md.c 97.8 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
/*
   md.c : Multiple Devices driver for Linux
	  Copyright (C) 1998, 1999, 2000 Ingo Molnar

     completely rewritten, based on the MD driver code from Marc Zyngier

   Changes:

   - RAID-1/RAID-5 extensions by Miguel de Icaza, Gadi Oxman, Ingo Molnar
   - RAID-6 extensions by H. Peter Anvin <hpa@zytor.com>
   - boot support for linear and striped mode by Harald Hoyer <HarryH@Royal.Net>
   - kerneld support by Boris Tobotras <boris@xtalk.msk.su>
   - kmod support by: Cyrus Durgin
   - RAID0 bugfixes: Mark Anthony Lisher <markal@iname.com>
   - Devfs support by Richard Gooch <rgooch@atnf.csiro.au>

   - lots of fixes and improvements to the RAID1/RAID5 and generic
     RAID code (such as request based resynchronization):

     Neil Brown <neilb@cse.unsw.edu.au>.

22 23 24
   - persistent bitmap code
     Copyright (C) 2003-2004, Paul Clements, SteelEye Technology, Inc.

L
Linus Torvalds 已提交
25 26 27 28 29 30 31 32 33 34 35 36
   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; either version 2, or (at your option)
   any later version.

   You should have received a copy of the GNU General Public License
   (for example /usr/src/linux/COPYING); if not, write to the Free
   Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
*/

#include <linux/module.h>
#include <linux/config.h>
37
#include <linux/kthread.h>
L
Linus Torvalds 已提交
38 39
#include <linux/linkage.h>
#include <linux/raid/md.h>
40
#include <linux/raid/bitmap.h>
L
Linus Torvalds 已提交
41 42 43 44 45 46 47
#include <linux/sysctl.h>
#include <linux/devfs_fs_kernel.h>
#include <linux/buffer_head.h> /* for invalidate_bdev */
#include <linux/suspend.h>

#include <linux/init.h>

48 49
#include <linux/file.h>

L
Linus Torvalds 已提交
50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76
#ifdef CONFIG_KMOD
#include <linux/kmod.h>
#endif

#include <asm/unaligned.h>

#define MAJOR_NR MD_MAJOR
#define MD_DRIVER

/* 63 partitions with the alternate major number (mdp) */
#define MdpMinorShift 6

#define DEBUG 0
#define dprintk(x...) ((void)(DEBUG && printk(x)))


#ifndef MODULE
static void autostart_arrays (int part);
#endif

static mdk_personality_t *pers[MAX_PERSONALITY];
static DEFINE_SPINLOCK(pers_lock);

/*
 * Current RAID-1,4,5 parallel reconstruction 'guaranteed speed limit'
 * is 1000 KB/sec, so the extra system load does not show up that much.
 * Increase it if you want to have more _guaranteed_ speed. Note that
77
 * the RAID driver will use the maximum available bandwidth if the IO
L
Linus Torvalds 已提交
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
 * subsystem is idle. There is also an 'absolute maximum' reconstruction
 * speed limit - in case reconstruction slows down your system despite
 * idle IO detection.
 *
 * you can change it via /proc/sys/dev/raid/speed_limit_min and _max.
 */

static int sysctl_speed_limit_min = 1000;
static int sysctl_speed_limit_max = 200000;

static struct ctl_table_header *raid_table_header;

static ctl_table raid_table[] = {
	{
		.ctl_name	= DEV_RAID_SPEED_LIMIT_MIN,
		.procname	= "speed_limit_min",
		.data		= &sysctl_speed_limit_min,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= &proc_dointvec,
	},
	{
		.ctl_name	= DEV_RAID_SPEED_LIMIT_MAX,
		.procname	= "speed_limit_max",
		.data		= &sysctl_speed_limit_max,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= &proc_dointvec,
	},
	{ .ctl_name = 0 }
};

static ctl_table raid_dir_table[] = {
	{
		.ctl_name	= DEV_RAID,
		.procname	= "raid",
		.maxlen		= 0,
		.mode		= 0555,
		.child		= raid_table,
	},
	{ .ctl_name = 0 }
};

static ctl_table raid_root_table[] = {
	{
		.ctl_name	= CTL_DEV,
		.procname	= "dev",
		.maxlen		= 0,
		.mode		= 0555,
		.child		= raid_dir_table,
	},
	{ .ctl_name = 0 }
};

static struct block_device_operations md_fops;

/*
 * Enables to iterate over all existing md arrays
 * all_mddevs_lock protects this list.
 */
static LIST_HEAD(all_mddevs);
static DEFINE_SPINLOCK(all_mddevs_lock);


/*
 * iterates through all used mddevs in the system.
 * We take care to grab the all_mddevs_lock whenever navigating
 * the list, and to always hold a refcount when unlocked.
 * Any code which breaks out of this loop while own
 * a reference to the current mddev and must mddev_put it.
 */
#define ITERATE_MDDEV(mddev,tmp)					\
									\
	for (({ spin_lock(&all_mddevs_lock); 				\
		tmp = all_mddevs.next;					\
		mddev = NULL;});					\
	     ({ if (tmp != &all_mddevs)					\
			mddev_get(list_entry(tmp, mddev_t, all_mddevs));\
		spin_unlock(&all_mddevs_lock);				\
		if (mddev) mddev_put(mddev);				\
		mddev = list_entry(tmp, mddev_t, all_mddevs);		\
		tmp != &all_mddevs;});					\
	     ({ spin_lock(&all_mddevs_lock);				\
		tmp = tmp->next;})					\
		)


static int md_fail_request (request_queue_t *q, struct bio *bio)
{
	bio_io_error(bio, bio->bi_size);
	return 0;
}

static inline mddev_t *mddev_get(mddev_t *mddev)
{
	atomic_inc(&mddev->active);
	return mddev;
}

static void mddev_put(mddev_t *mddev)
{
	if (!atomic_dec_and_lock(&mddev->active, &all_mddevs_lock))
		return;
	if (!mddev->raid_disks && list_empty(&mddev->disks)) {
		list_del(&mddev->all_mddevs);
		blk_put_queue(mddev->queue);
		kfree(mddev);
	}
	spin_unlock(&all_mddevs_lock);
}

static mddev_t * mddev_find(dev_t unit)
{
	mddev_t *mddev, *new = NULL;

 retry:
	spin_lock(&all_mddevs_lock);
	list_for_each_entry(mddev, &all_mddevs, all_mddevs)
		if (mddev->unit == unit) {
			mddev_get(mddev);
			spin_unlock(&all_mddevs_lock);
199
			kfree(new);
L
Linus Torvalds 已提交
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
			return mddev;
		}

	if (new) {
		list_add(&new->all_mddevs, &all_mddevs);
		spin_unlock(&all_mddevs_lock);
		return new;
	}
	spin_unlock(&all_mddevs_lock);

	new = (mddev_t *) kmalloc(sizeof(*new), GFP_KERNEL);
	if (!new)
		return NULL;

	memset(new, 0, sizeof(*new));

	new->unit = unit;
	if (MAJOR(unit) == MD_MAJOR)
		new->md_minor = MINOR(unit);
	else
		new->md_minor = MINOR(unit) >> MdpMinorShift;

	init_MUTEX(&new->reconfig_sem);
	INIT_LIST_HEAD(&new->disks);
	INIT_LIST_HEAD(&new->all_mddevs);
	init_timer(&new->safemode_timer);
	atomic_set(&new->active, 1);
227
	spin_lock_init(&new->write_lock);
228
	init_waitqueue_head(&new->sb_wait);
L
Linus Torvalds 已提交
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

	new->queue = blk_alloc_queue(GFP_KERNEL);
	if (!new->queue) {
		kfree(new);
		return NULL;
	}

	blk_queue_make_request(new->queue, md_fail_request);

	goto retry;
}

static inline int mddev_lock(mddev_t * mddev)
{
	return down_interruptible(&mddev->reconfig_sem);
}

static inline void mddev_lock_uninterruptible(mddev_t * mddev)
{
	down(&mddev->reconfig_sem);
}

static inline int mddev_trylock(mddev_t * mddev)
{
	return down_trylock(&mddev->reconfig_sem);
}

static inline void mddev_unlock(mddev_t * mddev)
{
	up(&mddev->reconfig_sem);

260
	md_wakeup_thread(mddev->thread);
L
Linus Torvalds 已提交
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
}

mdk_rdev_t * find_rdev_nr(mddev_t *mddev, int nr)
{
	mdk_rdev_t * rdev;
	struct list_head *tmp;

	ITERATE_RDEV(mddev,rdev,tmp) {
		if (rdev->desc_nr == nr)
			return rdev;
	}
	return NULL;
}

static mdk_rdev_t * find_rdev(mddev_t * mddev, dev_t dev)
{
	struct list_head *tmp;
	mdk_rdev_t *rdev;

	ITERATE_RDEV(mddev,rdev,tmp) {
		if (rdev->bdev->bd_dev == dev)
			return rdev;
	}
	return NULL;
}

287
static inline sector_t calc_dev_sboffset(struct block_device *bdev)
L
Linus Torvalds 已提交
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
{
	sector_t size = bdev->bd_inode->i_size >> BLOCK_SIZE_BITS;
	return MD_NEW_SIZE_BLOCKS(size);
}

static sector_t calc_dev_size(mdk_rdev_t *rdev, unsigned chunk_size)
{
	sector_t size;

	size = rdev->sb_offset;

	if (chunk_size)
		size &= ~((sector_t)chunk_size/1024 - 1);
	return size;
}

static int alloc_disk_sb(mdk_rdev_t * rdev)
{
	if (rdev->sb_page)
		MD_BUG();

	rdev->sb_page = alloc_page(GFP_KERNEL);
	if (!rdev->sb_page) {
		printk(KERN_ALERT "md: out of memory.\n");
		return -EINVAL;
	}

	return 0;
}

static void free_disk_sb(mdk_rdev_t * rdev)
{
	if (rdev->sb_page) {
		page_cache_release(rdev->sb_page);
		rdev->sb_loaded = 0;
		rdev->sb_page = NULL;
		rdev->sb_offset = 0;
		rdev->size = 0;
	}
}


330 331 332 333 334 335 336 337 338 339 340
static int super_written(struct bio *bio, unsigned int bytes_done, int error)
{
	mdk_rdev_t *rdev = bio->bi_private;
	if (bio->bi_size)
		return 1;

	if (error || !test_bit(BIO_UPTODATE, &bio->bi_flags))
		md_error(rdev->mddev, rdev);

	if (atomic_dec_and_test(&rdev->mddev->pending_writes))
		wake_up(&rdev->mddev->sb_wait);
N
Neil Brown 已提交
341
	bio_put(bio);
342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364
	return 0;
}

void md_super_write(mddev_t *mddev, mdk_rdev_t *rdev,
		   sector_t sector, int size, struct page *page)
{
	/* write first size bytes of page to sector of rdev
	 * Increment mddev->pending_writes before returning
	 * and decrement it on completion, waking up sb_wait
	 * if zero is reached.
	 * If an error occurred, call md_error
	 */
	struct bio *bio = bio_alloc(GFP_NOIO, 1);

	bio->bi_bdev = rdev->bdev;
	bio->bi_sector = sector;
	bio_add_page(bio, page, size, 0);
	bio->bi_private = rdev;
	bio->bi_end_io = super_written;
	atomic_inc(&mddev->pending_writes);
	submit_bio((1<<BIO_RW)|(1<<BIO_RW_SYNC), bio);
}

L
Linus Torvalds 已提交
365 366 367 368 369 370 371 372 373
static int bi_complete(struct bio *bio, unsigned int bytes_done, int error)
{
	if (bio->bi_size)
		return 1;

	complete((struct completion*)bio->bi_private);
	return 0;
}

374
int sync_page_io(struct block_device *bdev, sector_t sector, int size,
L
Linus Torvalds 已提交
375 376
		   struct page *page, int rw)
{
377
	struct bio *bio = bio_alloc(GFP_NOIO, 1);
L
Linus Torvalds 已提交
378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396
	struct completion event;
	int ret;

	rw |= (1 << BIO_RW_SYNC);

	bio->bi_bdev = bdev;
	bio->bi_sector = sector;
	bio_add_page(bio, page, size, 0);
	init_completion(&event);
	bio->bi_private = &event;
	bio->bi_end_io = bi_complete;
	submit_bio(rw, bio);
	wait_for_completion(&event);

	ret = test_bit(BIO_UPTODATE, &bio->bi_flags);
	bio_put(bio);
	return ret;
}

397
static int read_disk_sb(mdk_rdev_t * rdev, int size)
L
Linus Torvalds 已提交
398 399 400 401 402 403 404 405 406 407
{
	char b[BDEVNAME_SIZE];
	if (!rdev->sb_page) {
		MD_BUG();
		return -EINVAL;
	}
	if (rdev->sb_loaded)
		return 0;


408
	if (!sync_page_io(rdev->bdev, rdev->sb_offset<<1, size, rdev->sb_page, READ))
L
Linus Torvalds 已提交
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
		goto fail;
	rdev->sb_loaded = 1;
	return 0;

fail:
	printk(KERN_WARNING "md: disabled device %s, could not read superblock.\n",
		bdevname(rdev->bdev,b));
	return -EINVAL;
}

static int uuid_equal(mdp_super_t *sb1, mdp_super_t *sb2)
{
	if (	(sb1->set_uuid0 == sb2->set_uuid0) &&
		(sb1->set_uuid1 == sb2->set_uuid1) &&
		(sb1->set_uuid2 == sb2->set_uuid2) &&
		(sb1->set_uuid3 == sb2->set_uuid3))

		return 1;

	return 0;
}


static int sb_equal(mdp_super_t *sb1, mdp_super_t *sb2)
{
	int ret;
	mdp_super_t *tmp1, *tmp2;

	tmp1 = kmalloc(sizeof(*tmp1),GFP_KERNEL);
	tmp2 = kmalloc(sizeof(*tmp2),GFP_KERNEL);

	if (!tmp1 || !tmp2) {
		ret = 0;
		printk(KERN_INFO "md.c: sb1 is not equal to sb2!\n");
		goto abort;
	}

	*tmp1 = *sb1;
	*tmp2 = *sb2;

	/*
	 * nr_disks is not constant
	 */
	tmp1->nr_disks = 0;
	tmp2->nr_disks = 0;

	if (memcmp(tmp1, tmp2, MD_SB_GENERIC_CONSTANT_WORDS * 4))
		ret = 0;
	else
		ret = 1;

abort:
461 462
	kfree(tmp1);
	kfree(tmp2);
L
Linus Torvalds 已提交
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
	return ret;
}

static unsigned int calc_sb_csum(mdp_super_t * sb)
{
	unsigned int disk_csum, csum;

	disk_csum = sb->sb_csum;
	sb->sb_csum = 0;
	csum = csum_partial((void *)sb, MD_SB_BYTES, 0);
	sb->sb_csum = disk_csum;
	return csum;
}


