srv0start.c 54 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
/*****************************************************************************

Copyright (c) 1996, 2009, Innobase Oy. All Rights Reserved.
Copyright (c) 2008, Google Inc.

Portions of this file contain modifications contributed and copyrighted by
Google, Inc. Those modifications are gratefully acknowledged and are described
briefly in the InnoDB documentation. The contributions by Google are
incorporated with their permission, and subject to the conditions contained in
the file COPYING.Google.

This program is free software; you can redistribute it and/or modify it under
the terms of the GNU General Public License as published by the Free Software
Foundation; version 2 of the License.

This program is distributed in the hope that it will be useful, but WITHOUT
ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.

You should have received a copy of the GNU General Public License along with
this program; if not, write to the Free Software Foundation, Inc., 59 Temple
Place, Suite 330, Boston, MA 02111-1307 USA

*****************************************************************************/
Sergey Vojtovich's avatar
Sergey Vojtovich committed
25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50
/***********************************************************************

Copyright (c) 1995, 2009, Innobase Oy. All Rights Reserved.
Copyright (c) 2009, Percona Inc.

Portions of this file contain modifications contributed and copyrighted
by Percona Inc.. Those modifications are
gratefully acknowledged and are described briefly in the InnoDB
documentation. The contributions by Percona Inc. are incorporated with
their permission, and subject to the conditions contained in the file
COPYING.Percona.

This program is free software; you can redistribute it and/or modify it
under the terms of the GNU General Public License as published by the
Free Software Foundation; version 2 of the License.

This program is distributed in the hope that it will be useful, but
WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General
Public License for more details.

You should have received a copy of the GNU General Public License along
with this program; if not, write to the Free Software Foundation, Inc.,
59 Temple Place, Suite 330, Boston, MA 02111-1307 USA

***********************************************************************/
51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035

/********************************************************************//**
@file srv/srv0start.c
Starts the InnoDB database server

Created 2/16/1996 Heikki Tuuri
*************************************************************************/

#include "ut0mem.h"
#include "mem0mem.h"
#include "data0data.h"
#include "data0type.h"
#include "dict0dict.h"
#include "buf0buf.h"
#include "os0file.h"
#include "os0thread.h"
#include "fil0fil.h"
#include "fsp0fsp.h"
#include "rem0rec.h"
#include "mtr0mtr.h"
#include "log0log.h"
#include "log0recv.h"
#include "page0page.h"
#include "page0cur.h"
#include "trx0trx.h"
#include "trx0sys.h"
#include "btr0btr.h"
#include "btr0cur.h"
#include "rem0rec.h"
#include "ibuf0ibuf.h"
#include "srv0start.h"
#include "srv0srv.h"
#ifndef UNIV_HOTBACKUP
# include "os0proc.h"
# include "sync0sync.h"
# include "buf0flu.h"
# include "buf0rea.h"
# include "dict0boot.h"
# include "dict0load.h"
# include "que0que.h"
# include "usr0sess.h"
# include "lock0lock.h"
# include "trx0roll.h"
# include "trx0purge.h"
# include "lock0lock.h"
# include "pars0pars.h"
# include "btr0sea.h"
# include "rem0cmp.h"
# include "dict0crea.h"
# include "row0ins.h"
# include "row0sel.h"
# include "row0upd.h"
# include "row0row.h"
# include "row0mysql.h"
# include "btr0pcur.h"

/** Log sequence number immediately after startup */
UNIV_INTERN ib_uint64_t	srv_start_lsn;
/** Log sequence number at shutdown */
UNIV_INTERN ib_uint64_t	srv_shutdown_lsn;

#ifdef HAVE_DARWIN_THREADS
# include <sys/utsname.h>
/** TRUE if the F_FULLFSYNC option is available */
UNIV_INTERN ibool	srv_have_fullfsync = FALSE;
#endif

/** TRUE if a raw partition is in use */
UNIV_INTERN ibool	srv_start_raw_disk_in_use = FALSE;

/** TRUE if the server is being started, before rolling back any
incomplete transactions */
UNIV_INTERN ibool	srv_startup_is_before_trx_rollback_phase = FALSE;
/** TRUE if the server is being started */
UNIV_INTERN ibool	srv_is_being_started = FALSE;
/** TRUE if the server was successfully started */
UNIV_INTERN ibool	srv_was_started = FALSE;
/** TRUE if innobase_start_or_create_for_mysql() has been called */
static ibool	srv_start_has_been_called = FALSE;

/** At a shutdown this value climbs from SRV_SHUTDOWN_NONE to
SRV_SHUTDOWN_CLEANUP and then to SRV_SHUTDOWN_LAST_PHASE, and so on */
UNIV_INTERN enum srv_shutdown_state	srv_shutdown_state = SRV_SHUTDOWN_NONE;

/** Files comprising the system tablespace */
static os_file_t	files[1000];

/** Mutex protecting the ios count */
static mutex_t		ios_mutex;
/** Count of I/O operations in io_handler_thread() */
static ulint		ios;

/** io_handler_thread parameters for thread identification */
static ulint		n[SRV_MAX_N_IO_THREADS + 5];
/** io_handler_thread identifiers */
static os_thread_id_t	thread_ids[SRV_MAX_N_IO_THREADS + 5];

/** We use this mutex to test the return value of pthread_mutex_trylock
   on successful locking. HP-UX does NOT return 0, though Linux et al do. */
static os_fast_mutex_t	srv_os_test_mutex;

/** Name of srv_monitor_file */
static char*	srv_monitor_file_name;
#endif /* !UNIV_HOTBACKUP */

/** */
#define SRV_N_PENDING_IOS_PER_THREAD	OS_AIO_N_PENDING_IOS_PER_THREAD
#define SRV_MAX_N_PENDING_SYNC_IOS	100


/*********************************************************************//**
Convert a numeric string that optionally ends in G or M, to a number
containing megabytes.
@return	next character in string */
static
char*
srv_parse_megabytes(
/*================*/
	char*	str,	/*!< in: string containing a quantity in bytes */
	ulint*	megs)	/*!< out: the number in megabytes */
{
	char*	endp;
	ulint	size;

	size = strtoul(str, &endp, 10);

	str = endp;

	switch (*str) {
	case 'G': case 'g':
		size *= 1024;
		/* fall through */
	case 'M': case 'm':
		str++;
		break;
	default:
		size /= 1024 * 1024;
		break;
	}

	*megs = size;
	return(str);
}

/*********************************************************************//**
Reads the data files and their sizes from a character string given in
the .cnf file.
@return	TRUE if ok, FALSE on parse error */
UNIV_INTERN
ibool
srv_parse_data_file_paths_and_sizes(
/*================================*/
	char*	str)	/*!< in/out: the data file path string */
{
	char*	input_str;
	char*	path;
	ulint	size;
	ulint	i	= 0;

	srv_auto_extend_last_data_file = FALSE;
	srv_last_file_size_max = 0;
	srv_data_file_names = NULL;
	srv_data_file_sizes = NULL;
	srv_data_file_is_raw_partition = NULL;

	input_str = str;

	/* First calculate the number of data files and check syntax:
	path:size[M | G];path:size[M | G]... . Note that a Windows path may
	contain a drive name and a ':'. */

	while (*str != '\0') {
		path = str;

		while ((*str != ':' && *str != '\0')
		       || (*str == ':'
			   && (*(str + 1) == '\\' || *(str + 1) == '/'
			       || *(str + 1) == ':'))) {
			str++;
		}

		if (*str == '\0') {
			return(FALSE);
		}

		str++;

		str = srv_parse_megabytes(str, &size);

		if (0 == strncmp(str, ":autoextend",
				 (sizeof ":autoextend") - 1)) {

			str += (sizeof ":autoextend") - 1;

			if (0 == strncmp(str, ":max:",
					 (sizeof ":max:") - 1)) {

				str += (sizeof ":max:") - 1;

				str = srv_parse_megabytes(str, &size);
			}

			if (*str != '\0') {

				return(FALSE);
			}
		}

		if (strlen(str) >= 6
		    && *str == 'n'
		    && *(str + 1) == 'e'
		    && *(str + 2) == 'w') {
			str += 3;
		}

		if (*str == 'r' && *(str + 1) == 'a' && *(str + 2) == 'w') {
			str += 3;
		}

		if (size == 0) {
			return(FALSE);
		}

		i++;

		if (*str == ';') {
			str++;
		} else if (*str != '\0') {

			return(FALSE);
		}
	}

	if (i == 0) {
		/* If innodb_data_file_path was defined it must contain
		at least one data file definition */

		return(FALSE);
	}

	srv_data_file_names = malloc(i * sizeof *srv_data_file_names);
	srv_data_file_sizes = malloc(i * sizeof *srv_data_file_sizes);
	srv_data_file_is_raw_partition = malloc(
		i * sizeof *srv_data_file_is_raw_partition);

	srv_n_data_files = i;

