
	}
out:
	return error;
}

1151
SYSCALL_DEFINE3(syslog, int, type, char __user *, buf, int, len)
L
Linus Torvalds 已提交
1152
{
1153
	return do_syslog(type, buf, len, SYSLOG_FROM_CALL);
L
Linus Torvalds 已提交
1154 1155
}

1156 1157 1158 1159 1160 1161 1162 1163 1164 1165
#ifdef	CONFIG_KGDB_KDB
/* kdb dmesg command needs access to the syslog buffer.  do_syslog()
 * uses locks so it cannot be used during debugging.  Just tell kdb
 * where the start and end of the physical and logical logs are.  This
 * is equivalent to do_syslog(3).
 */
void kdb_syslog_data(char *syslog_data[4])
{
	syslog_data[0] = log_buf;
	syslog_data[1] = log_buf + log_buf_len;
1166 1167
	syslog_data[2] = log_buf + log_first_idx;
	syslog_data[3] = log_buf + log_next_idx;
1168 1169 1170
}
#endif	/* CONFIG_KGDB_KDB */

1171
static bool __read_mostly ignore_loglevel;
1172

1173
static int __init ignore_loglevel_setup(char *str)
1174 1175 1176 1177
{
	ignore_loglevel = 1;
	printk(KERN_INFO "debug: ignoring loglevel setting.\n");

1178
	return 0;
1179 1180
}

1181
early_param("ignore_loglevel", ignore_loglevel_setup);
1182
module_param(ignore_loglevel, bool, S_IRUGO | S_IWUSR);
1183 1184
MODULE_PARM_DESC(ignore_loglevel, "ignore loglevel setting, to"
	"print all kernel messages to the console.");
1185

L
Linus Torvalds 已提交
1186 1187 1188
/*
 * Call the console drivers, asking them to write out
 * log_buf[start] to log_buf[end - 1].
1189
 * The console_lock must be held.
L
Linus Torvalds 已提交
1190
 */
1191
static void call_console_drivers(int level, const char *text, size_t len)
L
Linus Torvalds 已提交
1192
{
1193
	struct console *con;
L
Linus Torvalds 已提交
1194

1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213
	trace_console(text, 0, len, len);

	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);
	}
L
Linus Torvalds 已提交
1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225
}

/*
 * 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) &&
J
Jesper Juhl 已提交
1226
			!time_after(jiffies, oops_timestamp + 30 * HZ))
L
Linus Torvalds 已提交
1227 1228 1229 1230
		return;

	oops_timestamp = jiffies;

1231
	debug_locks_off();
L
Linus Torvalds 已提交
1232
	/* If a crash is occurring, make sure we can't deadlock */
1233
	raw_spin_lock_init(&logbuf_lock);
L
Linus Torvalds 已提交
1234
	/* And make sure that we print immediately */
1235
	sema_init(&console_sem, 1);
L
Linus Torvalds 已提交
1236 1237
}

1238 1239 1240 1241 1242
/* Check if we have any console registered that can be called early in boot. */
static int have_callable_console(void)
{
	struct console *con;

1243
	for_each_console(con)
1244 1245 1246 1247 1248 1249
		if (con->flags & CON_ANYTIME)
			return 1;

	return 0;
}

1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265
/*
 * 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
1266
 * console_lock held, and 'console_locked' set) if it
1267 1268 1269 1270 1271 1272
 * 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.
 */
1273
static int console_trylock_for_printk(unsigned int cpu)
N
Namhyung Kim 已提交
1274
	__releases(&logbuf_lock)
1275
{
1276
	int retval = 0, wake = 0;
1277

1278
	if (console_trylock()) {
1279 1280 1281 1282 1283 1284 1285 1286 1287 1288
		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;
1289
			wake = 1;
1290 1291 1292
			retval = 0;
		}
	}
1293
	logbuf_cpu = UINT_MAX;
1294 1295
	if (wake)
		up(&console_sem);
1296
	raw_spin_unlock(&logbuf_lock);
1297 1298
	return retval;
}
1299

1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313
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();
		}
	}
}

1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394
/*
 * 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 */
	bool flushed:1;			/* buffer sealed and committed */
} cont;

static void cont_flush(void)
{
	if (cont.flushed)
		return;
	if (cont.len == 0)
		return;

	log_store(cont.facility, cont.level, LOG_NOCONS, cont.ts_nsec,
		  NULL, 0, cont.buf, cont.len);

	cont.flushed = true;
}

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)) {
		cont_flush();
		return false;
	}

	if (!cont.len) {
		cont.facility = facility;
		cont.level = level;
		cont.owner = current;
		cont.ts_nsec = local_clock();
		cont.cons = 0;
		cont.flushed = false;
	}

	memcpy(cont.buf + cont.len, text, len);
	cont.len += len;
	return true;
}

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

	if (cont.cons == 0) {
		textlen += print_time(cont.ts_nsec, text);
		size -= textlen;
	}

	len = cont.len - cont.cons;
	if (len > 0) {
		if (len+1 > size)
			len = size-1;
		memcpy(text + textlen, cont.buf + cont.cons, len);
		textlen += len;
		cont.cons = cont.len;
	}

	if (cont.flushed) {
		text[textlen++] = '\n';
		/* got everything, release buffer */
		cont.len = 0;
	}
	return textlen;
}

1395 1396 1397
asmlinkage int vprintk_emit(int facility, int level,
			    const char *dict, size_t dictlen,
			    const char *fmt, va_list args)
L
Linus Torvalds 已提交
1398
{
1399 1400 1401
	static int recursion_bug;
	static char textbuf[LOG_LINE_MAX];
	char *text = textbuf;
1402
	size_t text_len;
1403
	unsigned long flags;
1404
	int this_cpu;
1405
	bool newline = false;
1406
	bool prefix = false;
1407
	int printed_len = 0;
L
Linus Torvalds 已提交
1408

R
Randy Dunlap 已提交
1409
	boot_delay_msec();
1410
	printk_delay();
R
Randy Dunlap 已提交
1411

L
Linus Torvalds 已提交
1412
	/* This stops the holder of console_sem just where we want him */
1413
	local_irq_save(flags);
1414 1415 1416 1417 1418
	this_cpu = smp_processor_id();

	/*
	 * Ouch, printk recursed into itself!
	 */
1419
	if (unlikely(logbuf_cpu == this_cpu)) {
1420 1421 1422 1423 1424 1425 1426
		/*
		 * If a crash is occurring during printk() on this CPU,
		 * then try to get the crash message out but make sure
		 * we can't deadlock. Otherwise just return to avoid the
		 * recursion and return - but flag the recursion so that
		 * it can be printed at the next appropriate moment:
		 */
1427
		if (!oops_in_progress && !lockdep_recursing(current)) {
1428
			recursion_bug = 1;
1429 1430 1431 1432 1433
			goto out_restore_irqs;
		}
		zap_locks();
	}