/*
 * Handle superblock details.
 * We want to be able to handle multiple superblock formats
 * so we have a common interface to them all, and an array of
 * different handlers.
 * We rely on user-space to write the initial superblock, and support
 * reading and updating of superblocks.
 * Interface methods are:
 *   int load_super(mdk_rdev_t *dev, mdk_rdev_t *refdev, int minor_version)
 *      loads and validates a superblock on dev.
 *      if refdev != NULL, compare superblocks on both devices
 *    Return:
 *      0 - dev has a superblock that is compatible with refdev
 *      1 - dev has a superblock that is compatible and newer than refdev
 *          so dev should be used as the refdev in future
 *     -EINVAL superblock incompatible or invalid
 *     -othererror e.g. -EIO
 *
 *   int validate_super(mddev_t *mddev, mdk_rdev_t *dev)
 *      Verify that dev is acceptable into mddev.
 *       The first time, mddev->raid_disks will be 0, and data from
 *       dev should be merged in.  Subsequent calls check that dev
 *       is new enough.  Return 0 or -EINVAL
 *
 *   void sync_super(mddev_t *mddev, mdk_rdev_t *dev)
 *     Update the superblock for rdev with data in mddev
 *     This does not write to disc.
 *
 */

struct super_type  {
	char 		*name;
	struct module	*owner;
	int		(*load_super)(mdk_rdev_t *rdev, mdk_rdev_t *refdev, int minor_version);
	int		(*validate_super)(mddev_t *mddev, mdk_rdev_t *rdev);
	void		(*sync_super)(mddev_t *mddev, mdk_rdev_t *rdev);
};

/*
 * load_super for 0.90.0 
 */
static int super_90_load(mdk_rdev_t *rdev, mdk_rdev_t *refdev, int minor_version)
{
	char b[BDEVNAME_SIZE], b2[BDEVNAME_SIZE];
	mdp_super_t *sb;
	int ret;
	sector_t sb_offset;

	/*
	 * Calculate the position of the superblock,
	 * it's at the end of the disk.
	 *
	 * It also happens to be a multiple of 4Kb.
	 */
	sb_offset = calc_dev_sboffset(rdev->bdev);
	rdev->sb_offset = sb_offset;

535
	ret = read_disk_sb(rdev, MD_SB_BYTES);
L
Linus Torvalds 已提交
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
	if (ret) return ret;

	ret = -EINVAL;

	bdevname(rdev->bdev, b);
	sb = (mdp_super_t*)page_address(rdev->sb_page);

	if (sb->md_magic != MD_SB_MAGIC) {
		printk(KERN_ERR "md: invalid raid superblock magic on %s\n",
		       b);
		goto abort;
	}

	if (sb->major_version != 0 ||
	    sb->minor_version != 90) {
		printk(KERN_WARNING "Bad version number %d.%d on %s\n",
			sb->major_version, sb->minor_version,
			b);
		goto abort;
	}

	if (sb->raid_disks <= 0)
		goto abort;

	if (csum_fold(calc_sb_csum(sb)) != csum_fold(sb->sb_csum)) {
		printk(KERN_WARNING "md: invalid superblock checksum on %s\n",
			b);
		goto abort;
	}

	rdev->preferred_minor = sb->md_minor;
	rdev->data_offset = 0;
568
	rdev->sb_size = MD_SB_BYTES;
L
Linus Torvalds 已提交
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

	if (sb->level == LEVEL_MULTIPATH)
		rdev->desc_nr = -1;
	else
		rdev->desc_nr = sb->this_disk.number;

	if (refdev == 0)
		ret = 1;
	else {
		__u64 ev1, ev2;
		mdp_super_t *refsb = (mdp_super_t*)page_address(refdev->sb_page);
		if (!uuid_equal(refsb, sb)) {
			printk(KERN_WARNING "md: %s has different UUID to %s\n",
				b, bdevname(refdev->bdev,b2));
			goto abort;
		}
		if (!sb_equal(refsb, sb)) {
			printk(KERN_WARNING "md: %s has same UUID"
			       " but different superblock to %s\n",
			       b, bdevname(refdev->bdev, b2));
			goto abort;
		}
		ev1 = md_event(sb);
		ev2 = md_event(refsb);
		if (ev1 > ev2)
			ret = 1;
		else 
			ret = 0;
	}
	rdev->size = calc_dev_size(rdev, sb->chunk_size);

 abort:
	return ret;
}

/*
 * validate_super for 0.90.0
 */
static int super_90_validate(mddev_t *mddev, mdk_rdev_t *rdev)
{
	mdp_disk_t *desc;
	mdp_super_t *sb = (mdp_super_t *)page_address(rdev->sb_page);

612 613
	rdev->raid_disk = -1;
	rdev->in_sync = 0;
L
Linus Torvalds 已提交
614 615 616 617 618 619 620 621 622 623 624 625 626
	if (mddev->raid_disks == 0) {
		mddev->major_version = 0;
		mddev->minor_version = sb->minor_version;
		mddev->patch_version = sb->patch_version;
		mddev->persistent = ! sb->not_persistent;
		mddev->chunk_size = sb->chunk_size;
		mddev->ctime = sb->ctime;
		mddev->utime = sb->utime;
		mddev->level = sb->level;
		mddev->layout = sb->layout;
		mddev->raid_disks = sb->raid_disks;
		mddev->size = sb->size;
		mddev->events = md_event(sb);
627
		mddev->bitmap_offset = 0;
628
		mddev->default_bitmap_offset = MD_SB_BYTES >> 9;
L
Linus Torvalds 已提交
629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645

		if (sb->state & (1<<MD_SB_CLEAN))
			mddev->recovery_cp = MaxSector;
		else {
			if (sb->events_hi == sb->cp_events_hi && 
				sb->events_lo == sb->cp_events_lo) {
				mddev->recovery_cp = sb->recovery_cp;
			} else
				mddev->recovery_cp = 0;
		}

		memcpy(mddev->uuid+0, &sb->set_uuid0, 4);
		memcpy(mddev->uuid+4, &sb->set_uuid1, 4);
		memcpy(mddev->uuid+8, &sb->set_uuid2, 4);
		memcpy(mddev->uuid+12,&sb->set_uuid3, 4);

		mddev->max_disks = MD_SB_DISKS;
646 647 648

		if (sb->state & (1<<MD_SB_BITMAP_PRESENT) &&
		    mddev->bitmap_file == NULL) {
649
			if (mddev->level != 1 && mddev->level != 5 && mddev->level != 6) {
650 651 652 653
				/* FIXME use a better test */
				printk(KERN_WARNING "md: bitmaps only support for raid1\n");
				return -EINVAL;
			}
654
			mddev->bitmap_offset = mddev->default_bitmap_offset;
655 656
		}

657 658 659
	} else if (mddev->pers == NULL) {
		/* Insist on good event counter while assembling */
		__u64 ev1 = md_event(sb);
L
Linus Torvalds 已提交
660 661 662
		++ev1;
		if (ev1 < mddev->events) 
			return -EINVAL;
663 664 665 666 667 668 669 670 671 672
	} else if (mddev->bitmap) {
		/* if adding to array with a bitmap, then we can accept an
		 * older device ... but not too old.
		 */
		__u64 ev1 = md_event(sb);
		if (ev1 < mddev->bitmap->events_cleared)
			return 0;
	} else /* just a hot-add of a new device, leave raid_disk at -1 */
		return 0;

L
Linus Torvalds 已提交
673
	if (mddev->level != LEVEL_MULTIPATH) {
674
		rdev->faulty = 0;
675
		rdev->flags = 0;
L
Linus Torvalds 已提交
676 677 678 679 680 681 682 683 684
		desc = sb->disks + rdev->desc_nr;

		if (desc->state & (1<<MD_DISK_FAULTY))
			rdev->faulty = 1;
		else if (desc->state & (1<<MD_DISK_SYNC) &&
			 desc->raid_disk < mddev->raid_disks) {
			rdev->in_sync = 1;
			rdev->raid_disk = desc->raid_disk;
		}
685 686
		if (desc->state & (1<<MD_DISK_WRITEMOSTLY))
			set_bit(WriteMostly, &rdev->flags);
687 688
	} else /* MULTIPATH are always insync */
		rdev->in_sync = 1;
L
Linus Torvalds 已提交
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
	return 0;
}

/*
 * sync_super for 0.90.0
 */
static void super_90_sync(mddev_t *mddev, mdk_rdev_t *rdev)
{
	mdp_super_t *sb;
	struct list_head *tmp;
	mdk_rdev_t *rdev2;
	int next_spare = mddev->raid_disks;

	/* make rdev->sb match mddev data..
	 *
	 * 1/ zero out disks
	 * 2/ Add info for each disk, keeping track of highest desc_nr (next_spare);
	 * 3/ any empty disks < next_spare become removed
	 *
	 * disks[0] gets initialised to REMOVED because
	 * we cannot be sure from other fields if it has
	 * been initialised or not.
	 */
	int i;
	int active=0, working=0,failed=0,spare=0,nr_disks=0;

715 716
	rdev->sb_size = MD_SB_BYTES;

L
Linus Torvalds 已提交
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
	sb = (mdp_super_t*)page_address(rdev->sb_page);

	memset(sb, 0, sizeof(*sb));

	sb->md_magic = MD_SB_MAGIC;
	sb->major_version = mddev->major_version;
	sb->minor_version = mddev->minor_version;
	sb->patch_version = mddev->patch_version;
	sb->gvalid_words  = 0; /* ignored */
	memcpy(&sb->set_uuid0, mddev->uuid+0, 4);
	memcpy(&sb->set_uuid1, mddev->uuid+4, 4);
	memcpy(&sb->set_uuid2, mddev->uuid+8, 4);
	memcpy(&sb->set_uuid3, mddev->uuid+12,4);

	sb->ctime = mddev->ctime;
	sb->level = mddev->level;
	sb->size  = mddev->size;
	sb->raid_disks = mddev->raid_disks;
	sb->md_minor = mddev->md_minor;
	sb->not_persistent = !mddev->persistent;
	sb->utime = mddev->utime;
	sb->state = 0;
	sb->events_hi = (mddev->events>>32);
	sb->events_lo = (u32)mddev->events;

	if (mddev->in_sync)
	{
		sb->recovery_cp = mddev->recovery_cp;
		sb->cp_events_hi = (mddev->events>>32);
		sb->cp_events_lo = (u32)mddev->events;
		if (mddev->recovery_cp == MaxSector)
			sb->state = (1<< MD_SB_CLEAN);
	} else
		sb->recovery_cp = 0;

	sb->layout = mddev->layout;
	sb->chunk_size = mddev->chunk_size;

755 756 757
	if (mddev->bitmap && mddev->bitmap_file == NULL)
		sb->state |= (1<<MD_SB_BITMAP_PRESENT);

L
Linus Torvalds 已提交
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
	sb->disks[0].state = (1<<MD_DISK_REMOVED);
	ITERATE_RDEV(mddev,rdev2,tmp) {
		mdp_disk_t *d;
		if (rdev2->raid_disk >= 0 && rdev2->in_sync && !rdev2->faulty)
			rdev2->desc_nr = rdev2->raid_disk;
		else
			rdev2->desc_nr = next_spare++;
		d = &sb->disks[rdev2->desc_nr];
		nr_disks++;
		d->number = rdev2->desc_nr;
		d->major = MAJOR(rdev2->bdev->bd_dev);
		d->minor = MINOR(rdev2->bdev->bd_dev);
		if (rdev2->raid_disk >= 0 && rdev->in_sync && !rdev2->faulty)
			d->raid_disk = rdev2->raid_disk;
		else
			d->raid_disk = rdev2->desc_nr; /* compatibility */
		if (rdev2->faulty) {
			d->state = (1<<MD_DISK_FAULTY);
			failed++;
		} else if (rdev2->in_sync) {
			d->state = (1<<MD_DISK_ACTIVE);
			d->state |= (1<<MD_DISK_SYNC);
			active++;
			working++;
		} else {
			d->state = 0;
			spare++;
			working++;
		}
787 788
		if (test_bit(WriteMostly, &rdev2->flags))
			d->state |= (1<<MD_DISK_WRITEMOSTLY);
L
Linus Torvalds 已提交
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
	}
	
	/* now set the "removed" and "faulty" bits on any missing devices */
	for (i=0 ; i < mddev->raid_disks ; i++) {
		mdp_disk_t *d = &sb->disks[i];
		if (d->state == 0 && d->number == 0) {
			d->number = i;
			d->raid_disk = i;
			d->state = (1<<MD_DISK_REMOVED);
			d->state |= (1<<MD_DISK_FAULTY);
			failed++;
		}
	}
	sb->nr_disks = nr_disks;
	sb->active_disks = active;
	sb->working_disks = working;
	sb->failed_disks = failed;
	sb->spare_disks = spare;

	sb->this_disk = sb->disks[rdev->desc_nr];
	sb->sb_csum = calc_sb_csum(sb);
}

/*
 * version 1 superblock
 */

static unsigned int calc_sb_1_csum(struct mdp_superblock_1 * sb)
{
	unsigned int disk_csum, csum;
	unsigned long long newcsum;
	int size = 256 + le32_to_cpu(sb->max_dev)*2;
	unsigned int *isuper = (unsigned int*)sb;
	int i;

	disk_csum = sb->sb_csum;
	sb->sb_csum = 0;
	newcsum = 0;
	for (i=0; size>=4; size -= 4 )
		newcsum += le32_to_cpu(*isuper++);

	if (size == 2)
		newcsum += le16_to_cpu(*(unsigned short*) isuper);

	csum = (newcsum & 0xffffffff) + (newcsum >> 32);
	sb->sb_csum = disk_csum;
	return cpu_to_le32(csum);
}

static int super_1_load(mdk_rdev_t *rdev, mdk_rdev_t *refdev, int minor_version)
{
	struct mdp_superblock_1 *sb;
	int ret;
	sector_t sb_offset;
	char b[BDEVNAME_SIZE], b2[BDEVNAME_SIZE];
844
	int bmask;
L
Linus Torvalds 已提交
845 846 847 848 849 850 851 852 853 854 855 856 857

	/*
	 * Calculate the position of the superblock.
	 * It is always aligned to a 4K boundary and
	 * depeding on minor_version, it can be:
	 * 0: At least 8K, but less than 12K, from end of device
	 * 1: At start of device
	 * 2: 4K from start of device.
	 */
	switch(minor_version) {
	case 0:
		sb_offset = rdev->bdev->bd_inode->i_size >> 9;
		sb_offset -= 8*2;
858
		sb_offset &= ~(sector_t)(4*2-1);
L
Linus Torvalds 已提交
859 860 861 862 863 864 865 866 867 868 869 870 871 872
		/* convert from sectors to K */
		sb_offset /= 2;
		break;
	case 1:
		sb_offset = 0;
		break;
	case 2:
		sb_offset = 4;
		break;
	default:
		return -EINVAL;
	}
	rdev->sb_offset = sb_offset;

873 874 875 876
	/* superblock is rarely larger than 1K, but it can be larger,
	 * and it is safe to read 4k, so we do that
	 */
	ret = read_disk_sb(rdev, 4096);
L
Linus Torvalds 已提交
877 878 879 880 881 882 883 884 885
	if (ret) return ret;


	sb = (struct mdp_superblock_1*)page_address(rdev->sb_page);

	if (sb->magic != cpu_to_le32(MD_SB_MAGIC) ||
	    sb->major_version != cpu_to_le32(1) ||
	    le32_to_cpu(sb->max_dev) > (4096-256)/2 ||
	    le64_to_cpu(sb->super_offset) != (rdev->sb_offset<<1) ||
886
	    (le32_to_cpu(sb->feature_map) & ~MD_FEATURE_ALL) != 0)
L
Linus Torvalds 已提交
887 888 889 890 891 892 893 894 895 896 897 898 899 900 901
		return -EINVAL;

	if (calc_sb_1_csum(sb) != sb->sb_csum) {
		printk("md: invalid superblock checksum on %s\n",
			bdevname(rdev->bdev,b));
		return -EINVAL;
	}
	if (le64_to_cpu(sb->data_size) < 10) {
		printk("md: data_size too small on %s\n",
		       bdevname(rdev->bdev,b));
		return -EINVAL;
	}
	rdev->preferred_minor = 0xffff;
	rdev->data_offset = le64_to_cpu(sb->data_offset);