	/* Then store the actual values to our arrays */

	str = input_str;
	i = 0;

	while (*str != '\0') {
		path = str;

		/* Note that we must step over the ':' in a Windows path;
		a Windows path normally looks like C:\ibdata\ibdata1:1G, but
		a Windows raw partition may have a specification like
		\\.\C::1Gnewraw or \\.\PHYSICALDRIVE2:1Gnewraw */

		while ((*str != ':' && *str != '\0')
		       || (*str == ':'
			   && (*(str + 1) == '\\' || *(str + 1) == '/'
			       || *(str + 1) == ':'))) {
			str++;
		}

		if (*str == ':') {
			/* Make path a null-terminated string */
			*str = '\0';
			str++;
		}

		str = srv_parse_megabytes(str, &size);

		srv_data_file_names[i] = path;
		srv_data_file_sizes[i] = size;

		if (0 == strncmp(str, ":autoextend",
				 (sizeof ":autoextend") - 1)) {

			srv_auto_extend_last_data_file = TRUE;

			str += (sizeof ":autoextend") - 1;

			if (0 == strncmp(str, ":max:",
					 (sizeof ":max:") - 1)) {

				str += (sizeof ":max:") - 1;

				str = srv_parse_megabytes(
					str, &srv_last_file_size_max);
			}

			if (*str != '\0') {

				return(FALSE);
			}
		}

		(srv_data_file_is_raw_partition)[i] = 0;

		if (strlen(str) >= 6
		    && *str == 'n'
		    && *(str + 1) == 'e'
		    && *(str + 2) == 'w') {
			str += 3;
			(srv_data_file_is_raw_partition)[i] = SRV_NEW_RAW;
		}

		if (*str == 'r' && *(str + 1) == 'a' && *(str + 2) == 'w') {
			str += 3;

			if ((srv_data_file_is_raw_partition)[i] == 0) {
				(srv_data_file_is_raw_partition)[i] = SRV_OLD_RAW;
			}
		}

		i++;

		if (*str == ';') {
			str++;
		}
	}

	return(TRUE);
}

/*********************************************************************//**
Reads log group home directories from a character string given in
the .cnf file.
@return	TRUE if ok, FALSE on parse error */
UNIV_INTERN
ibool
srv_parse_log_group_home_dirs(
/*==========================*/
	char*	str)	/*!< in/out: character string */
{
	char*	input_str;
	char*	path;
	ulint	i	= 0;

	srv_log_group_home_dirs = NULL;

	input_str = str;

	/* First calculate the number of directories and check syntax:
	path;path;... */

	while (*str != '\0') {
		path = str;

		while (*str != ';' && *str != '\0') {
			str++;
		}

		i++;

		if (*str == ';') {
			str++;
		} else if (*str != '\0') {

			return(FALSE);
		}
	}

	if (i != 1) {
		/* If innodb_log_group_home_dir was defined it must
		contain exactly one path definition under current MySQL */

		return(FALSE);
	}

	srv_log_group_home_dirs = malloc(i * sizeof *srv_log_group_home_dirs);

	/* Then store the actual values to our array */

	str = input_str;
	i = 0;

	while (*str != '\0') {
		path = str;

		while (*str != ';' && *str != '\0') {
			str++;
		}

		if (*str == ';') {
			*str = '\0';
			str++;
		}

		srv_log_group_home_dirs[i] = path;

		i++;
	}

	return(TRUE);
}

/*********************************************************************//**
Frees the memory allocated by srv_parse_data_file_paths_and_sizes()
and srv_parse_log_group_home_dirs(). */
UNIV_INTERN
void
srv_free_paths_and_sizes(void)
/*==========================*/
{
	free(srv_data_file_names);
	srv_data_file_names = NULL;
	free(srv_data_file_sizes);
	srv_data_file_sizes = NULL;
	free(srv_data_file_is_raw_partition);
	srv_data_file_is_raw_partition = NULL;
	free(srv_log_group_home_dirs);
	srv_log_group_home_dirs = NULL;
}

#ifndef UNIV_HOTBACKUP
/********************************************************************//**
I/o-handler thread function.
@return	OS_THREAD_DUMMY_RETURN */
static
os_thread_ret_t
io_handler_thread(
/*==============*/
	void*	arg)	/*!< in: pointer to the number of the segment in
			the aio array */
{
	ulint	segment;
	ulint	i;

	segment = *((ulint*)arg);

#ifdef UNIV_DEBUG_THREAD_CREATION
	fprintf(stderr, "Io handler thread %lu starts, id %lu\n", segment,
		os_thread_pf(os_thread_get_curr_id()));
#endif
	for (i = 0;; i++) {
		fil_aio_wait(segment);

		mutex_enter(&ios_mutex);
		ios++;
		mutex_exit(&ios_mutex);
	}

	/* We count the number of threads in os_thread_exit(). A created
	thread should always use that to exit and not use return() to exit.
	The thread actually never comes here because it is exited in an
	os_event_wait(). */

	os_thread_exit(NULL);

	OS_THREAD_DUMMY_RETURN;
}
#endif /* !UNIV_HOTBACKUP */

#ifdef __WIN__
#define SRV_PATH_SEPARATOR	'\\'
#else
#define SRV_PATH_SEPARATOR	'/'
#endif

/*********************************************************************//**
Normalizes a directory path for Windows: converts slashes to backslashes. */
UNIV_INTERN
void
srv_normalize_path_for_win(
/*=======================*/
	char*	str __attribute__((unused)))	/*!< in/out: null-terminated
						character string */
{
#ifdef __WIN__
	for (; *str; str++) {

		if (*str == '/') {
			*str = '\\';
		}
	}
#endif
}

/*********************************************************************//**
Adds a slash or a backslash to the end of a string if it is missing
and the string is not empty.
@return	string which has the separator if the string is not empty */
UNIV_INTERN
char*
srv_add_path_separator_if_needed(
/*=============================*/
	char*	str)	/*!< in: null-terminated character string */
{
	char*	out_str;
	ulint	len	= ut_strlen(str);

	if (len == 0 || str[len - 1] == SRV_PATH_SEPARATOR) {

		return(str);
	}

	out_str = ut_malloc(len + 2);
	memcpy(out_str, str, len);
	out_str[len] = SRV_PATH_SEPARATOR;
	out_str[len + 1] = 0;

	return(out_str);
}

#ifndef UNIV_HOTBACKUP
/*********************************************************************//**
Calculates the low 32 bits when a file size which is given as a number
database pages is converted to the number of bytes.
@return	low 32 bytes of file size when expressed in bytes */
static
ulint
srv_calc_low32(
/*===========*/
	ulint	file_size)	/*!< in: file size in database pages */
{
	return(0xFFFFFFFFUL & (file_size << UNIV_PAGE_SIZE_SHIFT));
}

/*********************************************************************//**
Calculates the high 32 bits when a file size which is given as a number
database pages is converted to the number of bytes.
@return	high 32 bytes of file size when expressed in bytes */
static
ulint
srv_calc_high32(
/*============*/
	ulint	file_size)	/*!< in: file size in database pages */
{
	return(file_size >> (32 - UNIV_PAGE_SIZE_SHIFT));
}

/*********************************************************************//**
Creates or opens the log files and closes them.
@return	DB_SUCCESS or error code */
static
ulint
open_or_create_log_file(
/*====================*/
	ibool	create_new_db,		/*!< in: TRUE if we should create a
					new database */
	ibool*	log_file_created,	/*!< out: TRUE if new log file
					created */
	ibool	log_file_has_been_opened,/*!< in: TRUE if a log file has been
					opened before: then it is an error
					to try to create another log file */
	ulint	k,			/*!< in: log group number */
	ulint	i)			/*!< in: log file number in group */
{
	ibool	ret;
	ulint	size;
	ulint	size_high;
	char	name[10000];

	UT_NOT_USED(create_new_db);