1434
	lockdep_off();
1435
	raw_spin_lock(&logbuf_lock);
1436
	logbuf_cpu = this_cpu;
L
Linus Torvalds 已提交
1437

1438
	if (recursion_bug) {
1439 1440 1441
		static const char recursion_msg[] =
			"BUG: recent printk recursion!";

1442
		recursion_bug = 0;
1443 1444
		printed_len += strlen(recursion_msg);
		/* emit KERN_CRIT message */
1445 1446
		log_store(0, 2, LOG_DEFAULT, 0,
			  NULL, 0, recursion_msg, printed_len);
1447
	}
L
Linus Torvalds 已提交
1448

1449 1450 1451 1452
	/*
	 * The printf needs to come first; we need the syslog
	 * prefix which might be passed-in as a parameter.
	 */
1453
	text_len = vscnprintf(text, sizeof(textbuf), fmt, args);
1454

1455
	/* mark and strip a trailing newline */
1456 1457
	if (text_len && text[text_len-1] == '\n') {
		text_len--;
1458 1459
		newline = true;
	}
1460

1461
	/* strip syslog prefix and extract log level or control flags */
1462 1463 1464 1465 1466 1467
	if (text[0] == '<' && text[1] && text[2] == '>') {
		switch (text[1]) {
		case '0' ... '7':
			if (level == -1)
				level = text[1] - '0';
		case 'd':	/* KERN_DEFAULT */
1468 1469
			prefix = true;
		case 'c':	/* KERN_CONT */
1470
			text += 3;
1471
			text_len -= 3;
1472 1473 1474
		}
	}

1475 1476
	if (level == -1)
		level = default_message_loglevel;
1477

1478 1479 1480
	if (dict) {
		prefix = true;
		newline = true;
1481
	}
1482

1483
	if (!newline) {
1484 1485 1486 1487 1488 1489
		/*
		 * Flush the conflicting buffer. An earlier newline was missing,
		 * or another task also prints continuation lines.
		 */
		if (cont.len && (prefix || cont.owner != current))
			cont_flush();
1490

1491 1492 1493 1494
		/* buffer line if possible, otherwise store it right away */
		if (!cont_add(facility, level, text, text_len))
			log_store(facility, level, LOG_DEFAULT, 0,
				  dict, dictlen, text, text_len);
1495
	} else {
1496
		bool stored = false;
1497

1498
		/*
1499 1500 1501 1502
		 * If an earlier newline was missing and it was the same task,
		 * either merge it with the current buffer and flush, or if
		 * there was a race with interrupts (prefix == true) then just
		 * flush it out and store this line separately.
1503 1504
		 */
		if (cont.len && cont.owner == current) {
1505 1506
			if (!prefix)
				stored = cont_add(facility, level, text, text_len);
1507
			cont_flush();
1508
		}
1509 1510 1511 1512

		if (!stored)
			log_store(facility, level, LOG_DEFAULT, 0,
				  dict, dictlen, text, text_len);
L
Linus Torvalds 已提交
1513
	}
1514
	printed_len += text_len;
L
Linus Torvalds 已提交
1515

1516
	/*
1517 1518 1519
	 * Try to acquire and then immediately release the console semaphore.
	 * The release will print out buffers and wake up /dev/kmsg and syslog()
	 * users.
1520
	 *
1521 1522
	 * The console_trylock_for_printk() function will release 'logbuf_lock'
	 * regardless of whether it actually gets the console semaphore or not.
1523
	 */
1524 1525
	if (console_trylock_for_printk(this_cpu))
		console_unlock();
1526

1527
	lockdep_on();
1528
out_restore_irqs:
1529
	local_irq_restore(flags);
1530

L
Linus Torvalds 已提交
1531 1532
	return printed_len;
}
1533 1534 1535 1536 1537 1538
EXPORT_SYMBOL(vprintk_emit);

asmlinkage int vprintk(const char *fmt, va_list args)
{
	return vprintk_emit(0, -1, NULL, 0, fmt, args);
}
L
Linus Torvalds 已提交
1539 1540
EXPORT_SYMBOL(vprintk);

1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596
asmlinkage int printk_emit(int facility, int level,
			   const char *dict, size_t dictlen,
			   const char *fmt, ...)
{
	va_list args;
	int r;

	va_start(args, fmt);
	r = vprintk_emit(facility, level, dict, dictlen, fmt, args);
	va_end(args);

	return r;
}
EXPORT_SYMBOL(printk_emit);

/**
 * printk - print a kernel message
 * @fmt: format string
 *
 * This is printk(). It can be called from any context. We want it to work.
 *
 * We try to grab the console_lock. If we succeed, it's easy - we log the
 * output and call the console drivers.  If we fail to get the semaphore, we
 * place the output into the log buffer and return. The current holder of
 * the console_sem will notice the new output in console_unlock(); and will
 * send it to the consoles before releasing the lock.
 *
 * One effect of this deferred printing is that code which calls printk() and
 * then changes console_loglevel may break. This is because console_loglevel
 * is inspected when the actual printing occurs.
 *
 * See also:
 * printf(3)
 *
 * See the vsnprintf() documentation for format string extensions over C99.
 */
asmlinkage int printk(const char *fmt, ...)
{
	va_list args;
	int r;

#ifdef CONFIG_KGDB_KDB
	if (unlikely(kdb_trap_printk)) {
		va_start(args, fmt);
		r = vkdb_printf(fmt, args);
		va_end(args);
		return r;
	}
#endif
	va_start(args, fmt);
	r = vprintk_emit(0, -1, NULL, 0, fmt, args);
	va_end(args);

	return r;
}
EXPORT_SYMBOL(printk);
1597

M
Matt Mackall 已提交
1598 1599
#else

1600
#define LOG_LINE_MAX 0
1601 1602 1603 1604 1605 1606
static struct cont {
	size_t len;
	size_t cons;
	u8 level;
	bool flushed:1;
} cont;
1607 1608 1609
static struct log *log_from_idx(u32 idx) { return NULL; }
static u32 log_next(u32 idx) { return 0; }
static void call_console_drivers(int level, const char *text, size_t len) {}
1610 1611
static size_t msg_print_text(const struct log *msg, bool syslog,
			     char *buf, size_t size) { return 0; }
1612
static size_t cont_print_text(char *text, size_t size) { return 0; }
M
Matt Mackall 已提交
1613

1614
#endif /* CONFIG_PRINTK */
M
Matt Mackall 已提交
1615

1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645
static int __add_preferred_console(char *name, int idx, char *options,
				   char *brl_options)
{
	struct console_cmdline *c;
	int i;