902
	rdev->sb_size = le32_to_cpu(sb->max_dev) * 2 + 256;
903
	bmask = queue_hardsect_size(rdev->bdev->bd_disk->queue)-1;
904 905 906
	if (rdev->sb_size & bmask)
		rdev-> sb_size = (rdev->sb_size | bmask)+1;

L
Linus Torvalds 已提交
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
	if (refdev == 0)
		return 1;
	else {
		__u64 ev1, ev2;
		struct mdp_superblock_1 *refsb = 
			(struct mdp_superblock_1*)page_address(refdev->sb_page);

		if (memcmp(sb->set_uuid, refsb->set_uuid, 16) != 0 ||
		    sb->level != refsb->level ||
		    sb->layout != refsb->layout ||
		    sb->chunksize != refsb->chunksize) {
			printk(KERN_WARNING "md: %s has strangely different"
				" superblock to %s\n",
				bdevname(rdev->bdev,b),
				bdevname(refdev->bdev,b2));
			return -EINVAL;
		}
		ev1 = le64_to_cpu(sb->events);
		ev2 = le64_to_cpu(refsb->events);

		if (ev1 > ev2)
			return 1;
	}
	if (minor_version) 
		rdev->size = ((rdev->bdev->bd_inode->i_size>>9) - le64_to_cpu(sb->data_offset)) / 2;
	else
		rdev->size = rdev->sb_offset;
	if (rdev->size < le64_to_cpu(sb->data_size)/2)
		return -EINVAL;
	rdev->size = le64_to_cpu(sb->data_size)/2;
	if (le32_to_cpu(sb->chunksize))
		rdev->size &= ~((sector_t)le32_to_cpu(sb->chunksize)/2 - 1);
	return 0;
}

static int super_1_validate(mddev_t *mddev, mdk_rdev_t *rdev)
{
	struct mdp_superblock_1 *sb = (struct mdp_superblock_1*)page_address(rdev->sb_page);

946 947
	rdev->raid_disk = -1;
	rdev->in_sync = 0;
L
Linus Torvalds 已提交
948 949 950 951 952 953 954 955 956 957 958 959
	if (mddev->raid_disks == 0) {
		mddev->major_version = 1;
		mddev->patch_version = 0;
		mddev->persistent = 1;
		mddev->chunk_size = le32_to_cpu(sb->chunksize) << 9;
		mddev->ctime = le64_to_cpu(sb->ctime) & ((1ULL << 32)-1);
		mddev->utime = le64_to_cpu(sb->utime) & ((1ULL << 32)-1);
		mddev->level = le32_to_cpu(sb->level);
		mddev->layout = le32_to_cpu(sb->layout);
		mddev->raid_disks = le32_to_cpu(sb->raid_disks);
		mddev->size = le64_to_cpu(sb->size)/2;
		mddev->events = le64_to_cpu(sb->events);
960
		mddev->bitmap_offset = 0;
961
		mddev->default_bitmap_offset = 0;
962
		mddev->default_bitmap_offset = 1024;
L
Linus Torvalds 已提交
963 964 965 966 967
		
		mddev->recovery_cp = le64_to_cpu(sb->resync_offset);
		memcpy(mddev->uuid, sb->set_uuid, 16);

		mddev->max_disks =  (4096-256)/2;
968

969
		if ((le32_to_cpu(sb->feature_map) & MD_FEATURE_BITMAP_OFFSET) &&
970 971 972 973 974 975 976
		    mddev->bitmap_file == NULL ) {
			if (mddev->level != 1) {
				printk(KERN_WARNING "md: bitmaps only supported for raid1\n");
				return -EINVAL;
			}
			mddev->bitmap_offset = (__s32)le32_to_cpu(sb->bitmap_offset);
		}
977 978 979
	} else if (mddev->pers == NULL) {
		/* Insist of good event counter while assembling */
		__u64 ev1 = le64_to_cpu(sb->events);
L
Linus Torvalds 已提交
980 981 982
		++ev1;
		if (ev1 < mddev->events)
			return -EINVAL;
983 984 985 986 987 988 989 990 991
	} else if (mddev->bitmap) {
		/* If adding to array with a bitmap, then we can accept an
		 * older device, but not too old.
		 */
		__u64 ev1 = le64_to_cpu(sb->events);
		if (ev1 < mddev->bitmap->events_cleared)
			return 0;
	} else /* just a hot-add of a new device, leave raid_disk at -1 */
		return 0;
L
Linus Torvalds 已提交
992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009

	if (mddev->level != LEVEL_MULTIPATH) {
		int role;
		rdev->desc_nr = le32_to_cpu(sb->dev_number);
		role = le16_to_cpu(sb->dev_roles[rdev->desc_nr]);
		switch(role) {
		case 0xffff: /* spare */
			rdev->faulty = 0;
			break;
		case 0xfffe: /* faulty */
			rdev->faulty = 1;
			break;
		default:
			rdev->in_sync = 1;
			rdev->faulty = 0;
			rdev->raid_disk = role;
			break;
		}
1010 1011 1012
		rdev->flags = 0;
		if (sb->devflags & WriteMostly1)
			set_bit(WriteMostly, &rdev->flags);
1013 1014 1015
	} else /* MULTIPATH are always insync */
		rdev->in_sync = 1;

L
Linus Torvalds 已提交
1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041
	return 0;
}

static void super_1_sync(mddev_t *mddev, mdk_rdev_t *rdev)
{
	struct mdp_superblock_1 *sb;
	struct list_head *tmp;
	mdk_rdev_t *rdev2;
	int max_dev, i;
	/* make rdev->sb match mddev and rdev data. */

	sb = (struct mdp_superblock_1*)page_address(rdev->sb_page);

	sb->feature_map = 0;
	sb->pad0 = 0;
	memset(sb->pad1, 0, sizeof(sb->pad1));
	memset(sb->pad2, 0, sizeof(sb->pad2));
	memset(sb->pad3, 0, sizeof(sb->pad3));

	sb->utime = cpu_to_le64((__u64)mddev->utime);
	sb->events = cpu_to_le64(mddev->events);
	if (mddev->in_sync)
		sb->resync_offset = cpu_to_le64(mddev->recovery_cp);
	else
		sb->resync_offset = cpu_to_le64(0);

1042 1043
	if (mddev->bitmap && mddev->bitmap_file == NULL) {
		sb->bitmap_offset = cpu_to_le32((__u32)mddev->bitmap_offset);
1044
		sb->feature_map = cpu_to_le32(MD_FEATURE_BITMAP_OFFSET);
1045 1046
	}

L
Linus Torvalds 已提交
1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070
	max_dev = 0;
	ITERATE_RDEV(mddev,rdev2,tmp)
		if (rdev2->desc_nr+1 > max_dev)
			max_dev = rdev2->desc_nr+1;
	
	sb->max_dev = cpu_to_le32(max_dev);
	for (i=0; i<max_dev;i++)
		sb->dev_roles[i] = cpu_to_le16(0xfffe);
	
	ITERATE_RDEV(mddev,rdev2,tmp) {
		i = rdev2->desc_nr;
		if (rdev2->faulty)
			sb->dev_roles[i] = cpu_to_le16(0xfffe);
		else if (rdev2->in_sync)
			sb->dev_roles[i] = cpu_to_le16(rdev2->raid_disk);
		else
			sb->dev_roles[i] = cpu_to_le16(0xffff);
	}

	sb->recovery_offset = cpu_to_le64(0); /* not supported yet */
	sb->sb_csum = calc_sb_1_csum(sb);
}


A
Adrian Bunk 已提交
1071
static struct super_type super_types[] = {
L
Linus Torvalds 已提交
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 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 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 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310
	[0] = {
		.name	= "0.90.0",
		.owner	= THIS_MODULE,
		.load_super	= super_90_load,
		.validate_super	= super_90_validate,
		.sync_super	= super_90_sync,
	},
	[1] = {
		.name	= "md-1",
		.owner	= THIS_MODULE,
		.load_super	= super_1_load,
		.validate_super	= super_1_validate,
		.sync_super	= super_1_sync,
	},
};
	
static mdk_rdev_t * match_dev_unit(mddev_t *mddev, mdk_rdev_t *dev)
{
	struct list_head *tmp;
	mdk_rdev_t *rdev;

	ITERATE_RDEV(mddev,rdev,tmp)
		if (rdev->bdev->bd_contains == dev->bdev->bd_contains)
			return rdev;

	return NULL;
}

static int match_mddev_units(mddev_t *mddev1, mddev_t *mddev2)
{
	struct list_head *tmp;
	mdk_rdev_t *rdev;

	ITERATE_RDEV(mddev1,rdev,tmp)
		if (match_dev_unit(mddev2, rdev))
			return 1;

	return 0;
}

static LIST_HEAD(pending_raid_disks);

static int bind_rdev_to_array(mdk_rdev_t * rdev, mddev_t * mddev)
{
	mdk_rdev_t *same_pdev;
	char b[BDEVNAME_SIZE], b2[BDEVNAME_SIZE];

	if (rdev->mddev) {
		MD_BUG();
		return -EINVAL;
	}
	same_pdev = match_dev_unit(mddev, rdev);
	if (same_pdev)
		printk(KERN_WARNING
			"%s: WARNING: %s appears to be on the same physical"
	 		" disk as %s. True\n     protection against single-disk"
			" failure might be compromised.\n",
			mdname(mddev), bdevname(rdev->bdev,b),
			bdevname(same_pdev->bdev,b2));

	/* Verify rdev->desc_nr is unique.
	 * If it is -1, assign a free number, else
	 * check number is not in use
	 */
	if (rdev->desc_nr < 0) {
		int choice = 0;
		if (mddev->pers) choice = mddev->raid_disks;
		while (find_rdev_nr(mddev, choice))
			choice++;
		rdev->desc_nr = choice;
	} else {
		if (find_rdev_nr(mddev, rdev->desc_nr))
			return -EBUSY;
	}
			
	list_add(&rdev->same_set, &mddev->disks);
	rdev->mddev = mddev;
	printk(KERN_INFO "md: bind<%s>\n", bdevname(rdev->bdev,b));
	return 0;
}

static void unbind_rdev_from_array(mdk_rdev_t * rdev)
{
	char b[BDEVNAME_SIZE];
	if (!rdev->mddev) {
		MD_BUG();
		return;
	}
	list_del_init(&rdev->same_set);
	printk(KERN_INFO "md: unbind<%s>\n", bdevname(rdev->bdev,b));
	rdev->mddev = NULL;
}

/*
 * prevent the device from being mounted, repartitioned or
 * otherwise reused by a RAID array (or any other kernel
 * subsystem), by bd_claiming the device.
 */
static int lock_rdev(mdk_rdev_t *rdev, dev_t dev)
{
	int err = 0;
	struct block_device *bdev;
	char b[BDEVNAME_SIZE];

	bdev = open_by_devnum(dev, FMODE_READ|FMODE_WRITE);
	if (IS_ERR(bdev)) {
		printk(KERN_ERR "md: could not open %s.\n",
			__bdevname(dev, b));
		return PTR_ERR(bdev);
	}
	err = bd_claim(bdev, rdev);
	if (err) {
		printk(KERN_ERR "md: could not bd_claim %s.\n",
			bdevname(bdev, b));
		blkdev_put(bdev);
		return err;
	}
	rdev->bdev = bdev;
	return err;
}

static void unlock_rdev(mdk_rdev_t *rdev)
{
	struct block_device *bdev = rdev->bdev;
	rdev->bdev = NULL;
	if (!bdev)
		MD_BUG();
	bd_release(bdev);
	blkdev_put(bdev);
}

void md_autodetect_dev(dev_t dev);

static void export_rdev(mdk_rdev_t * rdev)
{
	char b[BDEVNAME_SIZE];
	printk(KERN_INFO "md: export_rdev(%s)\n",
		bdevname(rdev->bdev,b));
	if (rdev->mddev)
		MD_BUG();
	free_disk_sb(rdev);
	list_del_init(&rdev->same_set);
#ifndef MODULE
	md_autodetect_dev(rdev->bdev->bd_dev);
#endif
	unlock_rdev(rdev);
	kfree(rdev);
}

static void kick_rdev_from_array(mdk_rdev_t * rdev)
{
	unbind_rdev_from_array(rdev);
	export_rdev(rdev);
}

static void export_array(mddev_t *mddev)
{
	struct list_head *tmp;
	mdk_rdev_t *rdev;

	ITERATE_RDEV(mddev,rdev,tmp) {
		if (!rdev->mddev) {
			MD_BUG();
			continue;
		}
		kick_rdev_from_array(rdev);
	}
	if (!list_empty(&mddev->disks))
		MD_BUG();
	mddev->raid_disks = 0;
	mddev->major_version = 0;
}

static void print_desc(mdp_disk_t *desc)
{
	printk(" DISK<N:%d,(%d,%d),R:%d,S:%d>\n", desc->number,
		desc->major,desc->minor,desc->raid_disk,desc->state);
}

static void print_sb(mdp_super_t *sb)
{
	int i;

	printk(KERN_INFO 
		"md:  SB: (V:%d.%d.%d) ID:<%08x.%08x.%08x.%08x> CT:%08x\n",
		sb->major_version, sb->minor_version, sb->patch_version,
		sb->set_uuid0, sb->set_uuid1, sb->set_uuid2, sb->set_uuid3,
		sb->ctime);
	printk(KERN_INFO "md:     L%d S%08d ND:%d RD:%d md%d LO:%d CS:%d\n",
		sb->level, sb->size, sb->nr_disks, sb->raid_disks,
		sb->md_minor, sb->layout, sb->chunk_size);
	printk(KERN_INFO "md:     UT:%08x ST:%d AD:%d WD:%d"
		" FD:%d SD:%d CSUM:%08x E:%08lx\n",
		sb->utime, sb->state, sb->active_disks, sb->working_disks,
		sb->failed_disks, sb->spare_disks,
		sb->sb_csum, (unsigned long)sb->events_lo);

	printk(KERN_INFO);
	for (i = 0; i < MD_SB_DISKS; i++) {
		mdp_disk_t *desc;

		desc = sb->disks + i;
		if (desc->number || desc->major || desc->minor ||
		    desc->raid_disk || (desc->state && (desc->state != 4))) {
			printk("     D %2d: ", i);
			print_desc(desc);
		}
	}
	printk(KERN_INFO "md:     THIS: ");
	print_desc(&sb->this_disk);

}

static void print_rdev(mdk_rdev_t *rdev)
{
	char b[BDEVNAME_SIZE];
	printk(KERN_INFO "md: rdev %s, SZ:%08llu F:%d S:%d DN:%u\n",
		bdevname(rdev->bdev,b), (unsigned long long)rdev->size,
	       	rdev->faulty, rdev->in_sync, rdev->desc_nr);
	if (rdev->sb_loaded) {
		printk(KERN_INFO "md: rdev superblock:\n");
		print_sb((mdp_super_t*)page_address(rdev->sb_page));
	} else
		printk(KERN_INFO "md: no rdev superblock!\n");
}

void md_print_devices(void)
{
	struct list_head *tmp, *tmp2;
	mdk_rdev_t *rdev;
	mddev_t *mddev;
	char b[BDEVNAME_SIZE];

	printk("\n");
	printk("md:	**********************************\n");
	printk("md:	* <COMPLETE RAID STATE PRINTOUT> *\n");
	printk("md:	**********************************\n");
	ITERATE_MDDEV(mddev,tmp) {

1311 1312 1313 1314
		if (mddev->bitmap)
			bitmap_print_sb(mddev->bitmap);
		else
			printk("%s: ", mdname(mddev));
L
Linus Torvalds 已提交
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
		ITERATE_RDEV(mddev,rdev,tmp2)
			printk("<%s>", bdevname(rdev->bdev,b));
		printk("\n");