	*log_file_created = FALSE;

	srv_normalize_path_for_win(srv_log_group_home_dirs[k]);
	srv_log_group_home_dirs[k] = srv_add_path_separator_if_needed(
		srv_log_group_home_dirs[k]);

	ut_a(strlen(srv_log_group_home_dirs[k])
	     < (sizeof name) - 10 - sizeof "ib_logfile");
	sprintf(name, "%s%s%lu", srv_log_group_home_dirs[k],
		"ib_logfile", (ulong) i);

	files[i] = os_file_create(name, OS_FILE_CREATE, OS_FILE_NORMAL,
				  OS_LOG_FILE, &ret);
	if (ret == FALSE) {
		if (os_file_get_last_error(FALSE) != OS_FILE_ALREADY_EXISTS
#ifdef UNIV_AIX
		    /* AIX 5.1 after security patch ML7 may have errno set
		    to 0 here, which causes our function to return 100;
		    work around that AIX problem */
		    && os_file_get_last_error(FALSE) != 100
#endif
		    ) {
			fprintf(stderr,
				"InnoDB: Error in creating"
				" or opening %s\n", name);

			return(DB_ERROR);
		}

		files[i] = os_file_create(name, OS_FILE_OPEN, OS_FILE_AIO,
					  OS_LOG_FILE, &ret);
		if (!ret) {
			fprintf(stderr,
				"InnoDB: Error in opening %s\n", name);

			return(DB_ERROR);
		}

		ret = os_file_get_size(files[i], &size, &size_high);
		ut_a(ret);

		if (size != srv_calc_low32(srv_log_file_size)
		    || size_high != srv_calc_high32(srv_log_file_size)) {

			fprintf(stderr,
				"InnoDB: Error: log file %s is"
				" of different size %lu %lu bytes\n"
				"InnoDB: than specified in the .cnf"
				" file %lu %lu bytes!\n",
				name, (ulong) size_high, (ulong) size,
				(ulong) srv_calc_high32(srv_log_file_size),
				(ulong) srv_calc_low32(srv_log_file_size));

			return(DB_ERROR);
		}
	} else {
		*log_file_created = TRUE;

		ut_print_timestamp(stderr);

		fprintf(stderr,
			"  InnoDB: Log file %s did not exist:"
			" new to be created\n",
			name);
		if (log_file_has_been_opened) {

			return(DB_ERROR);
		}

		fprintf(stderr, "InnoDB: Setting log file %s size to %lu MB\n",
			name, (ulong) srv_log_file_size
			>> (20 - UNIV_PAGE_SIZE_SHIFT));

		fprintf(stderr,
			"InnoDB: Database physically writes the file"
			" full: wait...\n");

		ret = os_file_set_size(name, files[i],
				       srv_calc_low32(srv_log_file_size),
				       srv_calc_high32(srv_log_file_size));
		if (!ret) {
			fprintf(stderr,
				"InnoDB: Error in creating %s:"
				" probably out of disk space\n",
				name);

			return(DB_ERROR);
		}
	}

	ret = os_file_close(files[i]);
	ut_a(ret);

	if (i == 0) {
		/* Create in memory the file space object
		which is for this log group */

		fil_space_create(name,
				 2 * k + SRV_LOG_SPACE_FIRST_ID, 0, FIL_LOG);
	}

	ut_a(fil_validate());

	fil_node_create(name, srv_log_file_size,
			2 * k + SRV_LOG_SPACE_FIRST_ID, FALSE);
#ifdef UNIV_LOG_ARCHIVE
	/* If this is the first log group, create the file space object
	for archived logs.
	Under MySQL, no archiving ever done. */

	if (k == 0 && i == 0) {
		arch_space_id = 2 * k + 1 + SRV_LOG_SPACE_FIRST_ID;

		fil_space_create("arch_log_space", arch_space_id, 0, FIL_LOG);
	} else {
		arch_space_id = ULINT_UNDEFINED;
	}
#endif /* UNIV_LOG_ARCHIVE */
	if (i == 0) {
		log_group_init(k, srv_n_log_files,
			       srv_log_file_size * UNIV_PAGE_SIZE,
			       2 * k + SRV_LOG_SPACE_FIRST_ID,
			       SRV_LOG_SPACE_FIRST_ID + 1); /* dummy arch
							    space id */
	}

	return(DB_SUCCESS);
}

/*********************************************************************//**
Creates or opens database data files and closes them.
@return	DB_SUCCESS or error code */
static
ulint
open_or_create_data_files(
/*======================*/
	ibool*		create_new_db,	/*!< out: TRUE if new database should be
					created */
#ifdef UNIV_LOG_ARCHIVE
	ulint*		min_arch_log_no,/*!< out: min of archived log
					numbers in data files */
	ulint*		max_arch_log_no,/*!< out: max of archived log
					numbers in data files */
#endif /* UNIV_LOG_ARCHIVE */
	ib_uint64_t*	min_flushed_lsn,/*!< out: min of flushed lsn
					values in data files */
	ib_uint64_t*	max_flushed_lsn,/*!< out: max of flushed lsn
					values in data files */
	ulint*		sum_of_new_sizes)/*!< out: sum of sizes of the
					new files added */
{
	ibool	ret;
	ulint	i;
	ibool	one_opened	= FALSE;
	ibool	one_created	= FALSE;
	ulint	size;
	ulint	size_high;
	ulint	rounded_size_pages;
	char	name[10000];

	if (srv_n_data_files >= 1000) {
		fprintf(stderr, "InnoDB: can only have < 1000 data files\n"
			"InnoDB: you have defined %lu\n",
			(ulong) srv_n_data_files);
		return(DB_ERROR);
	}

	*sum_of_new_sizes = 0;

	*create_new_db = FALSE;

	srv_normalize_path_for_win(srv_data_home);
	srv_data_home = srv_add_path_separator_if_needed(srv_data_home);

	for (i = 0; i < srv_n_data_files; i++) {
		srv_normalize_path_for_win(srv_data_file_names[i]);

		ut_a(strlen(srv_data_home) + strlen(srv_data_file_names[i])
		     < (sizeof name) - 1);
		sprintf(name, "%s%s", srv_data_home, srv_data_file_names[i]);

		if (srv_data_file_is_raw_partition[i] == 0) {

			/* First we try to create the file: if it already
			exists, ret will get value FALSE */

			files[i] = os_file_create(name, OS_FILE_CREATE,
						  OS_FILE_NORMAL,
						  OS_DATA_FILE, &ret);

			if (ret == FALSE && os_file_get_last_error(FALSE)
			    != OS_FILE_ALREADY_EXISTS
#ifdef UNIV_AIX
			    /* AIX 5.1 after security patch ML7 may have
			    errno set to 0 here, which causes our function
			    to return 100; work around that AIX problem */
			    && os_file_get_last_error(FALSE) != 100
#endif
			    ) {
				fprintf(stderr,
					"InnoDB: Error in creating"
					" or opening %s\n",
					name);

				return(DB_ERROR);
			}
		} else if (srv_data_file_is_raw_partition[i] == SRV_NEW_RAW) {
			/* The partition is opened, not created; then it is
			written over */

			srv_start_raw_disk_in_use = TRUE;
			srv_created_new_raw = TRUE;

			files[i] = os_file_create(name, OS_FILE_OPEN_RAW,
						  OS_FILE_NORMAL,
						  OS_DATA_FILE, &ret);
			if (!ret) {
				fprintf(stderr,
					"InnoDB: Error in opening %s\n", name);

				return(DB_ERROR);
			}
		} else if (srv_data_file_is_raw_partition[i] == SRV_OLD_RAW) {
			srv_start_raw_disk_in_use = TRUE;

			ret = FALSE;
		} else {
			ut_a(0);
		}

		if (ret == FALSE) {
			/* We open the data file */

			if (one_created) {
				fprintf(stderr,
					"InnoDB: Error: data files can only"
					" be added at the end\n");
				fprintf(stderr,
					"InnoDB: of a tablespace, but"
					" data file %s existed beforehand.\n",
					name);
				return(DB_ERROR);
			}

			if (srv_data_file_is_raw_partition[i] == SRV_OLD_RAW) {
				files[i] = os_file_create(
					name, OS_FILE_OPEN_RAW,
					OS_FILE_NORMAL, OS_DATA_FILE, &ret);
			} else if (i == 0) {
				files[i] = os_file_create(
					name, OS_FILE_OPEN_RETRY,
					OS_FILE_NORMAL, OS_DATA_FILE, &ret);
			} else {
				files[i] = os_file_create(
					name, OS_FILE_OPEN, OS_FILE_NORMAL,
					OS_DATA_FILE, &ret);
			}