	/*
	 *	See if this tty is not yet registered, and
	 *	if we have a slot free.
	 */
	for (i = 0; i < MAX_CMDLINECONSOLES && console_cmdline[i].name[0]; i++)
		if (strcmp(console_cmdline[i].name, name) == 0 &&
			  console_cmdline[i].index == idx) {
				if (!brl_options)
					selected_console = i;
				return 0;
		}
	if (i == MAX_CMDLINECONSOLES)
		return -E2BIG;
	if (!brl_options)
		selected_console = i;
	c = &console_cmdline[i];
	strlcpy(c->name, name, sizeof(c->name));
	c->options = options;
#ifdef CONFIG_A11Y_BRAILLE_CONSOLE
	c->brl_options = brl_options;
#endif
	c->index = idx;
	return 0;
}
1646 1647 1648 1649 1650
/*
 * Set up a list of consoles.  Called from init/main.c
 */
static int __init console_setup(char *str)
{
Y
Yinghai Lu 已提交
1651
	char buf[sizeof(console_cmdline[0].name) + 4]; /* 4 for index */
1652
	char *s, *options, *brl_options = NULL;
1653 1654
	int idx;

1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669
#ifdef CONFIG_A11Y_BRAILLE_CONSOLE
	if (!memcmp(str, "brl,", 4)) {
		brl_options = "";
		str += 4;
	} else if (!memcmp(str, "brl=", 4)) {
		brl_options = str + 4;
		str = strchr(brl_options, ',');
		if (!str) {
			printk(KERN_ERR "need port name after brl=\n");
			return 1;
		}
		*(str++) = 0;
	}
#endif

1670 1671 1672 1673
	/*
	 * Decode str into name, index, options.
	 */
	if (str[0] >= '0' && str[0] <= '9') {
Y
Yinghai Lu 已提交
1674 1675
		strcpy(buf, "ttyS");
		strncpy(buf + 4, str, sizeof(buf) - 5);
1676
	} else {
Y
Yinghai Lu 已提交
1677
		strncpy(buf, str, sizeof(buf) - 1);
1678
	}
Y
Yinghai Lu 已提交
1679
	buf[sizeof(buf) - 1] = 0;
1680 1681 1682 1683
	if ((options = strchr(str, ',')) != NULL)
		*(options++) = 0;
#ifdef __sparc__
	if (!strcmp(str, "ttya"))
Y
Yinghai Lu 已提交
1684
		strcpy(buf, "ttyS0");
1685
	if (!strcmp(str, "ttyb"))
Y
Yinghai Lu 已提交
1686
		strcpy(buf, "ttyS1");
1687
#endif
Y
Yinghai Lu 已提交
1688
	for (s = buf; *s; s++)
1689 1690 1691 1692 1693
		if ((*s >= '0' && *s <= '9') || *s == ',')
			break;
	idx = simple_strtoul(s, NULL, 10);
	*s = 0;

1694
	__add_preferred_console(buf, idx, options, brl_options);
1695
	console_set_on_cmdline = 1;
1696 1697 1698 1699
	return 1;
}
__setup("console=", console_setup);

1700 1701
/**
 * add_preferred_console - add a device to the list of preferred consoles.
1702 1703 1704
 * @name: device name
 * @idx: device index
 * @options: options for this console
1705 1706 1707 1708 1709 1710 1711 1712
 *
 * The last preferred console added will be used for kernel messages
 * and stdin/out/err for init.  Normally this is used by console_setup
 * above to handle user-supplied console arguments; however it can also
 * be used by arch-specific code either to override the user or more
 * commonly to provide a default console (ie from PROM variables) when
 * the user has not supplied one.
 */
1713
int add_preferred_console(char *name, int idx, char *options)
1714
{
1715
	return __add_preferred_console(name, idx, options, NULL);
1716 1717
}

1718
int update_console_cmdline(char *name, int idx, char *name_new, int idx_new, char *options)
1719 1720 1721 1722 1723 1724 1725 1726
{
	struct console_cmdline *c;
	int i;

	for (i = 0; i < MAX_CMDLINECONSOLES && console_cmdline[i].name[0]; i++)
		if (strcmp(console_cmdline[i].name, name) == 0 &&
			  console_cmdline[i].index == idx) {
				c = &console_cmdline[i];
1727
				strlcpy(c->name, name_new, sizeof(c->name));
1728 1729 1730 1731 1732 1733 1734 1735 1736
				c->name[sizeof(c->name) - 1] = 0;
				c->options = options;
				c->index = idx_new;
				return i;
		}
	/* not found */
	return -1;
}

1737
bool console_suspend_enabled = 1;
1738 1739 1740 1741 1742 1743 1744 1745
EXPORT_SYMBOL(console_suspend_enabled);

static int __init console_suspend_disable(char *str)
{
	console_suspend_enabled = 0;
	return 1;
}
__setup("no_console_suspend", console_suspend_disable);
1746 1747 1748 1749
module_param_named(console_suspend, console_suspend_enabled,
		bool, S_IRUGO | S_IWUSR);
MODULE_PARM_DESC(console_suspend, "suspend console during suspend"
	" and hibernate operations");
1750

1751 1752 1753 1754 1755 1756 1757
/**
 * suspend_console - suspend the console subsystem
 *
 * This disables printk() while we go into suspend states
 */
void suspend_console(void)
{
1758 1759
	if (!console_suspend_enabled)
		return;
1760
	printk("Suspending console(s) (use no_console_suspend to debug)\n");
1761
	console_lock();
1762
	console_suspended = 1;
1763
	up(&console_sem);
1764 1765 1766 1767
}

void resume_console(void)
{
1768 1769
	if (!console_suspend_enabled)
		return;
1770
	down(&console_sem);
1771
	console_suspended = 0;
1772
	console_unlock();
1773 1774
}

1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794
/**
 * console_cpu_notify - print deferred console messages after CPU hotplug
 * @self: notifier struct
 * @action: CPU hotplug event
 * @hcpu: unused
 *
 * If printk() is called from a CPU that is not online yet, the messages
 * will be spooled but will not show up on the console.  This function is
 * called when a new CPU comes online (or fails to come up), and ensures
 * that any such output gets printed.
 */
static int __cpuinit console_cpu_notify(struct notifier_block *self,
	unsigned long action, void *hcpu)
{
	switch (action) {
	case CPU_ONLINE:
	case CPU_DEAD:
	case CPU_DYING:
	case CPU_DOWN_FAILED:
	case CPU_UP_CANCELED:
1795 1796
		console_lock();
		console_unlock();
1797 1798 1799 1800
	}
	return NOTIFY_OK;
}