		ITERATE_RDEV(mddev,rdev,tmp2)
			print_rdev(rdev);
	}
	printk("md:	**********************************\n");
	printk("\n");
}


static void sync_sbs(mddev_t * mddev)
{
	mdk_rdev_t *rdev;
	struct list_head *tmp;

	ITERATE_RDEV(mddev,rdev,tmp) {
		super_types[mddev->major_version].
			sync_super(mddev, rdev);
		rdev->sb_loaded = 1;
	}
}

static void md_update_sb(mddev_t * mddev)
{
1341
	int err;
L
Linus Torvalds 已提交
1342 1343
	struct list_head *tmp;
	mdk_rdev_t *rdev;
1344
	int sync_req;
L
Linus Torvalds 已提交
1345 1346

repeat:
1347 1348
	spin_lock(&mddev->write_lock);
	sync_req = mddev->in_sync;
L
Linus Torvalds 已提交
1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360
	mddev->utime = get_seconds();
	mddev->events ++;

	if (!mddev->events) {
		/*
		 * oops, this 64-bit counter should never wrap.
		 * Either we are in around ~1 trillion A.C., assuming
		 * 1 reboot per second, or we have a bug:
		 */
		MD_BUG();
		mddev->events --;
	}
1361
	mddev->sb_dirty = 2;
L
Linus Torvalds 已提交
1362 1363 1364 1365 1366 1367
	sync_sbs(mddev);

	/*
	 * do not write anything to disk if using
	 * nonpersistent superblocks
	 */
1368 1369 1370
	if (!mddev->persistent) {
		mddev->sb_dirty = 0;
		spin_unlock(&mddev->write_lock);
1371
		wake_up(&mddev->sb_wait);
L
Linus Torvalds 已提交
1372
		return;
1373 1374
	}
	spin_unlock(&mddev->write_lock);
L
Linus Torvalds 已提交
1375 1376 1377 1378 1379

	dprintk(KERN_INFO 
		"md: updating %s RAID superblock on device (in sync %d)\n",
		mdname(mddev),mddev->in_sync);

1380
	err = bitmap_update_sb(mddev->bitmap);
L
Linus Torvalds 已提交
1381 1382 1383 1384 1385 1386 1387 1388
	ITERATE_RDEV(mddev,rdev,tmp) {
		char b[BDEVNAME_SIZE];
		dprintk(KERN_INFO "md: ");
		if (rdev->faulty)
			dprintk("(skipping faulty ");

		dprintk("%s ", bdevname(rdev->bdev,b));
		if (!rdev->faulty) {
1389
			md_super_write(mddev,rdev,
1390
				       rdev->sb_offset<<1, rdev->sb_size,
1391 1392 1393 1394 1395
				       rdev->sb_page);
			dprintk(KERN_INFO "(write) %s's sb offset: %llu\n",
				bdevname(rdev->bdev,b),
				(unsigned long long)rdev->sb_offset);

L
Linus Torvalds 已提交
1396 1397
		} else
			dprintk(")\n");
1398
		if (mddev->level == LEVEL_MULTIPATH)
L
Linus Torvalds 已提交
1399 1400 1401
			/* only need to write one superblock... */
			break;
	}
1402 1403 1404
	wait_event(mddev->sb_wait, atomic_read(&mddev->pending_writes)==0);
	/* if there was a failure, sb_dirty was set to 1, and we re-write super */

1405
	spin_lock(&mddev->write_lock);
1406
	if (mddev->in_sync != sync_req|| mddev->sb_dirty == 1) {
1407 1408 1409 1410 1411 1412
		/* have to write it out again */
		spin_unlock(&mddev->write_lock);
		goto repeat;
	}
	mddev->sb_dirty = 0;
	spin_unlock(&mddev->write_lock);
1413
	wake_up(&mddev->sb_wait);
1414

L
Linus Torvalds 已提交
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
}

/*
 * Import a device. If 'super_format' >= 0, then sanity check the superblock
 *
 * mark the device faulty if:
 *
 *   - the device is nonexistent (zero size)
 *   - the device has no valid superblock
 *
 * a faulty rdev _never_ has rdev->sb set.
 */
static mdk_rdev_t *md_import_device(dev_t newdev, int super_format, int super_minor)
{
	char b[BDEVNAME_SIZE];
	int err;
	mdk_rdev_t *rdev;
	sector_t size;

	rdev = (mdk_rdev_t *) kmalloc(sizeof(*rdev), GFP_KERNEL);
	if (!rdev) {
		printk(KERN_ERR "md: could not alloc mem for new device!\n");
		return ERR_PTR(-ENOMEM);
	}
	memset(rdev, 0, sizeof(*rdev));

	if ((err = alloc_disk_sb(rdev)))
		goto abort_free;

	err = lock_rdev(rdev, newdev);
	if (err)
		goto abort_free;

	rdev->desc_nr = -1;
	rdev->faulty = 0;
	rdev->in_sync = 0;
	rdev->data_offset = 0;
	atomic_set(&rdev->nr_pending, 0);

	size = rdev->bdev->bd_inode->i_size >> BLOCK_SIZE_BITS;
	if (!size) {
		printk(KERN_WARNING 
			"md: %s has zero or unknown size, marking faulty!\n",
			bdevname(rdev->bdev,b));
		err = -EINVAL;
		goto abort_free;
	}

	if (super_format >= 0) {
		err = super_types[super_format].
			load_super(rdev, NULL, super_minor);
		if (err == -EINVAL) {
			printk(KERN_WARNING 
				"md: %s has invalid sb, not importing!\n",
				bdevname(rdev->bdev,b));
			goto abort_free;
		}
		if (err < 0) {
			printk(KERN_WARNING 
				"md: could not read %s's sb, not importing!\n",
				bdevname(rdev->bdev,b));
			goto abort_free;
		}
	}
	INIT_LIST_HEAD(&rdev->same_set);

	return rdev;

abort_free:
	if (rdev->sb_page) {
		if (rdev->bdev)
			unlock_rdev(rdev);
		free_disk_sb(rdev);
	}
	kfree(rdev);
	return ERR_PTR(err);
}

/*
 * Check a full RAID array for plausibility
 */


1498
static void analyze_sbs(mddev_t * mddev)
L
Linus Torvalds 已提交
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
{
	int i;
	struct list_head *tmp;
	mdk_rdev_t *rdev, *freshest;
	char b[BDEVNAME_SIZE];

	freshest = NULL;
	ITERATE_RDEV(mddev,rdev,tmp)
		switch (super_types[mddev->major_version].
			load_super(rdev, freshest, mddev->minor_version)) {
		case 1:
			freshest = rdev;
			break;
		case 0:
			break;
		default:
			printk( KERN_ERR \
				"md: fatal superblock inconsistency in %s"
				" -- removing from array\n", 
				bdevname(rdev->bdev,b));
			kick_rdev_from_array(rdev);
		}


	super_types[mddev->major_version].
		validate_super(mddev, freshest);

	i = 0;
	ITERATE_RDEV(mddev,rdev,tmp) {
		if (rdev != freshest)
			if (super_types[mddev->major_version].
			    validate_super(mddev, rdev)) {
				printk(KERN_WARNING "md: kicking non-fresh %s"
					" from array!\n",
					bdevname(rdev->bdev,b));
				kick_rdev_from_array(rdev);
				continue;
			}
		if (mddev->level == LEVEL_MULTIPATH) {
			rdev->desc_nr = i++;
			rdev->raid_disk = rdev->desc_nr;
			rdev->in_sync = 1;
		}
	}



	if (mddev->recovery_cp != MaxSector &&
	    mddev->level >= 1)
		printk(KERN_ERR "md: %s: raid array is not clean"
		       " -- starting background reconstruction\n",
		       mdname(mddev));

}

int mdp_major = 0;

static struct kobject *md_probe(dev_t dev, int *part, void *data)
{
	static DECLARE_MUTEX(disks_sem);
	mddev_t *mddev = mddev_find(dev);
	struct gendisk *disk;
	int partitioned = (MAJOR(dev) != MD_MAJOR);
	int shift = partitioned ? MdpMinorShift : 0;
	int unit = MINOR(dev) >> shift;

	if (!mddev)
		return NULL;

	down(&disks_sem);
	if (mddev->gendisk) {
		up(&disks_sem);
		mddev_put(mddev);
		return NULL;
	}
	disk = alloc_disk(1 << shift);
	if (!disk) {
		up(&disks_sem);
		mddev_put(mddev);
		return NULL;
	}
	disk->major = MAJOR(dev);
	disk->first_minor = unit << shift;
	if (partitioned) {
		sprintf(disk->disk_name, "md_d%d", unit);
		sprintf(disk->devfs_name, "md/d%d", unit);
	} else {
		sprintf(disk->disk_name, "md%d", unit);
		sprintf(disk->devfs_name, "md/%d", unit);
	}
	disk->fops = &md_fops;
	disk->private_data = mddev;
	disk->queue = mddev->queue;
	add_disk(disk);
	mddev->gendisk = disk;
	up(&disks_sem);
	return NULL;
}

void md_wakeup_thread(mdk_thread_t *thread);

static void md_safemode_timeout(unsigned long data)
{
	mddev_t *mddev = (mddev_t *) data;

	mddev->safemode = 1;
	md_wakeup_thread(mddev->thread);
}


static int do_md_run(mddev_t * mddev)
{
	int pnum, err;
	int chunk_size;
	struct list_head *tmp;
	mdk_rdev_t *rdev;
	struct gendisk *disk;
	char b[BDEVNAME_SIZE];

1618 1619
	if (list_empty(&mddev->disks))
		/* cannot run an array with no devices.. */
L
Linus Torvalds 已提交
1620 1621 1622 1623 1624 1625 1626 1627
		return -EINVAL;

	if (mddev->pers)
		return -EBUSY;

	/*
	 * Analyze all RAID superblock(s)
	 */
1628 1629
	if (!mddev->raid_disks)
		analyze_sbs(mddev);
L
Linus Torvalds 已提交
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

	chunk_size = mddev->chunk_size;
	pnum = level_to_pers(mddev->level);

	if ((pnum != MULTIPATH) && (pnum != RAID1)) {
		if (!chunk_size) {
			/*
			 * 'default chunksize' in the old md code used to
			 * be PAGE_SIZE, baaad.
			 * we abort here to be on the safe side. We don't
			 * want to continue the bad practice.
			 */
			printk(KERN_ERR 
				"no chunksize specified, see 'man raidtab'\n");
			return -EINVAL;
		}
		if (chunk_size > MAX_CHUNK_SIZE) {
			printk(KERN_ERR "too big chunk_size: %d > %d\n",
				chunk_size, MAX_CHUNK_SIZE);
			return -EINVAL;
		}
		/*
		 * chunk-size has to be a power of 2 and multiples of PAGE_SIZE
		 */
		if ( (1 << ffz(~chunk_size)) != chunk_size) {
1655
			printk(KERN_ERR "chunk_size of %d not valid\n", chunk_size);
L
Linus Torvalds 已提交
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
			return -EINVAL;
		}
		if (chunk_size < PAGE_SIZE) {
			printk(KERN_ERR "too small chunk_size: %d < %ld\n",
				chunk_size, PAGE_SIZE);
			return -EINVAL;
		}

		/* devices must have minimum size of one chunk */
		ITERATE_RDEV(mddev,rdev,tmp) {
			if (rdev->faulty)
				continue;
			if (rdev->size < chunk_size / 1024) {
				printk(KERN_WARNING
					"md: Dev %s smaller than chunk_size:"
					" %lluk < %dk\n",
					bdevname(rdev->bdev,b),
					(unsigned long long)rdev->size,
					chunk_size / 1024);
				return -EINVAL;
			}
		}
	}

#ifdef CONFIG_KMOD
	if (!pers[pnum])
	{
		request_module("md-personality-%d", pnum);
	}
#endif

	/*
	 * Drop all container device buffers, from now on
	 * the only valid external interface is through the md
	 * device.
	 * Also find largest hardsector size
	 */
	ITERATE_RDEV(mddev,rdev,tmp) {
		if (rdev->faulty)
			continue;
		sync_blockdev(rdev->bdev);
		invalidate_bdev(rdev->bdev, 0);
	}

	md_probe(mddev->unit, NULL, NULL);
	disk = mddev->gendisk;
	if (!disk)
		return -ENOMEM;

	spin_lock(&pers_lock);
	if (!pers[pnum] || !try_module_get(pers[pnum]->owner)) {
		spin_unlock(&pers_lock);
		printk(KERN_WARNING "md: personality %d is not loaded!\n",
		       pnum);
		return -EINVAL;
	}

	mddev->pers = pers[pnum];
	spin_unlock(&pers_lock);

1716
	mddev->recovery = 0;
L
Linus Torvalds 已提交
1717 1718
	mddev->resync_max_sectors = mddev->size << 1; /* may be over-ridden by personality */

1719 1720 1721 1722 1723 1724 1725
	/* before we start the array running, initialise the bitmap */
	err = bitmap_create(mddev);
	if (err)
		printk(KERN_ERR "%s: failed to create bitmap (%d)\n",
			mdname(mddev), err);
	else
		err = mddev->pers->run(mddev);
L
Linus Torvalds 已提交
1726 1727 1728 1729
	if (err) {
		printk(KERN_ERR "md: pers->run() failed ...\n");
		module_put(mddev->pers->owner);
		mddev->pers = NULL;
1730 1731
		bitmap_destroy(mddev);
		return err;
L
Linus Torvalds 已提交
1732 1733 1734 1735 1736 1737 1738 1739 1740
	}
 	atomic_set(&mddev->writes_pending,0);
	mddev->safemode = 0;
	mddev->safemode_timer.function = md_safemode_timeout;
	mddev->safemode_timer.data = (unsigned long) mddev;
	mddev->safemode_delay = (20 * HZ)/1000 +1; /* 20 msec delay */
	mddev->in_sync = 1;
	
	set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
1741
	md_wakeup_thread(mddev->thread);
L
Linus Torvalds 已提交
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
	
	if (mddev->sb_dirty)
		md_update_sb(mddev);

	set_capacity(disk, mddev->array_size<<1);

	/* If we call blk_queue_make_request here, it will
	 * re-initialise max_sectors etc which may have been
	 * refined inside -> run.  So just set the bits we need to set.
	 * Most initialisation happended when we called
	 * blk_queue_make_request(..., md_fail_request)
	 * earlier.
	 */
	mddev->queue->queuedata = mddev;
	mddev->queue->make_request_fn = mddev->pers->make_request;

	mddev->changed = 1;
	return 0;
}

static int restart_array(mddev_t *mddev)
{
	struct gendisk *disk = mddev->gendisk;
	int err;