			if (!ret) {
				fprintf(stderr,
					"InnoDB: Error in opening %s\n", name);
				os_file_get_last_error(TRUE);

				return(DB_ERROR);
			}

			if (srv_data_file_is_raw_partition[i] == SRV_OLD_RAW) {

				goto skip_size_check;
			}

			ret = os_file_get_size(files[i], &size, &size_high);
			ut_a(ret);
			/* Round size downward to megabytes */

			rounded_size_pages
				= (size / (1024 * 1024) + 4096 * size_high)
					<< (20 - UNIV_PAGE_SIZE_SHIFT);

			if (i == srv_n_data_files - 1
			    && srv_auto_extend_last_data_file) {

				if (srv_data_file_sizes[i] > rounded_size_pages
				    || (srv_last_file_size_max > 0
					&& srv_last_file_size_max
					< rounded_size_pages)) {

					fprintf(stderr,
						"InnoDB: Error: auto-extending"
						" data file %s is"
						" of a different size\n"
						"InnoDB: %lu pages (rounded"
						" down to MB) than specified"
						" in the .cnf file:\n"
						"InnoDB: initial %lu pages,"
						" max %lu (relevant if"
						" non-zero) pages!\n",
						name,
						(ulong) rounded_size_pages,
						(ulong) srv_data_file_sizes[i],
						(ulong)
						srv_last_file_size_max);

					return(DB_ERROR);
				}

				srv_data_file_sizes[i] = rounded_size_pages;
			}

			if (rounded_size_pages != srv_data_file_sizes[i]) {

				fprintf(stderr,
					"InnoDB: Error: data file %s"
					" is of a different size\n"
					"InnoDB: %lu pages"
					" (rounded down to MB)\n"
					"InnoDB: than specified"
					" in the .cnf file %lu pages!\n",
					name,
					(ulong) rounded_size_pages,
					(ulong) srv_data_file_sizes[i]);

				return(DB_ERROR);
			}
skip_size_check:
			fil_read_flushed_lsn_and_arch_log_no(
				files[i], one_opened,
#ifdef UNIV_LOG_ARCHIVE
				min_arch_log_no, max_arch_log_no,
#endif /* UNIV_LOG_ARCHIVE */
				min_flushed_lsn, max_flushed_lsn);
			one_opened = TRUE;
		} else {
			/* We created the data file and now write it full of
			zeros */

			one_created = TRUE;

			if (i > 0) {
				ut_print_timestamp(stderr);
				fprintf(stderr,
					"  InnoDB: Data file %s did not"
					" exist: new to be created\n",
					name);
			} else {
				fprintf(stderr,
					"InnoDB: The first specified"
					" data file %s did not exist:\n"
					"InnoDB: a new database"
					" to be created!\n", name);
				*create_new_db = TRUE;
			}

			ut_print_timestamp(stderr);
			fprintf(stderr,
				"  InnoDB: Setting file %s size to %lu MB\n",
				name,
				(ulong) (srv_data_file_sizes[i]
					 >> (20 - UNIV_PAGE_SIZE_SHIFT)));

			fprintf(stderr,
				"InnoDB: Database physically writes the"
				" file full: wait...\n");

			ret = os_file_set_size(
				name, files[i],
				srv_calc_low32(srv_data_file_sizes[i]),
				srv_calc_high32(srv_data_file_sizes[i]));

			if (!ret) {
				fprintf(stderr,
					"InnoDB: Error in creating %s:"
					" probably out of disk space\n", name);

				return(DB_ERROR);
			}

			*sum_of_new_sizes = *sum_of_new_sizes
				+ srv_data_file_sizes[i];
		}

		ret = os_file_close(files[i]);
		ut_a(ret);

		if (i == 0) {
			fil_space_create(name, 0, 0, FIL_TABLESPACE);
		}

		ut_a(fil_validate());

		fil_node_create(name, srv_data_file_sizes[i], 0,
				srv_data_file_is_raw_partition[i] != 0);
	}

	ios = 0;

	mutex_create(&ios_mutex, SYNC_NO_ORDER_CHECK);

	return(DB_SUCCESS);
}

/****************************************************************//**
Starts InnoDB and creates a new database if database files
are not found and the user wants.
@return	DB_SUCCESS or error code */
UNIV_INTERN
int
innobase_start_or_create_for_mysql(void)
/*====================================*/
{
	buf_pool_t*	ret;
	ibool		create_new_db;
	ibool		log_file_created;
	ibool		log_created	= FALSE;
	ibool		log_opened	= FALSE;
	ib_uint64_t	min_flushed_lsn;
	ib_uint64_t	max_flushed_lsn;
#ifdef UNIV_LOG_ARCHIVE
	ulint		min_arch_log_no;
	ulint		max_arch_log_no;
#endif /* UNIV_LOG_ARCHIVE */
	ulint		sum_of_new_sizes;
	ulint		sum_of_data_file_sizes;
	ulint		tablespace_size_in_header;
	ulint		err;
	ulint		i;
Sergey Vojtovich's avatar
Sergey Vojtovich committed
1036
	ulint		io_limit;
1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164
	my_bool		srv_file_per_table_original_value
		= srv_file_per_table;
	mtr_t		mtr;
#ifdef HAVE_DARWIN_THREADS
# ifdef F_FULLFSYNC
	/* This executable has been compiled on Mac OS X 10.3 or later.
	Assume that F_FULLFSYNC is available at run-time. */
	srv_have_fullfsync = TRUE;
# else /* F_FULLFSYNC */
	/* This executable has been compiled on Mac OS X 10.2
	or earlier.  Determine if the executable is running
	on Mac OS X 10.3 or later. */
	struct utsname utsname;
	if (uname(&utsname)) {
		fputs("InnoDB: cannot determine Mac OS X version!\n", stderr);
	} else {
		srv_have_fullfsync = strcmp(utsname.release, "7.") >= 0;
	}
	if (!srv_have_fullfsync) {
		fputs("InnoDB: On Mac OS X, fsync() may be"
		      " broken on internal drives,\n"
		      "InnoDB: making transactions unsafe!\n", stderr);
	}
# endif /* F_FULLFSYNC */
#endif /* HAVE_DARWIN_THREADS */

	if (sizeof(ulint) != sizeof(void*)) {
		fprintf(stderr,
			"InnoDB: Error: size of InnoDB's ulint is %lu,"
			" but size of void* is %lu.\n"
			"InnoDB: The sizes should be the same"
			" so that on a 64-bit platform you can\n"
			"InnoDB: allocate more than 4 GB of memory.",
			(ulong)sizeof(ulint), (ulong)sizeof(void*));
	}

	/* System tables are created in tablespace 0.  Thus, we must
	temporarily clear srv_file_per_table.  This is ok, because the
	server will not accept connections (which could modify
	innodb_file_per_table) until this function has returned. */
	srv_file_per_table = FALSE;
#ifdef UNIV_DEBUG
	fprintf(stderr,
		"InnoDB: !!!!!!!! UNIV_DEBUG switched on !!!!!!!!!\n");
#endif

#ifdef UNIV_IBUF_DEBUG
	fprintf(stderr,
		"InnoDB: !!!!!!!! UNIV_IBUF_DEBUG switched on !!!!!!!!!\n"
		"InnoDB: Crash recovery will fail with UNIV_IBUF_DEBUG\n");
#endif

#ifdef UNIV_SYNC_DEBUG
	fprintf(stderr,
		"InnoDB: !!!!!!!! UNIV_SYNC_DEBUG switched on !!!!!!!!!\n");
#endif

#ifdef UNIV_SEARCH_DEBUG
	fprintf(stderr,
		"InnoDB: !!!!!!!! UNIV_SEARCH_DEBUG switched on !!!!!!!!!\n");
#endif

#ifdef UNIV_MEM_DEBUG
	fprintf(stderr,
		"InnoDB: !!!!!!!! UNIV_MEM_DEBUG switched on !!!!!!!!!\n");
#endif

	if (UNIV_LIKELY(srv_use_sys_malloc)) {
		fprintf(stderr,
			"InnoDB: The InnoDB memory heap is disabled\n");
	}