L
Linus Torvalds 已提交
1801
/**
1802
 * console_lock - lock the console system for exclusive use.
L
Linus Torvalds 已提交
1803
 *
1804
 * Acquires a lock which guarantees that the caller has
L
Linus Torvalds 已提交
1805 1806 1807 1808
 * exclusive access to the console system and the console_drivers list.
 *
 * Can sleep, returns nothing.
 */
1809
void console_lock(void)
L
Linus Torvalds 已提交
1810
{
1811
	BUG_ON(in_interrupt());
L
Linus Torvalds 已提交
1812
	down(&console_sem);
1813 1814
	if (console_suspended)
		return;
L
Linus Torvalds 已提交
1815 1816 1817
	console_locked = 1;
	console_may_schedule = 1;
}
1818
EXPORT_SYMBOL(console_lock);
L
Linus Torvalds 已提交
1819

1820 1821 1822 1823 1824 1825 1826 1827 1828
/**
 * console_trylock - try to lock the console system for exclusive use.
 *
 * Tried to acquire a lock which guarantees that the caller has
 * exclusive access to the console system and the console_drivers list.
 *
 * returns 1 on success, and 0 on failure to acquire the lock.
 */
int console_trylock(void)
L
Linus Torvalds 已提交
1829 1830
{
	if (down_trylock(&console_sem))
1831
		return 0;
1832 1833
	if (console_suspended) {
		up(&console_sem);
1834
		return 0;
1835
	}
L
Linus Torvalds 已提交
1836 1837
	console_locked = 1;
	console_may_schedule = 0;
1838
	return 1;
L
Linus Torvalds 已提交
1839
}
1840
EXPORT_SYMBOL(console_trylock);
L
Linus Torvalds 已提交
1841 1842 1843 1844 1845 1846

int is_console_locked(void)
{
	return console_locked;
}

1847
/*
1848
 * Delayed printk version, for scheduler-internal messages:
1849 1850 1851 1852 1853 1854
 */
#define PRINTK_BUF_SIZE		512

#define PRINTK_PENDING_WAKEUP	0x01
#define PRINTK_PENDING_SCHED	0x02

P
Peter Zijlstra 已提交
1855
static DEFINE_PER_CPU(int, printk_pending);
1856
static DEFINE_PER_CPU(char [PRINTK_BUF_SIZE], printk_sched_buf);
P
Peter Zijlstra 已提交
1857 1858

void printk_tick(void)
1859
{
1860
	if (__this_cpu_read(printk_pending)) {
1861 1862 1863 1864 1865 1866 1867
		int pending = __this_cpu_xchg(printk_pending, 0);
		if (pending & PRINTK_PENDING_SCHED) {
			char *buf = __get_cpu_var(printk_sched_buf);
			printk(KERN_WARNING "[sched_delayed] %s", buf);
		}
		if (pending & PRINTK_PENDING_WAKEUP)
			wake_up_interruptible(&log_wait);
P
Peter Zijlstra 已提交
1868 1869 1870 1871 1872
	}
}

int printk_needs_cpu(int cpu)
{
1873
	if (cpu_is_offline(cpu))
1874
		printk_tick();
1875
	return __this_cpu_read(printk_pending);
P
Peter Zijlstra 已提交
1876 1877 1878 1879 1880
}

void wake_up_klogd(void)
{
	if (waitqueue_active(&log_wait))
1881
		this_cpu_or(printk_pending, PRINTK_PENDING_WAKEUP);
1882 1883
}

1884 1885 1886 1887
/* the next printk record to write to the console */
static u64 console_seq;
static u32 console_idx;

L
Linus Torvalds 已提交
1888
/**
1889
 * console_unlock - unlock the console system
L
Linus Torvalds 已提交
1890
 *
1891
 * Releases the console_lock which the caller holds on the console system
L
Linus Torvalds 已提交
1892 1893
 * and the console driver list.
 *
1894 1895 1896
 * While the console_lock was held, console output may have been buffered
 * by printk().  If this is the case, console_unlock(); emits
 * the output prior to releasing the lock.
L
Linus Torvalds 已提交
1897
 *
1898
 * If there is output waiting, we wake /dev/kmsg and syslog() users.
L
Linus Torvalds 已提交
1899
 *
1900
 * console_unlock(); may be called from any context.
L
Linus Torvalds 已提交
1901
 */
1902
void console_unlock(void)
L
Linus Torvalds 已提交
1903
{
1904
	static char text[LOG_LINE_MAX];
1905
	static u64 seen_seq;
L
Linus Torvalds 已提交
1906
	unsigned long flags;
1907 1908
	bool wake_klogd = false;
	bool retry;
L
Linus Torvalds 已提交
1909

1910
	if (console_suspended) {
1911
		up(&console_sem);
1912 1913
		return;
	}
1914 1915 1916

	console_may_schedule = 0;

1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930
	/* flush buffered message fragment immediately to console */
	raw_spin_lock_irqsave(&logbuf_lock, flags);
	if (cont.len && (cont.cons < cont.len || cont.flushed)) {
		size_t len;

		len = cont_print_text(text, sizeof(text));
		raw_spin_unlock(&logbuf_lock);
		stop_critical_timings();
		call_console_drivers(cont.level, text, len);
		start_critical_timings();
		local_irq_restore(flags);
	} else
		raw_spin_unlock_irqrestore(&logbuf_lock, flags);

1931
again:
1932 1933
	for (;;) {
		struct log *msg;
1934
		size_t len;
1935 1936
		int level;

1937
		raw_spin_lock_irqsave(&logbuf_lock, flags);
1938 1939 1940 1941 1942 1943 1944 1945 1946 1947
		if (seen_seq != log_next_seq) {
			wake_klogd = true;
			seen_seq = log_next_seq;
		}

		if (console_seq < log_first_seq) {
			/* messages are gone, move to first one */
			console_seq = log_first_seq;
			console_idx = log_first_idx;
		}
1948
skip:
1949 1950 1951 1952
		if (console_seq == log_next_seq)
			break;

		msg = log_from_idx(console_idx);
1953 1954 1955 1956 1957 1958 1959 1960 1961
		if (msg->flags & LOG_NOCONS) {
			/*
			 * Skip record we have buffered and already printed
			 * directly to the console when we received it.
			 */
			console_idx = log_next(console_idx);
			console_seq++;
			goto skip;
		}
1962

1963
		level = msg->level;
1964
		len = msg_print_text(msg, false, text, sizeof(text));
1965 1966 1967

		console_idx = log_next(console_idx);
		console_seq++;
1968
		raw_spin_unlock(&logbuf_lock);
1969

1970
		stop_critical_timings();	/* don't trace print latency */
1971
		call_console_drivers(level, text, len);
1972
		start_critical_timings();
L
Linus Torvalds 已提交
1973 1974 1975
		local_irq_restore(flags);
	}
	console_locked = 0;
1976 1977 1978 1979 1980