	/*
	 * Complain if it has no devices
	 */
	err = -ENXIO;
	if (list_empty(&mddev->disks))
		goto out;

	if (mddev->pers) {
		err = -EBUSY;
		if (!mddev->ro)
			goto out;

		mddev->safemode = 0;
		mddev->ro = 0;
		set_disk_ro(disk, 0);

		printk(KERN_INFO "md: %s switched to read-write mode.\n",
			mdname(mddev));
		/*
		 * Kick recovery or resync if necessary
		 */
		set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
		md_wakeup_thread(mddev->thread);
		err = 0;
	} else {
		printk(KERN_ERR "md: %s has no personality assigned.\n",
			mdname(mddev));
		err = -EINVAL;
	}

out:
	return err;
}

static int do_md_stop(mddev_t * mddev, int ro)
{
	int err = 0;
	struct gendisk *disk = mddev->gendisk;

	if (mddev->pers) {
		if (atomic_read(&mddev->active)>2) {
			printk("md: %s still in use.\n",mdname(mddev));
			return -EBUSY;
		}

		if (mddev->sync_thread) {
			set_bit(MD_RECOVERY_INTR, &mddev->recovery);
			md_unregister_thread(mddev->sync_thread);
			mddev->sync_thread = NULL;
		}

		del_timer_sync(&mddev->safemode_timer);

		invalidate_partition(disk, 0);

		if (ro) {
			err  = -ENXIO;
			if (mddev->ro)
				goto out;
			mddev->ro = 1;
		} else {
1828 1829
			bitmap_flush(mddev);
			wait_event(mddev->sb_wait, atomic_read(&mddev->pending_writes)==0);
L
Linus Torvalds 已提交
1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846
			if (mddev->ro)
				set_disk_ro(disk, 0);
			blk_queue_make_request(mddev->queue, md_fail_request);
			mddev->pers->stop(mddev);
			module_put(mddev->pers->owner);
			mddev->pers = NULL;
			if (mddev->ro)
				mddev->ro = 0;
		}
		if (!mddev->in_sync) {
			/* mark array as shutdown cleanly */
			mddev->in_sync = 1;
			md_update_sb(mddev);
		}
		if (ro)
			set_disk_ro(disk, 1);
	}
1847 1848 1849 1850 1851 1852 1853

	bitmap_destroy(mddev);
	if (mddev->bitmap_file) {
		atomic_set(&mddev->bitmap_file->f_dentry->d_inode->i_writecount, 1);
		fput(mddev->bitmap_file);
		mddev->bitmap_file = NULL;
	}
1854
	mddev->bitmap_offset = 0;
1855

L
Linus Torvalds 已提交
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
	/*
	 * Free resources if final stop
	 */
	if (!ro) {
		struct gendisk *disk;
		printk(KERN_INFO "md: %s stopped.\n", mdname(mddev));

		export_array(mddev);

		mddev->array_size = 0;
		disk = mddev->gendisk;
		if (disk)
			set_capacity(disk, 0);
		mddev->changed = 1;
	} else
		printk(KERN_INFO "md: %s switched to read-only mode.\n",
			mdname(mddev));
	err = 0;
out:
	return err;
}

static void autorun_array(mddev_t *mddev)
{
	mdk_rdev_t *rdev;
	struct list_head *tmp;
	int err;

1884
	if (list_empty(&mddev->disks))
L
Linus Torvalds 已提交
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 2097 2098 2099
		return;

	printk(KERN_INFO "md: running: ");

	ITERATE_RDEV(mddev,rdev,tmp) {
		char b[BDEVNAME_SIZE];
		printk("<%s>", bdevname(rdev->bdev,b));
	}
	printk("\n");

	err = do_md_run (mddev);
	if (err) {
		printk(KERN_WARNING "md: do_md_run() returned %d\n", err);
		do_md_stop (mddev, 0);
	}
}

/*
 * lets try to run arrays based on all disks that have arrived
 * until now. (those are in pending_raid_disks)
 *
 * the method: pick the first pending disk, collect all disks with
 * the same UUID, remove all from the pending list and put them into
 * the 'same_array' list. Then order this list based on superblock
 * update time (freshest comes first), kick out 'old' disks and
 * compare superblocks. If everything's fine then run it.
 *
 * If "unit" is allocated, then bump its reference count
 */
static void autorun_devices(int part)
{
	struct list_head candidates;
	struct list_head *tmp;
	mdk_rdev_t *rdev0, *rdev;
	mddev_t *mddev;
	char b[BDEVNAME_SIZE];

	printk(KERN_INFO "md: autorun ...\n");
	while (!list_empty(&pending_raid_disks)) {
		dev_t dev;
		rdev0 = list_entry(pending_raid_disks.next,
					 mdk_rdev_t, same_set);

		printk(KERN_INFO "md: considering %s ...\n",
			bdevname(rdev0->bdev,b));
		INIT_LIST_HEAD(&candidates);
		ITERATE_RDEV_PENDING(rdev,tmp)
			if (super_90_load(rdev, rdev0, 0) >= 0) {
				printk(KERN_INFO "md:  adding %s ...\n",
					bdevname(rdev->bdev,b));
				list_move(&rdev->same_set, &candidates);
			}
		/*
		 * now we have a set of devices, with all of them having
		 * mostly sane superblocks. It's time to allocate the
		 * mddev.
		 */
		if (rdev0->preferred_minor < 0 || rdev0->preferred_minor >= MAX_MD_DEVS) {
			printk(KERN_INFO "md: unit number in %s is bad: %d\n",
			       bdevname(rdev0->bdev, b), rdev0->preferred_minor);
			break;
		}
		if (part)
			dev = MKDEV(mdp_major,
				    rdev0->preferred_minor << MdpMinorShift);
		else
			dev = MKDEV(MD_MAJOR, rdev0->preferred_minor);

		md_probe(dev, NULL, NULL);
		mddev = mddev_find(dev);
		if (!mddev) {
			printk(KERN_ERR 
				"md: cannot allocate memory for md drive.\n");
			break;
		}
		if (mddev_lock(mddev)) 
			printk(KERN_WARNING "md: %s locked, cannot run\n",
			       mdname(mddev));
		else if (mddev->raid_disks || mddev->major_version
			 || !list_empty(&mddev->disks)) {
			printk(KERN_WARNING 
				"md: %s already running, cannot run %s\n",
				mdname(mddev), bdevname(rdev0->bdev,b));
			mddev_unlock(mddev);
		} else {
			printk(KERN_INFO "md: created %s\n", mdname(mddev));
			ITERATE_RDEV_GENERIC(candidates,rdev,tmp) {
				list_del_init(&rdev->same_set);
				if (bind_rdev_to_array(rdev, mddev))
					export_rdev(rdev);
			}
			autorun_array(mddev);
			mddev_unlock(mddev);
		}
		/* on success, candidates will be empty, on error
		 * it won't...
		 */
		ITERATE_RDEV_GENERIC(candidates,rdev,tmp)
			export_rdev(rdev);
		mddev_put(mddev);
	}
	printk(KERN_INFO "md: ... autorun DONE.\n");
}

/*
 * import RAID devices based on one partition
 * if possible, the array gets run as well.
 */

static int autostart_array(dev_t startdev)
{
	char b[BDEVNAME_SIZE];
	int err = -EINVAL, i;
	mdp_super_t *sb = NULL;
	mdk_rdev_t *start_rdev = NULL, *rdev;

	start_rdev = md_import_device(startdev, 0, 0);
	if (IS_ERR(start_rdev))
		return err;


	/* NOTE: this can only work for 0.90.0 superblocks */
	sb = (mdp_super_t*)page_address(start_rdev->sb_page);
	if (sb->major_version != 0 ||
	    sb->minor_version != 90 ) {
		printk(KERN_WARNING "md: can only autostart 0.90.0 arrays\n");
		export_rdev(start_rdev);
		return err;
	}

	if (start_rdev->faulty) {
		printk(KERN_WARNING 
			"md: can not autostart based on faulty %s!\n",
			bdevname(start_rdev->bdev,b));
		export_rdev(start_rdev);
		return err;
	}
	list_add(&start_rdev->same_set, &pending_raid_disks);

	for (i = 0; i < MD_SB_DISKS; i++) {
		mdp_disk_t *desc = sb->disks + i;
		dev_t dev = MKDEV(desc->major, desc->minor);

		if (!dev)
			continue;
		if (dev == startdev)
			continue;
		if (MAJOR(dev) != desc->major || MINOR(dev) != desc->minor)
			continue;
		rdev = md_import_device(dev, 0, 0);
		if (IS_ERR(rdev))
			continue;

		list_add(&rdev->same_set, &pending_raid_disks);
	}

	/*
	 * possibly return codes
	 */
	autorun_devices(0);
	return 0;

}


static int get_version(void __user * arg)
{
	mdu_version_t ver;

	ver.major = MD_MAJOR_VERSION;
	ver.minor = MD_MINOR_VERSION;
	ver.patchlevel = MD_PATCHLEVEL_VERSION;

	if (copy_to_user(arg, &ver, sizeof(ver)))
		return -EFAULT;

	return 0;
}

static int get_array_info(mddev_t * mddev, void __user * arg)
{
	mdu_array_info_t info;
	int nr,working,active,failed,spare;
	mdk_rdev_t *rdev;
	struct list_head *tmp;

	nr=working=active=failed=spare=0;
	ITERATE_RDEV(mddev,rdev,tmp) {
		nr++;
		if (rdev->faulty)
			failed++;
		else {
			working++;
			if (rdev->in_sync)
				active++;	
			else
				spare++;
		}
	}

	info.major_version = mddev->major_version;
	info.minor_version = mddev->minor_version;
	info.patch_version = MD_PATCHLEVEL_VERSION;
	info.ctime         = mddev->ctime;
	info.level         = mddev->level;
	info.size          = mddev->size;
	info.nr_disks      = nr;
	info.raid_disks    = mddev->raid_disks;
	info.md_minor      = mddev->md_minor;
	info.not_persistent= !mddev->persistent;

	info.utime         = mddev->utime;
	info.state         = 0;
	if (mddev->in_sync)
		info.state = (1<<MD_SB_CLEAN);
2100 2101
	if (mddev->bitmap && mddev->bitmap_offset)
		info.state = (1<<MD_SB_BITMAP_PRESENT);
L
Linus Torvalds 已提交
2102 2103 2104 2105 2106 2107 2108 2109 2110 2111 2112 2113 2114 2115
	info.active_disks  = active;
	info.working_disks = working;
	info.failed_disks  = failed;
	info.spare_disks   = spare;

	info.layout        = mddev->layout;
	info.chunk_size    = mddev->chunk_size;

	if (copy_to_user(arg, &info, sizeof(info)))
		return -EFAULT;

	return 0;
}

2116
static int get_bitmap_file(mddev_t * mddev, void __user * arg)
2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148 2149 2150 2151
{
	mdu_bitmap_file_t *file = NULL; /* too big for stack allocation */
	char *ptr, *buf = NULL;
	int err = -ENOMEM;

	file = kmalloc(sizeof(*file), GFP_KERNEL);
	if (!file)
		goto out;

	/* bitmap disabled, zero the first byte and copy out */
	if (!mddev->bitmap || !mddev->bitmap->file) {
		file->pathname[0] = '\0';
		goto copy_out;
	}

	buf = kmalloc(sizeof(file->pathname), GFP_KERNEL);
	if (!buf)
		goto out;

	ptr = file_path(mddev->bitmap->file, buf, sizeof(file->pathname));
	if (!ptr)
		goto out;

	strcpy(file->pathname, ptr);

copy_out:
	err = 0;
	if (copy_to_user(arg, file, sizeof(*file)))
		err = -EFAULT;
out:
	kfree(buf);
	kfree(file);
	return err;
}

L
Linus Torvalds 已提交
2152 2153 2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170 2171 2172 2173 2174
static int get_disk_info(mddev_t * mddev, void __user * arg)
{
	mdu_disk_info_t info;
	unsigned int nr;
	mdk_rdev_t *rdev;

	if (copy_from_user(&info, arg, sizeof(info)))
		return -EFAULT;

	nr = info.number;

	rdev = find_rdev_nr(mddev, nr);
	if (rdev) {
		info.major = MAJOR(rdev->bdev->bd_dev);
		info.minor = MINOR(rdev->bdev->bd_dev);
		info.raid_disk = rdev->raid_disk;
		info.state = 0;
		if (rdev->faulty)
			info.state |= (1<<MD_DISK_FAULTY);
		else if (rdev->in_sync) {
			info.state |= (1<<MD_DISK_ACTIVE);
			info.state |= (1<<MD_DISK_SYNC);
		}
2175 2176
		if (test_bit(WriteMostly, &rdev->flags))
			info.state |= (1<<MD_DISK_WRITEMOSTLY);
L
Linus Torvalds 已提交
2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187 2188 2189 2190 2191 2192 2193 2194 2195 2196 2197 2198 2199 2200 2201 2202 2203 2204 2205 2206 2207 2208 2209 2210 2211 2212 2213 2214 2215 2216 2217 2218 2219 2220 2221 2222 2223 2224 2225 2226 2227 2228 2229 2230 2231 2232 2233 2234 2235 2236 2237 2238 2239 2240
	} else {
		info.major = info.minor = 0;
		info.raid_disk = -1;
		info.state = (1<<MD_DISK_REMOVED);
	}

	if (copy_to_user(arg, &info, sizeof(info)))
		return -EFAULT;

	return 0;
}

static int add_new_disk(mddev_t * mddev, mdu_disk_info_t *info)
{
	char b[BDEVNAME_SIZE], b2[BDEVNAME_SIZE];
	mdk_rdev_t *rdev;
	dev_t dev = MKDEV(info->major,info->minor);

	if (info->major != MAJOR(dev) || info->minor != MINOR(dev))
		return -EOVERFLOW;

	if (!mddev->raid_disks) {
		int err;
		/* expecting a device which has a superblock */
		rdev = md_import_device(dev, mddev->major_version, mddev->minor_version);
		if (IS_ERR(rdev)) {
			printk(KERN_WARNING 
				"md: md_import_device returned %ld\n",
				PTR_ERR(rdev));
			return PTR_ERR(rdev);
		}
		if (!list_empty(&mddev->disks)) {
			mdk_rdev_t *rdev0 = list_entry(mddev->disks.next,
							mdk_rdev_t, same_set);
			int err = super_types[mddev->major_version]
				.load_super(rdev, rdev0, mddev->minor_version);
			if (err < 0) {
				printk(KERN_WARNING 
					"md: %s has different UUID to %s\n",
					bdevname(rdev->bdev,b), 
					bdevname(rdev0->bdev,b2));
				export_rdev(rdev);
				return -EINVAL;
			}
		}
		err = bind_rdev_to_array(rdev, mddev);
		if (err)
			export_rdev(rdev);
		return err;
	}

	/*
	 * add_new_disk can be used once the array is assembled
	 * to add "hot spares".  They must already have a superblock
	 * written
	 */
	if (mddev->pers) {
		int err;
		if (!mddev->pers->hot_add_disk) {
			printk(KERN_WARNING 
				"%s: personality does not support diskops!\n",
			       mdname(mddev));
			return -EINVAL;
		}
2241 2242 2243 2244 2245
		if (mddev->persistent)
			rdev = md_import_device(dev, mddev->major_version,
						mddev->minor_version);
		else
			rdev = md_import_device(dev, -1, -1);
L
Linus Torvalds 已提交
2246 2247 2248 2249 2250 2251
		if (IS_ERR(rdev)) {
			printk(KERN_WARNING 
				"md: md_import_device returned %ld\n",
				PTR_ERR(rdev));
			return PTR_ERR(rdev);
		}
2252 2253 2254 2255 2256 2257 2258 2259 2260 2261 2262 2263
		/* set save_raid_disk if appropriate */
		if (!mddev->persistent) {
			if (info->state & (1<<MD_DISK_SYNC)  &&
			    info->raid_disk < mddev->raid_disks)
				rdev->raid_disk = info->raid_disk;
			else
				rdev->raid_disk = -1;
		} else
			super_types[mddev->major_version].
				validate_super(mddev, rdev);
		rdev->saved_raid_disk = rdev->raid_disk;

L
Linus Torvalds 已提交
2264
		rdev->in_sync = 0; /* just to be sure */
2265 2266 2267
		if (info->state & (1<<MD_DISK_WRITEMOSTLY))
			set_bit(WriteMostly, &rdev->flags);

L
Linus Torvalds 已提交
2268 2269 2270 2271
		rdev->raid_disk = -1;
		err = bind_rdev_to_array(rdev, mddev);
		if (err)
			export_rdev(rdev);
2272 2273

		set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
2274
		md_wakeup_thread(mddev->thread);
L
Linus Torvalds 已提交
2275 2276 2277 2278 2279 2280 2281 2282 2283 2284 2285 2286 2287 2288 2289 2290 2291 2292 2293 2294 2295 2296 2297 2298 2299 2300 2301 2302 2303 2304 2305 2306 2307
		return err;
	}