#ifdef HAVE_GCC_ATOMIC_BUILTINS
# ifdef INNODB_RW_LOCKS_USE_ATOMICS
	fprintf(stderr,
		"InnoDB: Mutexes and rw_locks use GCC atomic builtins.\n");
# else /* INNODB_RW_LOCKS_USE_ATOMICS */
	fprintf(stderr,
		"InnoDB: Mutexes use GCC atomic builtins, rw_locks do not.\n");
# endif /* INNODB_RW_LOCKS_USE_ATOMICS */
#elif defined(HAVE_SOLARIS_ATOMICS)
# ifdef INNODB_RW_LOCKS_USE_ATOMICS
	fprintf(stderr,
		"InnoDB: Mutexes and rw_locks use Solaris atomic functions.\n");
# else
	fprintf(stderr,
		"InnoDB: Mutexes use Solaris atomic functions.\n");
# endif /* INNODB_RW_LOCKS_USE_ATOMICS */
#elif HAVE_WINDOWS_ATOMICS
# ifdef INNODB_RW_LOCKS_USE_ATOMICS
	fprintf(stderr,
		"InnoDB: Mutexes and rw_locks use Windows interlocked functions.\n");
# else
	fprintf(stderr,
		"InnoDB: Mutexes use Windows interlocked functions.\n");
# endif /* INNODB_RW_LOCKS_USE_ATOMICS */
#else /* HAVE_GCC_ATOMIC_BUILTINS */
	fprintf(stderr,
		"InnoDB: Neither mutexes nor rw_locks use GCC atomic builtins.\n");
#endif /* HAVE_GCC_ATOMIC_BUILTINS */

	/* Since InnoDB does not currently clean up all its internal data
	structures in MySQL Embedded Server Library server_end(), we
	print an error message if someone tries to start up InnoDB a
	second time during the process lifetime. */

	if (srv_start_has_been_called) {
		fprintf(stderr,
			"InnoDB: Error:startup called second time"
			" during the process lifetime.\n"
			"InnoDB: In the MySQL Embedded Server Library"
			" you cannot call server_init()\n"
			"InnoDB: more than once during"
			" the process lifetime.\n");
	}

	srv_start_has_been_called = TRUE;

#ifdef UNIV_DEBUG
	log_do_write = TRUE;
#endif /* UNIV_DEBUG */
	/*	yydebug = TRUE; */

	srv_is_being_started = TRUE;
	srv_startup_is_before_trx_rollback_phase = TRUE;
	os_aio_use_native_aio = FALSE;

#ifdef __WIN__
Sergey Vojtovich's avatar
Sergey Vojtovich committed
1165 1166 1167 1168
	switch (os_get_os_version()) {
	case OS_WIN95:
	case OS_WIN31:
	case OS_WINNT:
1169 1170 1171 1172 1173 1174
		/* On Win 95, 98, ME, Win32 subsystem for Windows 3.1,
		and NT use simulated aio. In NT Windows provides async i/o,
		but when run in conjunction with InnoDB Hot Backup, it seemed
		to corrupt the data files. */

		os_aio_use_native_aio = FALSE;
Sergey Vojtovich's avatar
Sergey Vojtovich committed
1175 1176
		break;
	default:
1177 1178
		/* On Win 2000 and XP use async i/o */
		os_aio_use_native_aio = TRUE;
Sergey Vojtovich's avatar
Sergey Vojtovich committed
1179
		break;
1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296
	}
#endif
	if (srv_file_flush_method_str == NULL) {
		/* These are the default options */

		srv_unix_file_flush_method = SRV_UNIX_FSYNC;

		srv_win_file_flush_method = SRV_WIN_IO_UNBUFFERED;
#ifndef __WIN__
	} else if (0 == ut_strcmp(srv_file_flush_method_str, "fsync")) {
		srv_unix_file_flush_method = SRV_UNIX_FSYNC;

	} else if (0 == ut_strcmp(srv_file_flush_method_str, "O_DSYNC")) {
		srv_unix_file_flush_method = SRV_UNIX_O_DSYNC;

	} else if (0 == ut_strcmp(srv_file_flush_method_str, "O_DIRECT")) {
		srv_unix_file_flush_method = SRV_UNIX_O_DIRECT;

	} else if (0 == ut_strcmp(srv_file_flush_method_str, "littlesync")) {
		srv_unix_file_flush_method = SRV_UNIX_LITTLESYNC;

	} else if (0 == ut_strcmp(srv_file_flush_method_str, "nosync")) {
		srv_unix_file_flush_method = SRV_UNIX_NOSYNC;
#else
	} else if (0 == ut_strcmp(srv_file_flush_method_str, "normal")) {
		srv_win_file_flush_method = SRV_WIN_IO_NORMAL;
		os_aio_use_native_aio = FALSE;

	} else if (0 == ut_strcmp(srv_file_flush_method_str, "unbuffered")) {
		srv_win_file_flush_method = SRV_WIN_IO_UNBUFFERED;
		os_aio_use_native_aio = FALSE;

	} else if (0 == ut_strcmp(srv_file_flush_method_str,
				  "async_unbuffered")) {
		srv_win_file_flush_method = SRV_WIN_IO_UNBUFFERED;
#endif
	} else {
		fprintf(stderr,
			"InnoDB: Unrecognized value %s for"
			" innodb_flush_method\n",
			srv_file_flush_method_str);
		return(DB_ERROR);
	}

	/* Note that the call srv_boot() also changes the values of
	some variables to the units used by InnoDB internally */

	/* Set the maximum number of threads which can wait for a semaphore
	inside InnoDB: this is the 'sync wait array' size, as well as the
	maximum number of threads that can wait in the 'srv_conc array' for
	their time to enter InnoDB. */

#if defined(__NETWARE__)

	/* Create less event semaphores because Win 98/ME had
	difficulty creating 40000 event semaphores.  Comment from
	Novell, Inc.: also, these just take a lot of memory on
	NetWare. */
	srv_max_n_threads = 1000;
#else
	if (srv_buf_pool_size >= 1000 * 1024 * 1024) {
		/* If buffer pool is less than 1000 MB,
		assume fewer threads. */
		srv_max_n_threads = 50000;

	} else if (srv_buf_pool_size >= 8 * 1024 * 1024) {

		srv_max_n_threads = 10000;
	} else {
		srv_max_n_threads = 1000;	/* saves several MB of memory,
						especially in 64-bit
						computers */
	}
#endif
	err = srv_boot();

	if (err != DB_SUCCESS) {

		return((int) err);
	}

	mutex_create(&srv_monitor_file_mutex, SYNC_NO_ORDER_CHECK);

	if (srv_innodb_status) {
		srv_monitor_file_name = mem_alloc(
			strlen(fil_path_to_mysql_datadir)
			+ 20 + sizeof "/innodb_status.");
		sprintf(srv_monitor_file_name, "%s/innodb_status.%lu",
			fil_path_to_mysql_datadir, os_proc_get_number());
		srv_monitor_file = fopen(srv_monitor_file_name, "w+");
		if (!srv_monitor_file) {
			fprintf(stderr, "InnoDB: unable to create %s: %s\n",
				srv_monitor_file_name, strerror(errno));
			return(DB_ERROR);
		}
	} else {
		srv_monitor_file_name = NULL;
		srv_monitor_file = os_file_create_tmpfile();
		if (!srv_monitor_file) {
			return(DB_ERROR);
		}
	}

	mutex_create(&srv_dict_tmpfile_mutex, SYNC_DICT_OPERATION);

	srv_dict_tmpfile = os_file_create_tmpfile();
	if (!srv_dict_tmpfile) {
		return(DB_ERROR);
	}

	mutex_create(&srv_misc_tmpfile_mutex, SYNC_ANY_LATCH);

	srv_misc_tmpfile = os_file_create_tmpfile();
	if (!srv_misc_tmpfile) {
		return(DB_ERROR);
	}

Sergey Vojtovich's avatar
Sergey Vojtovich committed
1297 1298 1299 1300 1301 1302 1303 1304
	/* If user has set the value of innodb_file_io_threads then
	we'll emit a message telling the user that this parameter
	is now deprecated. */
	if (srv_n_file_io_threads != 4) {
		fprintf(stderr, "InnoDB: Warning:"
			" innodb_file_io_threads is deprecated."
			" Please use innodb_read_io_threads and"
			" innodb_write_io_threads instead\n");
1305 1306
	}