	/* Release the exclusive_console once it is used */
	if (unlikely(exclusive_console))
		exclusive_console = NULL;

1981
	raw_spin_unlock(&logbuf_lock);
1982

1983
	up(&console_sem);
1984 1985 1986 1987 1988 1989 1990

	/*
	 * Someone could have filled up the buffer again, so re-check if there's
	 * something to flush. In case we cannot trylock the console_sem again,
	 * there's a new owner and the console_unlock() from them will do the
	 * flush, no worries.
	 */
1991
	raw_spin_lock(&logbuf_lock);
1992
	retry = console_seq != log_next_seq;
P
Peter Zijlstra 已提交
1993 1994
	raw_spin_unlock_irqrestore(&logbuf_lock, flags);

1995 1996 1997
	if (retry && console_trylock())
		goto again;

1998 1999
	if (wake_klogd)
		wake_up_klogd();
L
Linus Torvalds 已提交
2000
}
2001
EXPORT_SYMBOL(console_unlock);
L
Linus Torvalds 已提交
2002

2003 2004
/**
 * console_conditional_schedule - yield the CPU if required
L
Linus Torvalds 已提交
2005 2006 2007 2008 2009
 *
 * If the console code is currently allowed to sleep, and
 * if this CPU should yield the CPU to another task, do
 * so here.
 *
2010
 * Must be called within console_lock();.
L
Linus Torvalds 已提交
2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030
 */
void __sched console_conditional_schedule(void)
{
	if (console_may_schedule)
		cond_resched();
}
EXPORT_SYMBOL(console_conditional_schedule);

void console_unblank(void)
{
	struct console *c;

	/*
	 * console_unblank can no longer be called in interrupt context unless
	 * oops_in_progress is set to 1..
	 */
	if (oops_in_progress) {
		if (down_trylock(&console_sem) != 0)
			return;
	} else
2031
		console_lock();
L
Linus Torvalds 已提交
2032 2033 2034

	console_locked = 1;
	console_may_schedule = 0;
2035
	for_each_console(c)
L
Linus Torvalds 已提交
2036 2037
		if ((c->flags & CON_ENABLED) && c->unblank)
			c->unblank();
2038
	console_unlock();
L
Linus Torvalds 已提交
2039 2040 2041 2042 2043 2044 2045 2046 2047 2048
}

/*
 * Return the console tty driver structure and its associated index
 */
struct tty_driver *console_device(int *index)
{
	struct console *c;
	struct tty_driver *driver = NULL;

2049
	console_lock();
2050
	for_each_console(c) {
L
Linus Torvalds 已提交
2051 2052 2053 2054 2055 2056
		if (!c->device)
			continue;
		driver = c->device(c, index);
		if (driver)
			break;
	}
2057
	console_unlock();
L
Linus Torvalds 已提交
2058 2059 2060 2061 2062 2063 2064 2065 2066 2067
	return driver;
}

/*
 * Prevent further output on the passed console device so that (for example)
 * serial drivers can disable console output before suspending a port, and can
 * re-enable output afterwards.
 */
void console_stop(struct console *console)
{
2068
	console_lock();
L
Linus Torvalds 已提交
2069
	console->flags &= ~CON_ENABLED;
2070
	console_unlock();
L
Linus Torvalds 已提交
2071 2072 2073 2074 2075
}
EXPORT_SYMBOL(console_stop);

void console_start(struct console *console)
{
2076
	console_lock();
L
Linus Torvalds 已提交
2077
	console->flags |= CON_ENABLED;
2078
	console_unlock();
L
Linus Torvalds 已提交
2079 2080 2081
}
EXPORT_SYMBOL(console_start);

2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093
static int __read_mostly keep_bootcon;

static int __init keep_bootcon_setup(char *str)
{
	keep_bootcon = 1;
	printk(KERN_INFO "debug: skip boot console de-registration.\n");

	return 0;
}

early_param("keep_bootcon", keep_bootcon_setup);

L
Linus Torvalds 已提交
2094 2095 2096 2097 2098
/*
 * The console driver calls this routine during kernel initialization
 * to register the console printing procedure with printk() and to
 * print any messages that were printed by the kernel before the
 * console driver was initialized.
2099 2100 2101 2102 2103 2104 2105 2106 2107 2108 2109 2110 2111
 *
 * This can happen pretty early during the boot process (because of
 * early_printk) - sometimes before setup_arch() completes - be careful
 * of what kernel features are used - they may not be initialised yet.
 *
 * There are two types of consoles - bootconsoles (early_printk) and
 * "real" consoles (everything which is not a bootconsole) which are
 * handled differently.
 *  - Any number of bootconsoles can be registered at any time.
 *  - As soon as a "real" console is registered, all bootconsoles
 *    will be unregistered automatically.
 *  - Once a "real" console is registered, any attempt to register a
 *    bootconsoles will be rejected
L
Linus Torvalds 已提交
2112
 */
2113
void register_console(struct console *newcon)
L
Linus Torvalds 已提交
2114
{
J
Jesper Juhl 已提交
2115
	int i;
L
Linus Torvalds 已提交
2116
	unsigned long flags;
2117
	struct console *bcon = NULL;
L
Linus Torvalds 已提交
2118

2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131
	/*
	 * before we register a new CON_BOOT console, make sure we don't
	 * already have a valid console
	 */
	if (console_drivers && newcon->flags & CON_BOOT) {
		/* find the last or real console */
		for_each_console(bcon) {
			if (!(bcon->flags & CON_BOOT)) {
				printk(KERN_INFO "Too late to register bootconsole %s%d\n",
					newcon->name, newcon->index);
				return;
			}
		}
2132 2133
	}

2134 2135 2136 2137
	if (console_drivers && console_drivers->flags & CON_BOOT)
		bcon = console_drivers;

	if (preferred_console < 0 || bcon || !console_drivers)
L
Linus Torvalds 已提交
2138 2139
		preferred_console = selected_console;

2140 2141
	if (newcon->early_setup)
		newcon->early_setup();
2142

L
Linus Torvalds 已提交
2143 2144 2145 2146 2147 2148
	/*
	 *	See if we want to use this console driver. If we
	 *	didn't select a console we take the first one
	 *	that registers here.
	 */
	if (preferred_console < 0) {
2149 2150 2151 2152 2153 2154 2155
		if (newcon->index < 0)
			newcon->index = 0;
		if (newcon->setup == NULL ||
		    newcon->setup(newcon, NULL) == 0) {
			newcon->flags |= CON_ENABLED;
			if (newcon->device) {
				newcon->flags |= CON_CONSDEV;
2156 2157
				preferred_console = 0;
			}
L
Linus Torvalds 已提交
2158 2159 2160 2161 2162 2163 2164
		}
	}