	/* otherwise, add_new_disk is only allowed
	 * for major_version==0 superblocks
	 */
	if (mddev->major_version != 0) {
		printk(KERN_WARNING "%s: ADD_NEW_DISK not supported\n",
		       mdname(mddev));
		return -EINVAL;
	}

	if (!(info->state & (1<<MD_DISK_FAULTY))) {
		int err;
		rdev = md_import_device (dev, -1, 0);
		if (IS_ERR(rdev)) {
			printk(KERN_WARNING 
				"md: error, md_import_device() returned %ld\n",
				PTR_ERR(rdev));
			return PTR_ERR(rdev);
		}
		rdev->desc_nr = info->number;
		if (info->raid_disk < mddev->raid_disks)
			rdev->raid_disk = info->raid_disk;
		else
			rdev->raid_disk = -1;

		rdev->faulty = 0;
		if (rdev->raid_disk < mddev->raid_disks)
			rdev->in_sync = (info->state & (1<<MD_DISK_SYNC));
		else
			rdev->in_sync = 0;

2308 2309 2310
		if (info->state & (1<<MD_DISK_WRITEMOSTLY))
			set_bit(WriteMostly, &rdev->flags);

L
Linus Torvalds 已提交
2311 2312 2313 2314 2315 2316 2317 2318 2319 2320 2321 2322 2323 2324 2325 2326 2327 2328 2329 2330 2331 2332 2333 2334 2335 2336 2337 2338 2339 2340 2341 2342 2343 2344 2345 2346 2347 2348 2349 2350 2351 2352 2353 2354 2355 2356 2357 2358 2359 2360 2361 2362 2363 2364 2365 2366 2367 2368 2369 2370 2371 2372 2373 2374 2375 2376 2377 2378 2379 2380 2381 2382 2383 2384 2385 2386 2387 2388 2389 2390 2391 2392 2393 2394 2395 2396 2397 2398 2399 2400 2401 2402 2403 2404 2405 2406 2407 2408 2409 2410 2411 2412 2413 2414 2415 2416 2417 2418 2419 2420 2421 2422 2423 2424 2425 2426 2427 2428 2429 2430 2431 2432 2433 2434 2435 2436 2437 2438 2439 2440 2441 2442 2443 2444 2445 2446 2447 2448
		err = bind_rdev_to_array(rdev, mddev);
		if (err) {
			export_rdev(rdev);
			return err;
		}

		if (!mddev->persistent) {
			printk(KERN_INFO "md: nonpersistent superblock ...\n");
			rdev->sb_offset = rdev->bdev->bd_inode->i_size >> BLOCK_SIZE_BITS;
		} else 
			rdev->sb_offset = calc_dev_sboffset(rdev->bdev);
		rdev->size = calc_dev_size(rdev, mddev->chunk_size);

		if (!mddev->size || (mddev->size > rdev->size))
			mddev->size = rdev->size;
	}

	return 0;
}

static int hot_remove_disk(mddev_t * mddev, dev_t dev)
{
	char b[BDEVNAME_SIZE];
	mdk_rdev_t *rdev;

	if (!mddev->pers)
		return -ENODEV;

	rdev = find_rdev(mddev, dev);
	if (!rdev)
		return -ENXIO;

	if (rdev->raid_disk >= 0)
		goto busy;

	kick_rdev_from_array(rdev);
	md_update_sb(mddev);

	return 0;
busy:
	printk(KERN_WARNING "md: cannot remove active disk %s from %s ... \n",
		bdevname(rdev->bdev,b), mdname(mddev));
	return -EBUSY;
}

static int hot_add_disk(mddev_t * mddev, dev_t dev)
{
	char b[BDEVNAME_SIZE];
	int err;
	unsigned int size;
	mdk_rdev_t *rdev;

	if (!mddev->pers)
		return -ENODEV;

	if (mddev->major_version != 0) {
		printk(KERN_WARNING "%s: HOT_ADD may only be used with"
			" version-0 superblocks.\n",
			mdname(mddev));
		return -EINVAL;
	}
	if (!mddev->pers->hot_add_disk) {
		printk(KERN_WARNING 
			"%s: personality does not support diskops!\n",
			mdname(mddev));
		return -EINVAL;
	}

	rdev = md_import_device (dev, -1, 0);
	if (IS_ERR(rdev)) {
		printk(KERN_WARNING 
			"md: error, md_import_device() returned %ld\n",
			PTR_ERR(rdev));
		return -EINVAL;
	}

	if (mddev->persistent)
		rdev->sb_offset = calc_dev_sboffset(rdev->bdev);
	else
		rdev->sb_offset =
			rdev->bdev->bd_inode->i_size >> BLOCK_SIZE_BITS;

	size = calc_dev_size(rdev, mddev->chunk_size);
	rdev->size = size;

	if (size < mddev->size) {
		printk(KERN_WARNING 
			"%s: disk size %llu blocks < array size %llu\n",
			mdname(mddev), (unsigned long long)size,
			(unsigned long long)mddev->size);
		err = -ENOSPC;
		goto abort_export;
	}

	if (rdev->faulty) {
		printk(KERN_WARNING 
			"md: can not hot-add faulty %s disk to %s!\n",
			bdevname(rdev->bdev,b), mdname(mddev));
		err = -EINVAL;
		goto abort_export;
	}
	rdev->in_sync = 0;
	rdev->desc_nr = -1;
	bind_rdev_to_array(rdev, mddev);

	/*
	 * The rest should better be atomic, we can have disk failures
	 * noticed in interrupt contexts ...
	 */

	if (rdev->desc_nr == mddev->max_disks) {
		printk(KERN_WARNING "%s: can not hot-add to full array!\n",
			mdname(mddev));
		err = -EBUSY;
		goto abort_unbind_export;
	}

	rdev->raid_disk = -1;

	md_update_sb(mddev);

	/*
	 * Kick recovery, maybe this spare has to be added to the
	 * array immediately.
	 */
	set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
	md_wakeup_thread(mddev->thread);

	return 0;

abort_unbind_export:
	unbind_rdev_from_array(rdev);

abort_export:
	export_rdev(rdev);
	return err;
}

2449 2450 2451 2452 2453 2454 2455 2456 2457 2458 2459 2460 2461 2462 2463 2464 2465 2466 2467 2468 2469
/* similar to deny_write_access, but accounts for our holding a reference
 * to the file ourselves */
static int deny_bitmap_write_access(struct file * file)
{
	struct inode *inode = file->f_mapping->host;

	spin_lock(&inode->i_lock);
	if (atomic_read(&inode->i_writecount) > 1) {
		spin_unlock(&inode->i_lock);
		return -ETXTBSY;
	}
	atomic_set(&inode->i_writecount, -1);
	spin_unlock(&inode->i_lock);

	return 0;
}

static int set_bitmap_file(mddev_t *mddev, int fd)
{
	int err;

2470 2471 2472 2473 2474 2475 2476
	if (mddev->pers) {
		if (!mddev->pers->quiesce)
			return -EBUSY;
		if (mddev->recovery || mddev->sync_thread)
			return -EBUSY;
		/* we should be able to change the bitmap.. */
	}
2477 2478


2479 2480 2481 2482
	if (fd >= 0) {
		if (mddev->bitmap)
			return -EEXIST; /* cannot add when bitmap is present */
		mddev->bitmap_file = fget(fd);
2483

2484 2485 2486 2487 2488 2489 2490 2491 2492 2493 2494 2495 2496 2497
		if (mddev->bitmap_file == NULL) {
			printk(KERN_ERR "%s: error: failed to get bitmap file\n",
			       mdname(mddev));
			return -EBADF;
		}

		err = deny_bitmap_write_access(mddev->bitmap_file);
		if (err) {
			printk(KERN_ERR "%s: error: bitmap file is already in use\n",
			       mdname(mddev));
			fput(mddev->bitmap_file);
			mddev->bitmap_file = NULL;
			return err;
		}
2498
		mddev->bitmap_offset = 0; /* file overrides offset */
2499 2500 2501 2502 2503 2504 2505 2506 2507 2508 2509 2510 2511 2512 2513 2514
	} else if (mddev->bitmap == NULL)
		return -ENOENT; /* cannot remove what isn't there */
	err = 0;
	if (mddev->pers) {
		mddev->pers->quiesce(mddev, 1);
		if (fd >= 0)
			err = bitmap_create(mddev);
		if (fd < 0 || err)
			bitmap_destroy(mddev);
		mddev->pers->quiesce(mddev, 0);
	} else if (fd < 0) {
		if (mddev->bitmap_file)
			fput(mddev->bitmap_file);
		mddev->bitmap_file = NULL;
	}

2515 2516 2517
	return err;
}

L
Linus Torvalds 已提交
2518 2519 2520 2521 2522 2523 2524 2525 2526 2527 2528 2529 2530 2531 2532 2533 2534 2535 2536 2537 2538 2539 2540 2541 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
/*
 * set_array_info is used two different ways
 * The original usage is when creating a new array.
 * In this usage, raid_disks is > 0 and it together with
 *  level, size, not_persistent,layout,chunksize determine the
 *  shape of the array.
 *  This will always create an array with a type-0.90.0 superblock.
 * The newer usage is when assembling an array.
 *  In this case raid_disks will be 0, and the major_version field is
 *  use to determine which style super-blocks are to be found on the devices.
 *  The minor and patch _version numbers are also kept incase the
 *  super_block handler wishes to interpret them.
 */
static int set_array_info(mddev_t * mddev, mdu_array_info_t *info)
{

	if (info->raid_disks == 0) {
		/* just setting version number for superblock loading */
		if (info->major_version < 0 ||
		    info->major_version >= sizeof(super_types)/sizeof(super_types[0]) ||
		    super_types[info->major_version].name == NULL) {
			/* maybe try to auto-load a module? */
			printk(KERN_INFO 
				"md: superblock version %d not known\n",
				info->major_version);
			return -EINVAL;
		}
		mddev->major_version = info->major_version;
		mddev->minor_version = info->minor_version;
		mddev->patch_version = info->patch_version;
		return 0;
	}
	mddev->major_version = MD_MAJOR_VERSION;
	mddev->minor_version = MD_MINOR_VERSION;
	mddev->patch_version = MD_PATCHLEVEL_VERSION;
	mddev->ctime         = get_seconds();

	mddev->level         = info->level;
	mddev->size          = info->size;
	mddev->raid_disks    = info->raid_disks;
	/* don't set md_minor, it is determined by which /dev/md* was
	 * openned
	 */
	if (info->state & (1<<MD_SB_CLEAN))
		mddev->recovery_cp = MaxSector;
	else
		mddev->recovery_cp = 0;
	mddev->persistent    = ! info->not_persistent;

	mddev->layout        = info->layout;
	mddev->chunk_size    = info->chunk_size;

	mddev->max_disks     = MD_SB_DISKS;

	mddev->sb_dirty      = 1;

	/*
	 * Generate a 128 bit UUID
	 */
	get_random_bytes(mddev->uuid, 16);

	return 0;
}

/*
 * update_array_info is used to change the configuration of an
 * on-line array.
 * The version, ctime,level,size,raid_disks,not_persistent, layout,chunk_size
 * fields in the info are checked against the array.
 * Any differences that cannot be handled will cause an error.
 * Normally, only one change can be managed at a time.
 */
static int update_array_info(mddev_t *mddev, mdu_array_info_t *info)
{
	int rv = 0;
	int cnt = 0;
2594 2595 2596 2597 2598
	int state = 0;

	/* calculate expected state,ignoring low bits */
	if (mddev->bitmap && mddev->bitmap_offset)
		state |= (1 << MD_SB_BITMAP_PRESENT);
L
Linus Torvalds 已提交
2599 2600 2601 2602 2603 2604 2605 2606

	if (mddev->major_version != info->major_version ||
	    mddev->minor_version != info->minor_version ||
/*	    mddev->patch_version != info->patch_version || */
	    mddev->ctime         != info->ctime         ||
	    mddev->level         != info->level         ||
/*	    mddev->layout        != info->layout        || */
	    !mddev->persistent	 != info->not_persistent||
2607 2608 2609 2610
	    mddev->chunk_size    != info->chunk_size    ||
	    /* ignore bottom 8 bits of state, and allow SB_BITMAP_PRESENT to change */
	    ((state^info->state) & 0xfffffe00)
		)
L
Linus Torvalds 已提交
2611 2612 2613 2614 2615
		return -EINVAL;
	/* Check there is only one change */
	if (mddev->size != info->size) cnt++;
	if (mddev->raid_disks != info->raid_disks) cnt++;
	if (mddev->layout != info->layout) cnt++;
2616
	if ((state ^ info->state) & (1<<MD_SB_BITMAP_PRESENT)) cnt++;
L
Linus Torvalds 已提交
2617 2618 2619 2620 2621 2622 2623 2624 2625 2626 2627 2628 2629 2630 2631 2632 2633 2634 2635 2636 2637 2638 2639 2640 2641 2642 2643 2644 2645 2646 2647 2648 2649 2650 2651 2652 2653 2654 2655 2656 2657 2658 2659 2660 2661 2662 2663 2664 2665 2666 2667 2668 2669 2670 2671 2672 2673 2674 2675 2676 2677 2678 2679 2680 2681 2682 2683 2684 2685 2686 2687 2688 2689 2690 2691 2692 2693 2694
	if (cnt == 0) return 0;
	if (cnt > 1) return -EINVAL;

	if (mddev->layout != info->layout) {
		/* Change layout
		 * we don't need to do anything at the md level, the
		 * personality will take care of it all.
		 */
		if (mddev->pers->reconfig == NULL)
			return -EINVAL;
		else
			return mddev->pers->reconfig(mddev, info->layout, -1);
	}
	if (mddev->size != info->size) {
		mdk_rdev_t * rdev;
		struct list_head *tmp;
		if (mddev->pers->resize == NULL)
			return -EINVAL;
		/* The "size" is the amount of each device that is used.
		 * This can only make sense for arrays with redundancy.
		 * linear and raid0 always use whatever space is available
		 * We can only consider changing the size if no resync
		 * or reconstruction is happening, and if the new size
		 * is acceptable. It must fit before the sb_offset or,
		 * if that is <data_offset, it must fit before the
		 * size of each device.
		 * If size is zero, we find the largest size that fits.
		 */
		if (mddev->sync_thread)
			return -EBUSY;
		ITERATE_RDEV(mddev,rdev,tmp) {
			sector_t avail;
			int fit = (info->size == 0);
			if (rdev->sb_offset > rdev->data_offset)
				avail = (rdev->sb_offset*2) - rdev->data_offset;
			else
				avail = get_capacity(rdev->bdev->bd_disk)
					- rdev->data_offset;
			if (fit && (info->size == 0 || info->size > avail/2))
				info->size = avail/2;
			if (avail < ((sector_t)info->size << 1))
				return -ENOSPC;
		}
		rv = mddev->pers->resize(mddev, (sector_t)info->size *2);
		if (!rv) {
			struct block_device *bdev;

			bdev = bdget_disk(mddev->gendisk, 0);
			if (bdev) {
				down(&bdev->bd_inode->i_sem);
				i_size_write(bdev->bd_inode, mddev->array_size << 10);
				up(&bdev->bd_inode->i_sem);
				bdput(bdev);
			}
		}
	}
	if (mddev->raid_disks    != info->raid_disks) {
		/* change the number of raid disks */
		if (mddev->pers->reshape == NULL)
			return -EINVAL;
		if (info->raid_disks <= 0 ||
		    info->raid_disks >= mddev->max_disks)
			return -EINVAL;
		if (mddev->sync_thread)
			return -EBUSY;
		rv = mddev->pers->reshape(mddev, info->raid_disks);
		if (!rv) {
			struct block_device *bdev;