Sergey Vojtovich's avatar
Sergey Vojtovich committed
1307 1308 1309 1310 1311
	/* Now overwrite the value on srv_n_file_io_threads */
	srv_n_file_io_threads = 2 + srv_n_read_io_threads
				+ srv_n_write_io_threads;

	ut_a(srv_n_file_io_threads <= SRV_MAX_N_IO_THREADS);
1312

Sergey Vojtovich's avatar
Sergey Vojtovich committed
1313 1314 1315 1316
	/* TODO: Investigate if SRV_N_PENDING_IOS_PER_THREAD (32) limit
	still applies to windows. */
	if (!os_aio_use_native_aio) {
		io_limit = 8 * SRV_N_PENDING_IOS_PER_THREAD;
1317
	} else {
Sergey Vojtovich's avatar
Sergey Vojtovich committed
1318
		io_limit = SRV_N_PENDING_IOS_PER_THREAD;
1319 1320
	}

Sergey Vojtovich's avatar
Sergey Vojtovich committed
1321 1322 1323 1324 1325
	os_aio_init(io_limit,
		    srv_n_read_io_threads,
		    srv_n_write_io_threads,
		    SRV_MAX_N_PENDING_SYNC_IOS);

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 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 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 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 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 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096
	fil_init(srv_file_per_table ? 50000 : 5000,
		 srv_max_n_open_files);

	ret = buf_pool_init();

	if (ret == NULL) {
		fprintf(stderr,
			"InnoDB: Fatal error: cannot allocate the memory"
			" for the buffer pool\n");

		return(DB_ERROR);
	}

#ifdef UNIV_DEBUG
	/* We have observed deadlocks with a 5MB buffer pool but
	the actual lower limit could very well be a little higher. */

	if (srv_buf_pool_size <= 5 * 1024 * 1024) {

		fprintf(stderr, "InnoDB: Warning: Small buffer pool size "
			"(%luM), the flst_validate() debug function "
			"can cause a deadlock if the buffer pool fills up.\n",
			srv_buf_pool_size / 1024 / 1024);
	}
#endif

	fsp_init();
	log_init();

	lock_sys_create(srv_lock_table_size);

	/* Create i/o-handler threads: */

	for (i = 0; i < srv_n_file_io_threads; i++) {
		n[i] = i;

		os_thread_create(io_handler_thread, n + i, thread_ids + i);
	}

#ifdef UNIV_LOG_ARCHIVE
	if (0 != ut_strcmp(srv_log_group_home_dirs[0], srv_arch_dir)) {
		fprintf(stderr,
			"InnoDB: Error: you must set the log group"
			" home dir in my.cnf the\n"
			"InnoDB: same as log arch dir.\n");

		return(DB_ERROR);
	}
#endif /* UNIV_LOG_ARCHIVE */

	if (srv_n_log_files * srv_log_file_size >= 262144) {
		fprintf(stderr,
			"InnoDB: Error: combined size of log files"
			" must be < 4 GB\n");

		return(DB_ERROR);
	}

	sum_of_new_sizes = 0;

	for (i = 0; i < srv_n_data_files; i++) {
#ifndef __WIN__
		if (sizeof(off_t) < 5 && srv_data_file_sizes[i] >= 262144) {
			fprintf(stderr,
				"InnoDB: Error: file size must be < 4 GB"
				" with this MySQL binary\n"
				"InnoDB: and operating system combination,"
				" in some OS's < 2 GB\n");

			return(DB_ERROR);
		}
#endif
		sum_of_new_sizes += srv_data_file_sizes[i];
	}

	if (sum_of_new_sizes < 640) {
		fprintf(stderr,
			"InnoDB: Error: tablespace size must be"
			" at least 10 MB\n");

		return(DB_ERROR);
	}

	err = open_or_create_data_files(&create_new_db,
#ifdef UNIV_LOG_ARCHIVE
					&min_arch_log_no, &max_arch_log_no,
#endif /* UNIV_LOG_ARCHIVE */
					&min_flushed_lsn, &max_flushed_lsn,
					&sum_of_new_sizes);
	if (err != DB_SUCCESS) {
		fprintf(stderr,
			"InnoDB: Could not open or create data files.\n"
			"InnoDB: If you tried to add new data files,"
			" and it failed here,\n"
			"InnoDB: you should now edit innodb_data_file_path"
			" in my.cnf back\n"
			"InnoDB: to what it was, and remove the"
			" new ibdata files InnoDB created\n"
			"InnoDB: in this failed attempt. InnoDB only wrote"
			" those files full of\n"
			"InnoDB: zeros, but did not yet use them in any way."
			" But be careful: do not\n"
			"InnoDB: remove old data files"
			" which contain your precious data!\n");

		return((int) err);
	}

#ifdef UNIV_LOG_ARCHIVE
	srv_normalize_path_for_win(srv_arch_dir);
	srv_arch_dir = srv_add_path_separator_if_needed(srv_arch_dir);
#endif /* UNIV_LOG_ARCHIVE */

	for (i = 0; i < srv_n_log_files; i++) {
		err = open_or_create_log_file(create_new_db, &log_file_created,
					      log_opened, 0, i);
		if (err != DB_SUCCESS) {

			return((int) err);
		}

		if (log_file_created) {
			log_created = TRUE;
		} else {
			log_opened = TRUE;
		}
		if ((log_opened && create_new_db)
		    || (log_opened && log_created)) {
			fprintf(stderr,
				"InnoDB: Error: all log files must be"
				" created at the same time.\n"
				"InnoDB: All log files must be"
				" created also in database creation.\n"
				"InnoDB: If you want bigger or smaller"
				" log files, shut down the\n"
				"InnoDB: database and make sure there"
				" were no errors in shutdown.\n"
				"InnoDB: Then delete the existing log files."
				" Edit the .cnf file\n"
				"InnoDB: and start the database again.\n");

			return(DB_ERROR);
		}
	}

	/* Open all log files and data files in the system tablespace: we
	keep them open until database shutdown */

	fil_open_log_and_system_tablespace_files();

	if (log_created && !create_new_db
#ifdef UNIV_LOG_ARCHIVE
	    && !srv_archive_recovery
#endif /* UNIV_LOG_ARCHIVE */
	    ) {
		if (max_flushed_lsn != min_flushed_lsn
#ifdef UNIV_LOG_ARCHIVE
		    || max_arch_log_no != min_arch_log_no
#endif /* UNIV_LOG_ARCHIVE */
		    ) {
			fprintf(stderr,
				"InnoDB: Cannot initialize created"
				" log files because\n"
				"InnoDB: data files were not in sync"
				" with each other\n"
				"InnoDB: or the data files are corrupt.\n");

			return(DB_ERROR);
		}

		if (max_flushed_lsn < (ib_uint64_t) 1000) {
			fprintf(stderr,
				"InnoDB: Cannot initialize created"
				" log files because\n"
				"InnoDB: data files are corrupt,"
				" or new data files were\n"
				"InnoDB: created when the database"
				" was started previous\n"
				"InnoDB: time but the database"
				" was not shut down\n"
				"InnoDB: normally after that.\n");

			return(DB_ERROR);
		}

		mutex_enter(&(log_sys->mutex));

#ifdef UNIV_LOG_ARCHIVE
		/* Do not + 1 arch_log_no because we do not use log
		archiving */
		recv_reset_logs(max_flushed_lsn, max_arch_log_no, TRUE);
#else
		recv_reset_logs(max_flushed_lsn, TRUE);
#endif /* UNIV_LOG_ARCHIVE */

		mutex_exit(&(log_sys->mutex));
	}

	trx_sys_file_format_init();

	if (create_new_db) {
		mtr_start(&mtr);
		fsp_header_init(0, sum_of_new_sizes, &mtr);

		mtr_commit(&mtr);

		trx_sys_create();
		dict_create();
		srv_startup_is_before_trx_rollback_phase = FALSE;

#ifdef UNIV_LOG_ARCHIVE
	} else if (srv_archive_recovery) {
		fprintf(stderr,
			"InnoDB: Starting archive"
			" recovery from a backup...\n");
		err = recv_recovery_from_archive_start(
			min_flushed_lsn, srv_archive_recovery_limit_lsn,
			min_arch_log_no);
		if (err != DB_SUCCESS) {

			return(DB_ERROR);
		}
		/* Since ibuf init is in dict_boot, and ibuf is needed
		in any disk i/o, first call dict_boot */

		dict_boot();
		trx_sys_init_at_db_start();
		srv_startup_is_before_trx_rollback_phase = FALSE;