	/*
	 *	See if this console matches one we selected on
	 *	the command line.
	 */
J
Jesper Juhl 已提交
2165 2166
	for (i = 0; i < MAX_CMDLINECONSOLES && console_cmdline[i].name[0];
			i++) {
2167
		if (strcmp(console_cmdline[i].name, newcon->name) != 0)
L
Linus Torvalds 已提交
2168
			continue;
2169 2170
		if (newcon->index >= 0 &&
		    newcon->index != console_cmdline[i].index)
L
Linus Torvalds 已提交
2171
			continue;
2172 2173
		if (newcon->index < 0)
			newcon->index = console_cmdline[i].index;
2174 2175
#ifdef CONFIG_A11Y_BRAILLE_CONSOLE
		if (console_cmdline[i].brl_options) {
2176 2177
			newcon->flags |= CON_BRL;
			braille_register_console(newcon,
2178 2179 2180 2181 2182 2183
					console_cmdline[i].index,
					console_cmdline[i].options,
					console_cmdline[i].brl_options);
			return;
		}
#endif
2184 2185
		if (newcon->setup &&
		    newcon->setup(newcon, console_cmdline[i].options) != 0)
L
Linus Torvalds 已提交
2186
			break;
2187 2188
		newcon->flags |= CON_ENABLED;
		newcon->index = console_cmdline[i].index;
2189
		if (i == selected_console) {
2190
			newcon->flags |= CON_CONSDEV;
2191 2192
			preferred_console = selected_console;
		}
L
Linus Torvalds 已提交
2193 2194 2195
		break;
	}

2196
	if (!(newcon->flags & CON_ENABLED))
L
Linus Torvalds 已提交
2197 2198
		return;

2199 2200 2201 2202 2203 2204 2205
	/*
	 * If we have a bootconsole, and are switching to a real console,
	 * don't print everything out again, since when the boot console, and
	 * the real console are the same physical device, it's annoying to
	 * see the beginning boot messages twice
	 */
	if (bcon && ((newcon->flags & (CON_CONSDEV | CON_BOOT)) == CON_CONSDEV))
2206
		newcon->flags &= ~CON_PRINTBUFFER;
L
Linus Torvalds 已提交
2207 2208 2209 2210 2211

	/*
	 *	Put this console in the list - keep the
	 *	preferred driver at the head of the list.
	 */
2212
	console_lock();
2213 2214 2215 2216 2217
	if ((newcon->flags & CON_CONSDEV) || console_drivers == NULL) {
		newcon->next = console_drivers;
		console_drivers = newcon;
		if (newcon->next)
			newcon->next->flags &= ~CON_CONSDEV;
L
Linus Torvalds 已提交
2218
	} else {
2219 2220
		newcon->next = console_drivers->next;
		console_drivers->next = newcon;
L
Linus Torvalds 已提交
2221
	}
2222
	if (newcon->flags & CON_PRINTBUFFER) {
L
Linus Torvalds 已提交
2223
		/*
2224
		 * console_unlock(); will print out the buffered messages
L
Linus Torvalds 已提交
2225 2226
		 * for us.
		 */
2227
		raw_spin_lock_irqsave(&logbuf_lock, flags);
2228 2229
		console_seq = syslog_seq;
		console_idx = syslog_idx;
2230
		raw_spin_unlock_irqrestore(&logbuf_lock, flags);
2231 2232 2233 2234 2235 2236
		/*
		 * We're about to replay the log buffer.  Only do this to the
		 * just-registered console to avoid excessive message spam to
		 * the already-registered consoles.
		 */
		exclusive_console = newcon;
L
Linus Torvalds 已提交
2237
	}
2238
	console_unlock();
2239
	console_sysfs_notify();
2240 2241 2242 2243 2244 2245 2246 2247

	/*
	 * By unregistering the bootconsoles after we enable the real console
	 * we get the "console xxx enabled" message on all the consoles -
	 * boot consoles, real consoles, etc - this is to ensure that end
	 * users know there might be something in the kernel's log buffer that
	 * went to the bootconsole (that they do not see on the real console)
	 */
2248 2249 2250
	if (bcon &&
	    ((newcon->flags & (CON_CONSDEV | CON_BOOT)) == CON_CONSDEV) &&
	    !keep_bootcon) {
2251 2252 2253 2254 2255 2256 2257 2258 2259 2260 2261 2262 2263
		/* we need to iterate through twice, to make sure we print
		 * everything out, before we unregister the console(s)
		 */
		printk(KERN_INFO "console [%s%d] enabled, bootconsole disabled\n",
			newcon->name, newcon->index);
		for_each_console(bcon)
			if (bcon->flags & CON_BOOT)
				unregister_console(bcon);
	} else {
		printk(KERN_INFO "%sconsole [%s%d] enabled\n",
			(newcon->flags & CON_BOOT) ? "boot" : "" ,
			newcon->name, newcon->index);
	}
L
Linus Torvalds 已提交
2264 2265 2266
}
EXPORT_SYMBOL(register_console);

J
Jesper Juhl 已提交
2267
int unregister_console(struct console *console)
L
Linus Torvalds 已提交
2268
{
J
Jesper Juhl 已提交
2269
        struct console *a, *b;
L
Linus Torvalds 已提交
2270 2271
	int res = 1;

2272 2273 2274 2275 2276
#ifdef CONFIG_A11Y_BRAILLE_CONSOLE
	if (console->flags & CON_BRL)
		return braille_unregister_console(console);
#endif

2277
	console_lock();
L
Linus Torvalds 已提交
2278 2279 2280
	if (console_drivers == console) {
		console_drivers=console->next;
		res = 0;
2281
	} else if (console_drivers) {
L
Linus Torvalds 已提交
2282 2283 2284 2285 2286 2287
		for (a=console_drivers->next, b=console_drivers ;
		     a; b=a, a=b->next) {
			if (a == console) {
				b->next = a->next;
				res = 0;
				break;
J
Jesper Juhl 已提交
2288
			}
L
Linus Torvalds 已提交
2289 2290
		}
	}
J
Jesper Juhl 已提交
2291

2292
	/*
2293 2294
	 * If this isn't the last console and it has CON_CONSDEV set, we
	 * need to set it on the next preferred console.
L
Linus Torvalds 已提交
2295
	 */
2296
	if (console_drivers != NULL && console->flags & CON_CONSDEV)
2297
		console_drivers->flags |= CON_CONSDEV;
L
Linus Torvalds 已提交
2298

2299
	console_unlock();
2300
	console_sysfs_notify();
L
Linus Torvalds 已提交
2301 2302 2303
	return res;
}
EXPORT_SYMBOL(unregister_console);
M
Matt Mackall 已提交
2304