			bdev = bdget_disk(mddev->gendisk, 0);
			if (bdev) {
				down(&bdev->bd_inode->i_sem);
				i_size_write(bdev->bd_inode, mddev->array_size << 10);
				up(&bdev->bd_inode->i_sem);
				bdput(bdev);
			}
		}
	}
2695 2696 2697 2698 2699 2700 2701 2702 2703 2704 2705 2706 2707 2708 2709 2710 2711 2712 2713 2714 2715 2716 2717 2718 2719 2720 2721 2722 2723
	if ((state ^ info->state) & (1<<MD_SB_BITMAP_PRESENT)) {
		if (mddev->pers->quiesce == NULL)
			return -EINVAL;
		if (mddev->recovery || mddev->sync_thread)
			return -EBUSY;
		if (info->state & (1<<MD_SB_BITMAP_PRESENT)) {
			/* add the bitmap */
			if (mddev->bitmap)
				return -EEXIST;
			if (mddev->default_bitmap_offset == 0)
				return -EINVAL;
			mddev->bitmap_offset = mddev->default_bitmap_offset;
			mddev->pers->quiesce(mddev, 1);
			rv = bitmap_create(mddev);
			if (rv)
				bitmap_destroy(mddev);
			mddev->pers->quiesce(mddev, 0);
		} else {
			/* remove the bitmap */
			if (!mddev->bitmap)
				return -ENOENT;
			if (mddev->bitmap->file)
				return -EINVAL;
			mddev->pers->quiesce(mddev, 1);
			bitmap_destroy(mddev);
			mddev->pers->quiesce(mddev, 0);
			mddev->bitmap_offset = 0;
		}
	}
L
Linus Torvalds 已提交
2724 2725 2726 2727 2728 2729 2730 2731 2732 2733 2734 2735 2736 2737 2738 2739 2740 2741 2742 2743 2744 2745 2746 2747 2748 2749 2750 2751 2752 2753 2754 2755 2756 2757 2758 2759 2760 2761 2762 2763 2764 2765 2766 2767 2768 2769 2770 2771 2772 2773 2774 2775 2776 2777 2778 2779 2780 2781 2782 2783 2784 2785 2786 2787 2788 2789 2790 2791 2792 2793 2794 2795 2796 2797 2798 2799 2800 2801 2802 2803 2804 2805 2806 2807 2808 2809 2810 2811 2812 2813 2814 2815 2816 2817 2818 2819 2820 2821 2822 2823 2824 2825 2826 2827 2828 2829 2830 2831 2832 2833 2834 2835 2836 2837 2838 2839 2840 2841 2842 2843 2844 2845 2846 2847 2848 2849 2850 2851 2852 2853 2854 2855 2856 2857 2858 2859 2860 2861 2862 2863 2864 2865 2866
	md_update_sb(mddev);
	return rv;
}

static int set_disk_faulty(mddev_t *mddev, dev_t dev)
{
	mdk_rdev_t *rdev;

	if (mddev->pers == NULL)
		return -ENODEV;

	rdev = find_rdev(mddev, dev);
	if (!rdev)
		return -ENODEV;

	md_error(mddev, rdev);
	return 0;
}

static int md_ioctl(struct inode *inode, struct file *file,
			unsigned int cmd, unsigned long arg)
{
	int err = 0;
	void __user *argp = (void __user *)arg;
	struct hd_geometry __user *loc = argp;
	mddev_t *mddev = NULL;

	if (!capable(CAP_SYS_ADMIN))
		return -EACCES;

	/*
	 * Commands dealing with the RAID driver but not any
	 * particular array:
	 */
	switch (cmd)
	{
		case RAID_VERSION:
			err = get_version(argp);
			goto done;

		case PRINT_RAID_DEBUG:
			err = 0;
			md_print_devices();
			goto done;

#ifndef MODULE
		case RAID_AUTORUN:
			err = 0;
			autostart_arrays(arg);
			goto done;
#endif
		default:;
	}

	/*
	 * Commands creating/starting a new array:
	 */

	mddev = inode->i_bdev->bd_disk->private_data;

	if (!mddev) {
		BUG();
		goto abort;
	}


	if (cmd == START_ARRAY) {
		/* START_ARRAY doesn't need to lock the array as autostart_array
		 * does the locking, and it could even be a different array
		 */
		static int cnt = 3;
		if (cnt > 0 ) {
			printk(KERN_WARNING
			       "md: %s(pid %d) used deprecated START_ARRAY ioctl. "
			       "This will not be supported beyond 2.6\n",
			       current->comm, current->pid);
			cnt--;
		}
		err = autostart_array(new_decode_dev(arg));
		if (err) {
			printk(KERN_WARNING "md: autostart failed!\n");
			goto abort;
		}
		goto done;
	}

	err = mddev_lock(mddev);
	if (err) {
		printk(KERN_INFO 
			"md: ioctl lock interrupted, reason %d, cmd %d\n",
			err, cmd);
		goto abort;
	}

	switch (cmd)
	{
		case SET_ARRAY_INFO:
			{
				mdu_array_info_t info;
				if (!arg)
					memset(&info, 0, sizeof(info));
				else if (copy_from_user(&info, argp, sizeof(info))) {
					err = -EFAULT;
					goto abort_unlock;
				}
				if (mddev->pers) {
					err = update_array_info(mddev, &info);
					if (err) {
						printk(KERN_WARNING "md: couldn't update"
						       " array info. %d\n", err);
						goto abort_unlock;
					}
					goto done_unlock;
				}
				if (!list_empty(&mddev->disks)) {
					printk(KERN_WARNING
					       "md: array %s already has disks!\n",
					       mdname(mddev));
					err = -EBUSY;
					goto abort_unlock;
				}
				if (mddev->raid_disks) {
					printk(KERN_WARNING
					       "md: array %s already initialised!\n",
					       mdname(mddev));
					err = -EBUSY;
					goto abort_unlock;
				}
				err = set_array_info(mddev, &info);
				if (err) {
					printk(KERN_WARNING "md: couldn't set"
					       " array info. %d\n", err);
					goto abort_unlock;
				}
			}
			goto done_unlock;

		default:;
	}

	/*
	 * Commands querying/configuring an existing array:
	 */
2867 2868 2869 2870
	/* if we are not initialised yet, only ADD_NEW_DISK, STOP_ARRAY,
	 * RUN_ARRAY, and SET_BITMAP_FILE are allowed */
	if (!mddev->raid_disks && cmd != ADD_NEW_DISK && cmd != STOP_ARRAY
			&& cmd != RUN_ARRAY && cmd != SET_BITMAP_FILE) {
L
Linus Torvalds 已提交
2871 2872 2873 2874 2875 2876 2877 2878 2879 2880 2881 2882 2883
		err = -ENODEV;
		goto abort_unlock;
	}

	/*
	 * Commands even a read-only array can execute:
	 */
	switch (cmd)
	{
		case GET_ARRAY_INFO:
			err = get_array_info(mddev, argp);
			goto done_unlock;

2884
		case GET_BITMAP_FILE:
2885
			err = get_bitmap_file(mddev, argp);
2886 2887
			goto done_unlock;

L
Linus Torvalds 已提交
2888 2889 2890 2891 2892 2893 2894 2895 2896 2897 2898 2899 2900 2901 2902 2903 2904 2905 2906 2907 2908 2909 2910 2911 2912 2913 2914 2915 2916 2917 2918 2919 2920 2921 2922 2923 2924 2925 2926 2927 2928 2929 2930 2931 2932 2933 2934 2935 2936 2937 2938 2939 2940 2941 2942 2943 2944 2945 2946 2947 2948 2949 2950 2951 2952 2953 2954 2955 2956 2957 2958 2959 2960 2961 2962 2963 2964 2965 2966 2967
		case GET_DISK_INFO:
			err = get_disk_info(mddev, argp);
			goto done_unlock;

		case RESTART_ARRAY_RW:
			err = restart_array(mddev);
			goto done_unlock;

		case STOP_ARRAY:
			err = do_md_stop (mddev, 0);
			goto done_unlock;

		case STOP_ARRAY_RO:
			err = do_md_stop (mddev, 1);
			goto done_unlock;

	/*
	 * We have a problem here : there is no easy way to give a CHS
	 * virtual geometry. We currently pretend that we have a 2 heads
	 * 4 sectors (with a BIG number of cylinders...). This drives
	 * dosfs just mad... ;-)
	 */
		case HDIO_GETGEO:
			if (!loc) {
				err = -EINVAL;
				goto abort_unlock;
			}
			err = put_user (2, (char __user *) &loc->heads);
			if (err)
				goto abort_unlock;
			err = put_user (4, (char __user *) &loc->sectors);
			if (err)
				goto abort_unlock;
			err = put_user(get_capacity(mddev->gendisk)/8,
					(short __user *) &loc->cylinders);
			if (err)
				goto abort_unlock;
			err = put_user (get_start_sect(inode->i_bdev),
						(long __user *) &loc->start);
			goto done_unlock;
	}

	/*
	 * The remaining ioctls are changing the state of the
	 * superblock, so we do not allow read-only arrays
	 * here:
	 */
	if (mddev->ro) {
		err = -EROFS;
		goto abort_unlock;
	}

	switch (cmd)
	{
		case ADD_NEW_DISK:
		{
			mdu_disk_info_t info;
			if (copy_from_user(&info, argp, sizeof(info)))
				err = -EFAULT;
			else
				err = add_new_disk(mddev, &info);
			goto done_unlock;
		}

		case HOT_REMOVE_DISK:
			err = hot_remove_disk(mddev, new_decode_dev(arg));
			goto done_unlock;

		case HOT_ADD_DISK:
			err = hot_add_disk(mddev, new_decode_dev(arg));
			goto done_unlock;

		case SET_DISK_FAULTY:
			err = set_disk_faulty(mddev, new_decode_dev(arg));
			goto done_unlock;

		case RUN_ARRAY:
			err = do_md_run (mddev);
			goto done_unlock;

2968 2969 2970 2971
		case SET_BITMAP_FILE:
			err = set_bitmap_file(mddev, (int)arg);
			goto done_unlock;

L
Linus Torvalds 已提交
2972 2973 2974 2975 2976 2977 2978 2979 2980 2981 2982 2983 2984 2985 2986 2987 2988 2989 2990 2991 2992 2993 2994 2995 2996 2997 2998 2999 3000 3001 3002 3003 3004 3005 3006 3007 3008 3009 3010 3011 3012 3013 3014 3015 3016 3017 3018 3019 3020 3021 3022 3023 3024 3025 3026 3027 3028 3029 3030 3031 3032 3033 3034 3035 3036 3037 3038 3039 3040 3041 3042 3043 3044 3045 3046 3047 3048 3049
		default:
			if (_IOC_TYPE(cmd) == MD_MAJOR)
				printk(KERN_WARNING "md: %s(pid %d) used"
					" obsolete MD ioctl, upgrade your"
					" software to use new ictls.\n",
					current->comm, current->pid);
			err = -EINVAL;
			goto abort_unlock;
	}

done_unlock:
abort_unlock:
	mddev_unlock(mddev);

	return err;
done:
	if (err)
		MD_BUG();
abort:
	return err;
}

static int md_open(struct inode *inode, struct file *file)
{
	/*
	 * Succeed if we can lock the mddev, which confirms that
	 * it isn't being stopped right now.
	 */
	mddev_t *mddev = inode->i_bdev->bd_disk->private_data;
	int err;

	if ((err = mddev_lock(mddev)))
		goto out;

	err = 0;
	mddev_get(mddev);
	mddev_unlock(mddev);

	check_disk_change(inode->i_bdev);
 out:
	return err;
}

static int md_release(struct inode *inode, struct file * file)
{
 	mddev_t *mddev = inode->i_bdev->bd_disk->private_data;

	if (!mddev)
		BUG();
	mddev_put(mddev);

	return 0;
}

static int md_media_changed(struct gendisk *disk)
{
	mddev_t *mddev = disk->private_data;

	return mddev->changed;
}

static int md_revalidate(struct gendisk *disk)
{
	mddev_t *mddev = disk->private_data;

	mddev->changed = 0;
	return 0;
}
static struct block_device_operations md_fops =
{
	.owner		= THIS_MODULE,
	.open		= md_open,
	.release	= md_release,
	.ioctl		= md_ioctl,
	.media_changed	= md_media_changed,
	.revalidate_disk= md_revalidate,
};

A
Adrian Bunk 已提交
3050
static int md_thread(void * arg)
L
Linus Torvalds 已提交
3051 3052 3053 3054 3055 3056 3057 3058 3059 3060 3061 3062 3063 3064 3065 3066
{
	mdk_thread_t *thread = arg;

	/*
	 * md_thread is a 'system-thread', it's priority should be very
	 * high. We avoid resource deadlocks individually in each
	 * raid personality. (RAID5 does preallocation) We also use RR and
	 * the very same RT priority as kswapd, thus we will never get
	 * into a priority inversion deadlock.
	 *
	 * we definitely have to have equal or higher priority than
	 * bdflush, otherwise bdflush will deadlock if there are too
	 * many dirty RAID5 blocks.
	 */

	complete(thread->event);
3067
	while (!kthread_should_stop()) {
L
Linus Torvalds 已提交
3068 3069
		void (*run)(mddev_t *);

3070
		wait_event_interruptible_timeout(thread->wqueue,
3071 3072
						 test_bit(THREAD_WAKEUP, &thread->flags)
						 || kthread_should_stop(),
3073
						 thread->timeout);
3074
		try_to_freeze();
L
Linus Torvalds 已提交
3075 3076 3077 3078 3079 3080 3081

		clear_bit(THREAD_WAKEUP, &thread->flags);

		run = thread->run;
		if (run)
			run(thread->mddev);
	}
3082

L
Linus Torvalds 已提交
3083 3084 3085 3086 3087 3088 3089 3090 3091 3092 3093 3094 3095 3096 3097 3098 3099 3100
	return 0;
}

void md_wakeup_thread(mdk_thread_t *thread)
{
	if (thread) {
		dprintk("md: waking up MD thread %s.\n", thread->tsk->comm);
		set_bit(THREAD_WAKEUP, &thread->flags);
		wake_up(&thread->wqueue);
	}
}

mdk_thread_t *md_register_thread(void (*run) (mddev_t *), mddev_t *mddev,
				 const char *name)
{
	mdk_thread_t *thread;
	struct completion event;