		/* Initialize the fsp free limit global variable in the log
		system */
		fsp_header_get_free_limit();

		recv_recovery_from_archive_finish();
#endif /* UNIV_LOG_ARCHIVE */
	} else {

		/* Check if we support the max format that is stamped
		on the system tablespace. 
		Note:  We are NOT allowed to make any modifications to
		the TRX_SYS_PAGE_NO page before recovery  because this
		page also contains the max_trx_id etc. important system
		variables that are required for recovery.  We need to
		ensure that we return the system to a state where normal
		recovery is guaranteed to work. We do this by
		invalidating the buffer cache, this will force the
		reread of the page and restoration to its last known
		consistent state, this is REQUIRED for the recovery
		process to work. */
		err = trx_sys_file_format_max_check(
			srv_check_file_format_at_startup);

		if (err != DB_SUCCESS) {
			return(err);
		}

		/* Invalidate the buffer pool to ensure that we reread
		the page that we read above, during recovery.
		Note that this is not as heavy weight as it seems. At
		this point there will be only ONE page in the buf_LRU
		and there must be no page in the buf_flush list. */
		buf_pool_invalidate();

		/* We always try to do a recovery, even if the database had
		been shut down normally: this is the normal startup path */

		err = recv_recovery_from_checkpoint_start(LOG_CHECKPOINT,
							  IB_ULONGLONG_MAX,
							  min_flushed_lsn,
							  max_flushed_lsn);
		if (err != DB_SUCCESS) {

			return(DB_ERROR);
		}

		/* Since the insert buffer init is in dict_boot, and the
		insert buffer is needed in any disk i/o, first we call
		dict_boot(). Note that trx_sys_init_at_db_start() only needs
		to access space 0, and the insert buffer at this stage already
		works for space 0. */

		dict_boot();
		trx_sys_init_at_db_start();

		if (srv_force_recovery < SRV_FORCE_NO_IBUF_MERGE) {
			/* The following call is necessary for the insert
			buffer to work with multiple tablespaces. We must
			know the mapping between space id's and .ibd file
			names.

			In a crash recovery, we check that the info in data
			dictionary is consistent with what we already know
			about space id's from the call of
			fil_load_single_table_tablespaces().

			In a normal startup, we create the space objects for
			every table in the InnoDB data dictionary that has
			an .ibd file.

			We also determine the maximum tablespace id used.

			TODO: We may have incomplete transactions in the
			data dictionary tables. Does that harm the scanning of
			the data dictionary below? */

			dict_check_tablespaces_and_store_max_id(
				recv_needed_recovery);
		}

		srv_startup_is_before_trx_rollback_phase = FALSE;

		/* Initialize the fsp free limit global variable in the log
		system */
		fsp_header_get_free_limit();

		/* recv_recovery_from_checkpoint_finish needs trx lists which
		are initialized in trx_sys_init_at_db_start(). */

		recv_recovery_from_checkpoint_finish();

		/* It is possible that file_format tag has never
		been set. In this case we initialize it to minimum
		value.  Important to note that we can do it ONLY after
		we have finished the recovery process so that the
		image of TRX_SYS_PAGE_NO is not stale. */
		trx_sys_file_format_tag_init();
	}

	if (!create_new_db && sum_of_new_sizes > 0) {
		/* New data file(s) were added */
		mtr_start(&mtr);

		fsp_header_inc_size(0, sum_of_new_sizes, &mtr);

		mtr_commit(&mtr);

		/* Immediately write the log record about increased tablespace
		size to disk, so that it is durable even if mysqld would crash
		quickly */

		log_buffer_flush_to_disk();
	}

#ifdef UNIV_LOG_ARCHIVE
	/* Archiving is always off under MySQL */
	if (!srv_log_archive_on) {
		ut_a(DB_SUCCESS == log_archive_noarchivelog());
	} else {
		mutex_enter(&(log_sys->mutex));

		start_archive = FALSE;

		if (log_sys->archiving_state == LOG_ARCH_OFF) {
			start_archive = TRUE;
		}

		mutex_exit(&(log_sys->mutex));

		if (start_archive) {
			ut_a(DB_SUCCESS == log_archive_archivelog());
		}
	}
#endif /* UNIV_LOG_ARCHIVE */

	/* fprintf(stderr, "Max allowed record size %lu\n",
	page_get_free_space_of_empty() / 2); */

	/* Create the thread which watches the timeouts for lock waits
	and prints InnoDB monitor info */

	os_thread_create(&srv_lock_timeout_and_monitor_thread, NULL,
			 thread_ids + 2 + SRV_MAX_N_IO_THREADS);

	/* Create the thread which warns of long semaphore waits */
	os_thread_create(&srv_error_monitor_thread, NULL,
			 thread_ids + 3 + SRV_MAX_N_IO_THREADS);
	srv_is_being_started = FALSE;

	if (trx_doublewrite == NULL) {
		/* Create the doublewrite buffer to a new tablespace */

		trx_sys_create_doublewrite_buf();
	}

	err = dict_create_or_check_foreign_constraint_tables();

	if (err != DB_SUCCESS) {
		return((int)DB_ERROR);
	}

	/* Create the master thread which does purge and other utility
	operations */

	os_thread_create(&srv_master_thread, NULL, thread_ids
			 + (1 + SRV_MAX_N_IO_THREADS));
#ifdef UNIV_DEBUG
	/* buf_debug_prints = TRUE; */
#endif /* UNIV_DEBUG */
	sum_of_data_file_sizes = 0;

	for (i = 0; i < srv_n_data_files; i++) {
		sum_of_data_file_sizes += srv_data_file_sizes[i];
	}

	tablespace_size_in_header = fsp_header_get_tablespace_size();

	if (!srv_auto_extend_last_data_file
	    && sum_of_data_file_sizes != tablespace_size_in_header) {

		fprintf(stderr,
			"InnoDB: Error: tablespace size"
			" stored in header is %lu pages, but\n"
			"InnoDB: the sum of data file sizes is %lu pages\n",
			(ulong) tablespace_size_in_header,
			(ulong) sum_of_data_file_sizes);

		if (srv_force_recovery == 0
		    && sum_of_data_file_sizes < tablespace_size_in_header) {
			/* This is a fatal error, the tail of a tablespace is
			missing */

			fprintf(stderr,
				"InnoDB: Cannot start InnoDB."
				" The tail of the system tablespace is\n"
				"InnoDB: missing. Have you edited"
				" innodb_data_file_path in my.cnf in an\n"
				"InnoDB: inappropriate way, removing"
				" ibdata files from there?\n"
				"InnoDB: You can set innodb_force_recovery=1"
				" in my.cnf to force\n"
				"InnoDB: a startup if you are trying"
				" to recover a badly corrupt database.\n");

			return(DB_ERROR);
		}
	}

	if (srv_auto_extend_last_data_file
	    && sum_of_data_file_sizes < tablespace_size_in_header) {

		fprintf(stderr,
			"InnoDB: Error: tablespace size stored in header"
			" is %lu pages, but\n"
			"InnoDB: the sum of data file sizes"
			" is only %lu pages\n",
			(ulong) tablespace_size_in_header,
			(ulong) sum_of_data_file_sizes);

		if (srv_force_recovery == 0) {

			fprintf(stderr,
				"InnoDB: Cannot start InnoDB. The tail of"
				" the system tablespace is\n"
				"InnoDB: missing. Have you edited"
				" innodb_data_file_path in my.cnf in an\n"
				"InnoDB: inappropriate way, removing"
				" ibdata files from there?\n"
				"InnoDB: You can set innodb_force_recovery=1"
				" in my.cnf to force\n"
				"InnoDB: a startup if you are trying to"
				" recover a badly corrupt database.\n");

			return(DB_ERROR);
		}
	}

	/* Check that os_fast_mutexes work as expected */
	os_fast_mutex_init(&srv_os_test_mutex);

	if (0 != os_fast_mutex_trylock(&srv_os_test_mutex)) {
		fprintf(stderr,
			"InnoDB: Error: pthread_mutex_trylock returns"
			" an unexpected value on\n"
			"InnoDB: success! Cannot continue.\n");
		exit(1);
	}

	os_fast_mutex_unlock(&srv_os_test_mutex);

	os_fast_mutex_lock(&srv_os_test_mutex);

	os_fast_mutex_unlock(&srv_os_test_mutex);

	os_fast_mutex_free(&srv_os_test_mutex);

	if (srv_print_verbose_log) {
		ut_print_timestamp(stderr);
		fprintf(stderr,
			" InnoDB Plugin %s started; "
			"log sequence number %llu\n",
			INNODB_VERSION_STR, srv_start_lsn);
	}

	if (srv_force_recovery > 0) {
		fprintf(stderr,
			"InnoDB: !!! innodb_force_recovery"
			" is set to %lu !!!\n",
			(ulong) srv_force_recovery);
	}

	fflush(stderr);

	if (trx_doublewrite_must_reset_space_ids) {
		/* Actually, we did not change the undo log format between
		4.0 and 4.1.1, and we would not need to run purge to
		completion. Note also that the purge algorithm in 4.1.1
		can process the the history list again even after a full
		purge, because our algorithm does not cut the end of the
		history list in all cases so that it would become empty
		after a full purge. That mean that we may purge 4.0 type
		undo log even after this phase.