2305
static int __init printk_late_init(void)
2306
{
2307 2308 2309
	struct console *con;

	for_each_console(con) {
2310
		if (!keep_bootcon && con->flags & CON_BOOT) {
2311
			printk(KERN_INFO "turn off boot console %s%d\n",
2312
				con->name, con->index);
2313
			unregister_console(con);
2314
		}
2315
	}
2316
	hotcpu_notifier(console_cpu_notify, 0);
2317 2318
	return 0;
}
2319
late_initcall(printk_late_init);
2320

2321
#if defined CONFIG_PRINTK
D
Dave Young 已提交
2322

2323 2324 2325 2326 2327 2328 2329 2330 2331 2332 2333 2334 2335 2336 2337 2338 2339 2340 2341 2342
int printk_sched(const char *fmt, ...)
{
	unsigned long flags;
	va_list args;
	char *buf;
	int r;

	local_irq_save(flags);
	buf = __get_cpu_var(printk_sched_buf);

	va_start(args, fmt);
	r = vsnprintf(buf, PRINTK_BUF_SIZE, fmt, args);
	va_end(args);

	__this_cpu_or(printk_pending, PRINTK_PENDING_SCHED);
	local_irq_restore(flags);

	return r;
}

L
Linus Torvalds 已提交
2343 2344 2345
/*
 * printk rate limiting, lifted from the networking subsystem.
 *
2346 2347
 * This enforces a rate limit: not more than 10 kernel messages
 * every 5s to make a denial-of-service attack impossible.
L
Linus Torvalds 已提交
2348
 */
2349 2350
DEFINE_RATELIMIT_STATE(printk_ratelimit_state, 5 * HZ, 10);

2351
int __printk_ratelimit(const char *func)
L
Linus Torvalds 已提交
2352
{
2353
	return ___ratelimit(&printk_ratelimit_state, func);
L
Linus Torvalds 已提交
2354
}
2355
EXPORT_SYMBOL(__printk_ratelimit);
2356 2357 2358 2359 2360 2361 2362 2363 2364 2365 2366 2367 2368

/**
 * printk_timed_ratelimit - caller-controlled printk ratelimiting
 * @caller_jiffies: pointer to caller's state
 * @interval_msecs: minimum interval between prints
 *
 * printk_timed_ratelimit() returns true if more than @interval_msecs
 * milliseconds have elapsed since the last time printk_timed_ratelimit()
 * returned true.
 */
bool printk_timed_ratelimit(unsigned long *caller_jiffies,
			unsigned int interval_msecs)
{
2369 2370 2371 2372 2373
	if (*caller_jiffies == 0
			|| !time_in_range(jiffies, *caller_jiffies,
					*caller_jiffies
					+ msecs_to_jiffies(interval_msecs))) {
		*caller_jiffies = jiffies;
2374 2375 2376 2377 2378
		return true;
	}
	return false;
}
EXPORT_SYMBOL(printk_timed_ratelimit);
2379 2380 2381 2382 2383 2384

static DEFINE_SPINLOCK(dump_list_lock);
static LIST_HEAD(dump_list);

/**
 * kmsg_dump_register - register a kernel log dumper.
R
Randy Dunlap 已提交
2385
 * @dumper: pointer to the kmsg_dumper structure
2386 2387 2388 2389 2390 2391 2392 2393 2394 2395 2396 2397 2398 2399 2400 2401 2402 2403
 *
 * Adds a kernel log dumper to the system. The dump callback in the
 * structure will be called when the kernel oopses or panics and must be
 * set. Returns zero on success and %-EINVAL or %-EBUSY otherwise.
 */
int kmsg_dump_register(struct kmsg_dumper *dumper)
{
	unsigned long flags;
	int err = -EBUSY;

	/* The dump callback needs to be set */
	if (!dumper->dump)
		return -EINVAL;

	spin_lock_irqsave(&dump_list_lock, flags);
	/* Don't allow registering multiple times */
	if (!dumper->registered) {
		dumper->registered = 1;
2404
		list_add_tail_rcu(&dumper->list, &dump_list);
2405 2406 2407 2408 2409 2410 2411 2412 2413 2414
		err = 0;
	}
	spin_unlock_irqrestore(&dump_list_lock, flags);

	return err;
}
EXPORT_SYMBOL_GPL(kmsg_dump_register);

/**
 * kmsg_dump_unregister - unregister a kmsg dumper.
R
Randy Dunlap 已提交
2415
 * @dumper: pointer to the kmsg_dumper structure
2416 2417 2418 2419 2420 2421 2422 2423 2424 2425 2426 2427
 *
 * Removes a dump device from the system. Returns zero on success and
 * %-EINVAL otherwise.
 */
int kmsg_dump_unregister(struct kmsg_dumper *dumper)
{
	unsigned long flags;
	int err = -EINVAL;

	spin_lock_irqsave(&dump_list_lock, flags);
	if (dumper->registered) {
		dumper->registered = 0;
2428
		list_del_rcu(&dumper->list);
2429 2430 2431
		err = 0;
	}
	spin_unlock_irqrestore(&dump_list_lock, flags);
2432
	synchronize_rcu();
2433 2434 2435 2436 2437

	return err;
}
EXPORT_SYMBOL_GPL(kmsg_dump_unregister);

2438 2439 2440
static bool always_kmsg_dump;
module_param_named(always_kmsg_dump, always_kmsg_dump, bool, S_IRUGO | S_IWUSR);

2441 2442 2443 2444
/**
 * kmsg_dump - dump kernel log to kernel message dumpers.
 * @reason: the reason (oops, panic etc) for dumping
 *
2445 2446 2447
 * Call each of the registered dumper's dump() callback, which can
 * retrieve the kmsg records with kmsg_dump_get_line() or
 * kmsg_dump_get_buffer().
2448 2449 2450 2451 2452 2453
 */
void kmsg_dump(enum kmsg_dump_reason reason)
{
	struct kmsg_dumper *dumper;
	unsigned long flags;

2454 2455 2456
	if ((reason > KMSG_DUMP_OOPS) && !always_kmsg_dump)
		return;

2457 2458 2459 2460 2461 2462 2463 2464 2465 2466 2467 2468 2469 2470 2471 2472 2473 2474 2475 2476 2477 2478 2479 2480 2481 2482 2483 2484 2485 2486 2487 2488 2489 2490 2491 2492 2493 2494 2495 2496 2497 2498 2499 2500 2501 2502 2503 2504 2505 2506 2507
	rcu_read_lock();
	list_for_each_entry_rcu(dumper, &dump_list, list) {
		if (dumper->max_reason && reason > dumper->max_reason)
			continue;

		/* initialize iterator with data about the stored records */
		dumper->active = true;

		raw_spin_lock_irqsave(&logbuf_lock, flags);
		dumper->cur_seq = clear_seq;
		dumper->cur_idx = clear_idx;
		dumper->next_seq = log_next_seq;
		dumper->next_idx = log_next_idx;
		raw_spin_unlock_irqrestore(&logbuf_lock, flags);