		The insert buffer record format changed between 4.0 and
		4.1.1. It is essential that the insert buffer is emptied
		here! */

		fprintf(stderr,
			"InnoDB: You are upgrading to an"
			" InnoDB version which allows multiple\n"
			"InnoDB: tablespaces. Wait that purge"
			" and insert buffer merge run to\n"
			"InnoDB: completion...\n");
		for (;;) {
			os_thread_sleep(1000000);

			if (0 == strcmp(srv_main_thread_op_info,
					"waiting for server activity")) {

				ut_a(ibuf_is_empty());

				break;
			}
		}
		fprintf(stderr,
			"InnoDB: Full purge and insert buffer merge"
			" completed.\n");

		trx_sys_mark_upgraded_to_multiple_tablespaces();

		fprintf(stderr,
			"InnoDB: You have now successfully upgraded"
			" to the multiple tablespaces\n"
			"InnoDB: format. You should NOT DOWNGRADE"
			" to an earlier version of\n"
			"InnoDB: InnoDB! But if you absolutely need to"
			" downgrade, see\n"
			"InnoDB: " REFMAN "multiple-tablespaces.html\n"
			"InnoDB: for instructions.\n");
	}

	if (srv_force_recovery == 0) {
		/* In the insert buffer we may have even bigger tablespace
		id's, because we may have dropped those tablespaces, but
		insert buffer merge has not had time to clean the records from
		the ibuf tree. */

		ibuf_update_max_tablespace_id();
	}

	srv_file_per_table = srv_file_per_table_original_value;

	srv_was_started = TRUE;

	return((int) DB_SUCCESS);
}

/****************************************************************//**
Shuts down the InnoDB database.
@return	DB_SUCCESS or error code */
UNIV_INTERN
int
innobase_shutdown_for_mysql(void)
/*=============================*/
{
	ulint	i;
#ifdef __NETWARE__
	extern ibool panic_shutdown;
#endif
	if (!srv_was_started) {
		if (srv_is_being_started) {
			ut_print_timestamp(stderr);
			fprintf(stderr,
				"  InnoDB: Warning: shutting down"
				" a not properly started\n"
				"InnoDB: or created database!\n");
		}

		return(DB_SUCCESS);
	}

	/* 1. Flush the buffer pool to disk, write the current lsn to
	the tablespace header(s), and copy all log data to archive.
	The step 1 is the real InnoDB shutdown. The remaining steps 2 - ...
	just free data structures after the shutdown. */


	if (srv_fast_shutdown == 2) {
		ut_print_timestamp(stderr);
		fprintf(stderr,
			"  InnoDB: MySQL has requested a very fast shutdown"
			" without flushing "
			"the InnoDB buffer pool to data files."
			" At the next mysqld startup "
			"InnoDB will do a crash recovery!\n");
	}

#ifdef __NETWARE__
	if (!panic_shutdown)
#endif
		logs_empty_and_mark_files_at_shutdown();

	if (srv_conc_n_threads != 0) {
		fprintf(stderr,
			"InnoDB: Warning: query counter shows %ld queries"
			" still\n"
			"InnoDB: inside InnoDB at shutdown\n",
			srv_conc_n_threads);
	}

	/* 2. Make all threads created by InnoDB to exit */

	srv_shutdown_state = SRV_SHUTDOWN_EXIT_THREADS;

	/* In a 'very fast' shutdown, we do not need to wait for these threads
	to die; all which counts is that we flushed the log; a 'very fast'
	shutdown is essentially a crash. */

	if (srv_fast_shutdown == 2) {
		return(DB_SUCCESS);
	}

	/* All threads end up waiting for certain events. Put those events
	to the signaled state. Then the threads will exit themselves in
	os_thread_event_wait(). */

	for (i = 0; i < 1000; i++) {
		/* NOTE: IF YOU CREATE THREADS IN INNODB, YOU MUST EXIT THEM
		HERE OR EARLIER */

		/* a. Let the lock timeout thread exit */
		os_event_set(srv_lock_timeout_thread_event);

		/* b. srv error monitor thread exits automatically, no need
		to do anything here */

		/* c. We wake the master thread so that it exits */
		srv_wake_master_thread();

		/* d. Exit the i/o threads */

		os_aio_wake_all_threads_at_shutdown();

		os_mutex_enter(os_sync_mutex);

		if (os_thread_count == 0) {
			/* All the threads have exited or are just exiting;
			NOTE that the threads may not have completed their
			exit yet. Should we use pthread_join() to make sure
			they have exited? Now we just sleep 0.1 seconds and
			hope that is enough! */

			os_mutex_exit(os_sync_mutex);

			os_thread_sleep(100000);

			break;
		}

		os_mutex_exit(os_sync_mutex);

		os_thread_sleep(100000);
	}

	if (i == 1000) {
		fprintf(stderr,
			"InnoDB: Warning: %lu threads created by InnoDB"
			" had not exited at shutdown!\n",
			(ulong) os_thread_count);
	}

	if (srv_monitor_file) {
		fclose(srv_monitor_file);
		srv_monitor_file = 0;
		if (srv_monitor_file_name) {
			unlink(srv_monitor_file_name);
			mem_free(srv_monitor_file_name);
		}
	}
	if (srv_dict_tmpfile) {
		fclose(srv_dict_tmpfile);
		srv_dict_tmpfile = 0;
	}

	if (srv_misc_tmpfile) {
		fclose(srv_misc_tmpfile);
		srv_misc_tmpfile = 0;
	}

	trx_sys_file_format_close();

	mutex_free(&srv_monitor_file_mutex);
	mutex_free(&srv_dict_tmpfile_mutex);
	mutex_free(&srv_misc_tmpfile_mutex);

	/* 3. Free all InnoDB's own mutexes and the os_fast_mutexes inside
	them */
	sync_close();

	/* 4. Free the os_conc_mutex and all os_events and os_mutexes */

	srv_free();
	os_sync_free();

	/* Check that all read views are closed except read view owned
	by a purge. */

	if (UT_LIST_GET_LEN(trx_sys->view_list) > 1) {
		fprintf(stderr,
			"InnoDB: Error: all read views were not closed"
			" before shutdown:\n"
			"InnoDB: %lu read views open \n",
			UT_LIST_GET_LEN(trx_sys->view_list) - 1);
	}

	/* 5. Free all allocated memory and the os_fast_mutex created in
	ut0mem.c */

	buf_pool_free();
	ut_free_all_mem();

	if (os_thread_count != 0
	    || os_event_count != 0
	    || os_mutex_count != 0
	    || os_fast_mutex_count != 0) {
		fprintf(stderr,
			"InnoDB: Warning: some resources were not"
			" cleaned up in shutdown:\n"
			"InnoDB: threads %lu, events %lu,"
			" os_mutexes %lu, os_fast_mutexes %lu\n",
			(ulong) os_thread_count, (ulong) os_event_count,
			(ulong) os_mutex_count, (ulong) os_fast_mutex_count);
	}

	if (dict_foreign_err_file) {
		fclose(dict_foreign_err_file);
	}
	if (lock_latest_err_file) {
		fclose(lock_latest_err_file);
	}

	if (srv_print_verbose_log) {
		ut_print_timestamp(stderr);
		fprintf(stderr,
			"  InnoDB: Shutdown completed;"
			" log sequence number %llu\n",
			srv_shutdown_lsn);
	}

	srv_was_started = FALSE;

	return((int) DB_SUCCESS);
}

#ifdef __NETWARE__
void set_panic_flag_for_netware()
{
	extern ibool panic_shutdown;
	panic_shutdown = TRUE;
}
#endif /* __NETWARE__ */
#endif /* !UNIV_HOTBACKUP */