		/* invoke dumper which will iterate over records */
		dumper->dump(dumper, reason);

		/* reset iterator */
		dumper->active = false;
	}
	rcu_read_unlock();
}

/**
 * kmsg_dump_get_line - retrieve one kmsg log line
 * @dumper: registered kmsg dumper
 * @syslog: include the "<4>" prefixes
 * @line: buffer to copy the line to
 * @size: maximum size of the buffer
 * @len: length of line placed into buffer
 *
 * Start at the beginning of the kmsg buffer, with the oldest kmsg
 * record, and copy one record into the provided buffer.
 *
 * Consecutive calls will return the next available record moving
 * towards the end of the buffer with the youngest messages.
 *
 * A return value of FALSE indicates that there are no more records to
 * read.
 */
bool kmsg_dump_get_line(struct kmsg_dumper *dumper, bool syslog,
			char *line, size_t size, size_t *len)
{
	unsigned long flags;
	struct log *msg;
	size_t l = 0;
	bool ret = false;

	if (!dumper->active)
		goto out;
2508

2509
	raw_spin_lock_irqsave(&logbuf_lock, flags);
2510 2511 2512 2513 2514
	if (dumper->cur_seq < log_first_seq) {
		/* messages are gone, move to first available one */
		dumper->cur_seq = log_first_seq;
		dumper->cur_idx = log_first_idx;
	}
2515

2516 2517 2518 2519 2520
	/* last entry */
	if (dumper->cur_seq >= log_next_seq) {
		raw_spin_unlock_irqrestore(&logbuf_lock, flags);
		goto out;
	}
2521

2522 2523 2524 2525 2526 2527 2528 2529 2530 2531 2532 2533 2534 2535 2536 2537 2538 2539 2540
	msg = log_from_idx(dumper->cur_idx);
	l = msg_print_text(msg, syslog,
			      line, size);

	dumper->cur_idx = log_next(dumper->cur_idx);
	dumper->cur_seq++;
	ret = true;
	raw_spin_unlock_irqrestore(&logbuf_lock, flags);
out:
	if (len)
		*len = l;
	return ret;
}
EXPORT_SYMBOL_GPL(kmsg_dump_get_line);

/**
 * kmsg_dump_get_buffer - copy kmsg log lines
 * @dumper: registered kmsg dumper
 * @syslog: include the "<4>" prefixes
R
Randy Dunlap 已提交
2541
 * @buf: buffer to copy the line to
2542 2543 2544 2545 2546 2547 2548 2549 2550 2551 2552 2553 2554 2555 2556 2557 2558 2559 2560 2561 2562 2563 2564 2565 2566 2567 2568 2569 2570 2571 2572 2573 2574 2575 2576 2577 2578 2579 2580 2581 2582 2583 2584 2585 2586 2587 2588 2589 2590 2591 2592 2593 2594 2595 2596 2597 2598
 * @size: maximum size of the buffer
 * @len: length of line placed into buffer
 *
 * Start at the end of the kmsg buffer and fill the provided buffer
 * with as many of the the *youngest* kmsg records that fit into it.
 * If the buffer is large enough, all available kmsg records will be
 * copied with a single call.
 *
 * Consecutive calls will fill the buffer with the next block of
 * available older records, not including the earlier retrieved ones.
 *
 * A return value of FALSE indicates that there are no more records to
 * read.
 */
bool kmsg_dump_get_buffer(struct kmsg_dumper *dumper, bool syslog,
			  char *buf, size_t size, size_t *len)
{
	unsigned long flags;
	u64 seq;
	u32 idx;
	u64 next_seq;
	u32 next_idx;
	size_t l = 0;
	bool ret = false;

	if (!dumper->active)
		goto out;

	raw_spin_lock_irqsave(&logbuf_lock, flags);
	if (dumper->cur_seq < log_first_seq) {
		/* messages are gone, move to first available one */
		dumper->cur_seq = log_first_seq;
		dumper->cur_idx = log_first_idx;
	}

	/* last entry */
	if (dumper->cur_seq >= dumper->next_seq) {
		raw_spin_unlock_irqrestore(&logbuf_lock, flags);
		goto out;
	}

	/* calculate length of entire buffer */
	seq = dumper->cur_seq;
	idx = dumper->cur_idx;
	while (seq < dumper->next_seq) {
		struct log *msg = log_from_idx(idx);

		l += msg_print_text(msg, true, NULL, 0);
		idx = log_next(idx);
		seq++;
	}

	/* move first record forward until length fits into the buffer */
	seq = dumper->cur_seq;
	idx = dumper->cur_idx;
	while (l > size && seq < dumper->next_seq) {
		struct log *msg = log_from_idx(idx);
2599

2600 2601 2602
		l -= msg_print_text(msg, true, NULL, 0);
		idx = log_next(idx);
		seq++;
2603
	}
2604 2605 2606 2607 2608 2609 2610 2611 2612 2613 2614 2615 2616 2617 2618 2619 2620 2621 2622

	/* last message in next interation */
	next_seq = seq;
	next_idx = idx;

	l = 0;
	while (seq < dumper->next_seq) {
		struct log *msg = log_from_idx(idx);

		l += msg_print_text(msg, syslog,
				    buf + l, size - l);

		idx = log_next(idx);
		seq++;
	}

	dumper->next_seq = next_seq;
	dumper->next_idx = next_idx;
	ret = true;
2623
	raw_spin_unlock_irqrestore(&logbuf_lock, flags);
2624 2625 2626 2627 2628 2629
out:
	if (len)
		*len = l;
	return ret;
}
EXPORT_SYMBOL_GPL(kmsg_dump_get_buffer);
2630

2631 2632 2633 2634 2635 2636 2637 2638 2639 2640 2641 2642 2643 2644 2645 2646 2647 2648
/**
 * kmsg_dump_rewind - reset the interator
 * @dumper: registered kmsg dumper
 *
 * Reset the dumper's iterator so that kmsg_dump_get_line() and
 * kmsg_dump_get_buffer() can be called again and used multiple
 * times within the same dumper.dump() callback.
 */
void kmsg_dump_rewind(struct kmsg_dumper *dumper)
{
	unsigned long flags;

	raw_spin_lock_irqsave(&logbuf_lock, flags);
	dumper->cur_seq = clear_seq;
	dumper->cur_idx = clear_idx;
	dumper->next_seq = log_next_seq;
	dumper->next_idx = log_next_idx;
	raw_spin_unlock_irqrestore(&logbuf_lock, flags);
2649
}
2650
EXPORT_SYMBOL_GPL(kmsg_dump_rewind);
2651
#endif