3101
	thread = kmalloc(sizeof(mdk_thread_t), GFP_KERNEL);
L
Linus Torvalds 已提交
3102 3103 3104 3105 3106 3107 3108 3109 3110 3111 3112
	if (!thread)
		return NULL;

	memset(thread, 0, sizeof(mdk_thread_t));
	init_waitqueue_head(&thread->wqueue);

	init_completion(&event);
	thread->event = &event;
	thread->run = run;
	thread->mddev = mddev;
	thread->name = name;
3113
	thread->timeout = MAX_SCHEDULE_TIMEOUT;
3114 3115
	thread->tsk = kthread_run(md_thread, thread, mdname(thread->mddev));
	if (IS_ERR(thread->tsk)) {
L
Linus Torvalds 已提交
3116 3117 3118 3119 3120 3121 3122 3123 3124
		kfree(thread);
		return NULL;
	}
	wait_for_completion(&event);
	return thread;
}

void md_unregister_thread(mdk_thread_t *thread)
{
3125
	dprintk("interrupting MD-thread pid %d\n", thread->tsk->pid);
3126 3127

	kthread_stop(thread->tsk);
L
Linus Torvalds 已提交
3128 3129 3130 3131 3132 3133 3134 3135 3136 3137 3138 3139
	kfree(thread);
}

void md_error(mddev_t *mddev, mdk_rdev_t *rdev)
{
	if (!mddev) {
		MD_BUG();
		return;
	}

	if (!rdev || rdev->faulty)
		return;
3140
/*
L
Linus Torvalds 已提交
3141 3142 3143 3144 3145
	dprintk("md_error dev:%s, rdev:(%d:%d), (caller: %p,%p,%p,%p).\n",
		mdname(mddev),
		MAJOR(rdev->bdev->bd_dev), MINOR(rdev->bdev->bd_dev),
		__builtin_return_address(0),__builtin_return_address(1),
		__builtin_return_address(2),__builtin_return_address(3));
3146
*/
L
Linus Torvalds 已提交
3147 3148 3149 3150 3151 3152 3153 3154 3155 3156 3157 3158 3159 3160 3161 3162 3163 3164 3165 3166 3167 3168 3169 3170 3171 3172 3173 3174 3175 3176 3177 3178 3179 3180 3181 3182 3183 3184 3185 3186 3187 3188 3189 3190 3191 3192 3193 3194 3195 3196 3197 3198 3199 3200 3201 3202 3203 3204 3205 3206 3207 3208 3209 3210 3211 3212 3213 3214 3215 3216 3217 3218 3219 3220 3221 3222 3223 3224 3225 3226 3227 3228 3229 3230 3231 3232 3233 3234 3235 3236 3237 3238 3239 3240 3241 3242 3243 3244 3245 3246 3247 3248 3249 3250 3251 3252 3253 3254 3255 3256 3257 3258 3259 3260 3261 3262 3263 3264 3265 3266 3267 3268 3269 3270 3271 3272 3273 3274 3275 3276 3277 3278 3279 3280 3281 3282 3283 3284 3285 3286 3287 3288 3289 3290 3291 3292 3293 3294 3295 3296 3297 3298 3299
	if (!mddev->pers->error_handler)
		return;
	mddev->pers->error_handler(mddev,rdev);
	set_bit(MD_RECOVERY_INTR, &mddev->recovery);
	set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
	md_wakeup_thread(mddev->thread);
}

/* seq_file implementation /proc/mdstat */

static void status_unused(struct seq_file *seq)
{
	int i = 0;
	mdk_rdev_t *rdev;
	struct list_head *tmp;

	seq_printf(seq, "unused devices: ");

	ITERATE_RDEV_PENDING(rdev,tmp) {
		char b[BDEVNAME_SIZE];
		i++;
		seq_printf(seq, "%s ",
			      bdevname(rdev->bdev,b));
	}
	if (!i)
		seq_printf(seq, "<none>");

	seq_printf(seq, "\n");
}


static void status_resync(struct seq_file *seq, mddev_t * mddev)
{
	unsigned long max_blocks, resync, res, dt, db, rt;

	resync = (mddev->curr_resync - atomic_read(&mddev->recovery_active))/2;

	if (test_bit(MD_RECOVERY_SYNC, &mddev->recovery))
		max_blocks = mddev->resync_max_sectors >> 1;
	else
		max_blocks = mddev->size;

	/*
	 * Should not happen.
	 */
	if (!max_blocks) {
		MD_BUG();
		return;
	}
	res = (resync/1024)*1000/(max_blocks/1024 + 1);
	{
		int i, x = res/50, y = 20-x;
		seq_printf(seq, "[");
		for (i = 0; i < x; i++)
			seq_printf(seq, "=");
		seq_printf(seq, ">");
		for (i = 0; i < y; i++)
			seq_printf(seq, ".");
		seq_printf(seq, "] ");
	}
	seq_printf(seq, " %s =%3lu.%lu%% (%lu/%lu)",
		      (test_bit(MD_RECOVERY_SYNC, &mddev->recovery) ?
		       "resync" : "recovery"),
		      res/10, res % 10, resync, max_blocks);

	/*
	 * We do not want to overflow, so the order of operands and
	 * the * 100 / 100 trick are important. We do a +1 to be
	 * safe against division by zero. We only estimate anyway.
	 *
	 * dt: time from mark until now
	 * db: blocks written from mark until now
	 * rt: remaining time
	 */
	dt = ((jiffies - mddev->resync_mark) / HZ);
	if (!dt) dt++;
	db = resync - (mddev->resync_mark_cnt/2);
	rt = (dt * ((max_blocks-resync) / (db/100+1)))/100;

	seq_printf(seq, " finish=%lu.%lumin", rt / 60, (rt % 60)/6);

	seq_printf(seq, " speed=%ldK/sec", db/dt);
}

static void *md_seq_start(struct seq_file *seq, loff_t *pos)
{
	struct list_head *tmp;
	loff_t l = *pos;
	mddev_t *mddev;

	if (l >= 0x10000)
		return NULL;
	if (!l--)
		/* header */
		return (void*)1;

	spin_lock(&all_mddevs_lock);
	list_for_each(tmp,&all_mddevs)
		if (!l--) {
			mddev = list_entry(tmp, mddev_t, all_mddevs);
			mddev_get(mddev);
			spin_unlock(&all_mddevs_lock);
			return mddev;
		}
	spin_unlock(&all_mddevs_lock);
	if (!l--)
		return (void*)2;/* tail */
	return NULL;
}

static void *md_seq_next(struct seq_file *seq, void *v, loff_t *pos)
{
	struct list_head *tmp;
	mddev_t *next_mddev, *mddev = v;
	
	++*pos;
	if (v == (void*)2)
		return NULL;

	spin_lock(&all_mddevs_lock);
	if (v == (void*)1)
		tmp = all_mddevs.next;
	else
		tmp = mddev->all_mddevs.next;
	if (tmp != &all_mddevs)
		next_mddev = mddev_get(list_entry(tmp,mddev_t,all_mddevs));
	else {
		next_mddev = (void*)2;
		*pos = 0x10000;
	}		
	spin_unlock(&all_mddevs_lock);

	if (v != (void*)1)
		mddev_put(mddev);
	return next_mddev;

}

static void md_seq_stop(struct seq_file *seq, void *v)
{
	mddev_t *mddev = v;

	if (mddev && v != (void*)1 && v != (void*)2)
		mddev_put(mddev);
}

static int md_seq_show(struct seq_file *seq, void *v)
{
	mddev_t *mddev = v;
	sector_t size;
	struct list_head *tmp2;
	mdk_rdev_t *rdev;
	int i;
3300
	struct bitmap *bitmap;
L
Linus Torvalds 已提交
3301 3302 3303 3304 3305 3306 3307 3308 3309 3310 3311 3312 3313 3314 3315 3316 3317 3318 3319 3320 3321 3322 3323 3324 3325 3326 3327 3328 3329 3330 3331 3332 3333

	if (v == (void*)1) {
		seq_printf(seq, "Personalities : ");
		spin_lock(&pers_lock);
		for (i = 0; i < MAX_PERSONALITY; i++)
			if (pers[i])
				seq_printf(seq, "[%s] ", pers[i]->name);

		spin_unlock(&pers_lock);
		seq_printf(seq, "\n");
		return 0;
	}
	if (v == (void*)2) {
		status_unused(seq);
		return 0;
	}

	if (mddev_lock(mddev)!=0) 
		return -EINTR;
	if (mddev->pers || mddev->raid_disks || !list_empty(&mddev->disks)) {
		seq_printf(seq, "%s : %sactive", mdname(mddev),
						mddev->pers ? "" : "in");
		if (mddev->pers) {
			if (mddev->ro)
				seq_printf(seq, " (read-only)");
			seq_printf(seq, " %s", mddev->pers->name);
		}

		size = 0;
		ITERATE_RDEV(mddev,rdev,tmp2) {
			char b[BDEVNAME_SIZE];
			seq_printf(seq, " %s[%d]",
				bdevname(rdev->bdev,b), rdev->desc_nr);
3334 3335
			if (test_bit(WriteMostly, &rdev->flags))
				seq_printf(seq, "(W)");
L
Linus Torvalds 已提交
3336 3337 3338
			if (rdev->faulty) {
				seq_printf(seq, "(F)");
				continue;
3339 3340
			} else if (rdev->raid_disk < 0)
				seq_printf(seq, "(S)"); /* spare */
L
Linus Torvalds 已提交
3341 3342 3343 3344 3345 3346 3347 3348 3349 3350 3351
			size += rdev->size;
		}

		if (!list_empty(&mddev->disks)) {
			if (mddev->pers)
				seq_printf(seq, "\n      %llu blocks",
					(unsigned long long)mddev->array_size);
			else
				seq_printf(seq, "\n      %llu blocks",
					(unsigned long long)size);
		}
3352 3353 3354 3355 3356 3357 3358 3359 3360
		if (mddev->persistent) {
			if (mddev->major_version != 0 ||
			    mddev->minor_version != 90) {
				seq_printf(seq," super %d.%d",
					   mddev->major_version,
					   mddev->minor_version);
			}
		} else
			seq_printf(seq, " super non-persistent");
L
Linus Torvalds 已提交
3361 3362 3363 3364

		if (mddev->pers) {
			mddev->pers->status (seq, mddev);
	 		seq_printf(seq, "\n      ");
3365
			if (mddev->curr_resync > 2) {
L
Linus Torvalds 已提交
3366
				status_resync (seq, mddev);
3367 3368 3369 3370 3371 3372 3373 3374 3375 3376 3377 3378 3379 3380 3381 3382 3383 3384 3385
				seq_printf(seq, "\n      ");
			} else if (mddev->curr_resync == 1 || mddev->curr_resync == 2)
				seq_printf(seq, "	resync=DELAYED\n      ");
		} else
			seq_printf(seq, "\n       ");

		if ((bitmap = mddev->bitmap)) {
			unsigned long chunk_kb;
			unsigned long flags;
			spin_lock_irqsave(&bitmap->lock, flags);
			chunk_kb = bitmap->chunksize >> 10;
			seq_printf(seq, "bitmap: %lu/%lu pages [%luKB], "
				"%lu%s chunk",
				bitmap->pages - bitmap->missing_pages,
				bitmap->pages,
				(bitmap->pages - bitmap->missing_pages)
					<< (PAGE_SHIFT - 10),
				chunk_kb ? chunk_kb : bitmap->chunksize,
				chunk_kb ? "KB" : "B");
3386 3387 3388 3389
			if (bitmap->file) {
				seq_printf(seq, ", file: ");
				seq_path(seq, bitmap->file->f_vfsmnt,
					 bitmap->file->f_dentry," \t\n");
3390
			}
3391

3392 3393
			seq_printf(seq, "\n");
			spin_unlock_irqrestore(&bitmap->lock, flags);
L
Linus Torvalds 已提交
3394 3395 3396 3397 3398 3399 3400 3401 3402 3403 3404 3405 3406 3407 3408 3409 3410 3411 3412 3413 3414 3415 3416 3417 3418 3419 3420 3421 3422 3423 3424 3425 3426 3427 3428 3429 3430 3431 3432 3433 3434 3435 3436 3437 3438 3439 3440 3441 3442 3443 3444 3445 3446 3447
		}

		seq_printf(seq, "\n");
	}
	mddev_unlock(mddev);
	
	return 0;
}

static struct seq_operations md_seq_ops = {
	.start  = md_seq_start,
	.next   = md_seq_next,
	.stop   = md_seq_stop,
	.show   = md_seq_show,
};

static int md_seq_open(struct inode *inode, struct file *file)
{
	int error;

	error = seq_open(file, &md_seq_ops);
	return error;
}

static struct file_operations md_seq_fops = {
	.open           = md_seq_open,
	.read           = seq_read,
	.llseek         = seq_lseek,
	.release	= seq_release,
};

int register_md_personality(int pnum, mdk_personality_t *p)
{
	if (pnum >= MAX_PERSONALITY) {
		printk(KERN_ERR
		       "md: tried to install personality %s as nr %d, but max is %lu\n",
		       p->name, pnum, MAX_PERSONALITY-1);
		return -EINVAL;
	}

	spin_lock(&pers_lock);
	if (pers[pnum]) {
		spin_unlock(&pers_lock);
		return -EBUSY;
	}

	pers[pnum] = p;
	printk(KERN_INFO "md: %s personality registered as nr %d\n", p->name, pnum);
	spin_unlock(&pers_lock);
	return 0;
}

int unregister_md_personality(int pnum)
{
3448
	if (pnum >= MAX_PERSONALITY)
L
Linus Torvalds 已提交
3449 3450 3451 3452 3453 3454 3455 3456 3457 3458 3459 3460 3461 3462 3463 3464 3465 3466 3467 3468 3469 3470 3471 3472 3473 3474 3475 3476 3477 3478 3479 3480 3481 3482 3483 3484 3485 3486 3487 3488 3489 3490 3491 3492 3493 3494 3495
		return -EINVAL;

	printk(KERN_INFO "md: %s personality unregistered\n", pers[pnum]->name);
	spin_lock(&pers_lock);
	pers[pnum] = NULL;
	spin_unlock(&pers_lock);
	return 0;
}

static int is_mddev_idle(mddev_t *mddev)
{
	mdk_rdev_t * rdev;
	struct list_head *tmp;
	int idle;
	unsigned long curr_events;

	idle = 1;
	ITERATE_RDEV(mddev,rdev,tmp) {
		struct gendisk *disk = rdev->bdev->bd_contains->bd_disk;
		curr_events = disk_stat_read(disk, read_sectors) + 
				disk_stat_read(disk, write_sectors) - 
				atomic_read(&disk->sync_io);
		/* Allow some slack between valud of curr_events and last_events,
		 * as there are some uninteresting races.
		 * Note: the following is an unsigned comparison.
		 */
		if ((curr_events - rdev->last_events + 32) > 64) {
			rdev->last_events = curr_events;
			idle = 0;
		}
	}
	return idle;
}

void md_done_sync(mddev_t *mddev, int blocks, int ok)
{
	/* another "blocks" (512byte) blocks have been synced */
	atomic_sub(blocks, &mddev->recovery_active);
	wake_up(&mddev->recovery_wait);
	if (!ok) {
		set_bit(MD_RECOVERY_ERR, &mddev->recovery);
		md_wakeup_thread(mddev->thread);
		// stop recovery, signal do_sync ....
	}
}


3496 3497
/* md_write_start(mddev, bi)
 * If we need to update some array metadata (e.g. 'active' flag
3498 3499
 * in superblock) before writing, schedule a superblock update
 * and wait for it to complete.
3500
 */
3501
void md_write_start(mddev_t *mddev, struct bio *bi)
L
Linus Torvalds 已提交
3502
{
3503
	if (bio_data_dir(bi) != WRITE)
3504
		return;
3505 3506 3507

	atomic_inc(&mddev->writes_pending);
	if (mddev->in_sync) {
3508 3509 3510 3511 3512 3513 3514
		spin_lock(&mddev->write_lock);
		if (mddev->in_sync) {
			mddev->in_sync = 0;
			mddev->sb_dirty = 1;
			md_wakeup_thread(mddev->thread);
		}
		spin_unlock(&mddev->write_lock);
3515
	}
3516
	wait_event(mddev->sb_wait, mddev->sb_dirty==0);
L
Linus Torvalds 已提交
3517 3518 3519 3520 3521 3522 3523 3524 3525 3526 3527 3528
}

void md_write_end(mddev_t *mddev)
{
	if (atomic_dec_and_test(&mddev->writes_pending)) {
		if (mddev->safemode == 2)
			md_wakeup_thread(mddev->thread);
		else
			mod_timer(&mddev->safemode_timer, jiffies + mddev->safemode_delay);
	}
}

A
Adrian Bunk 已提交
3529
static DECLARE_WAIT_QUEUE_HEAD(resync_wait);
L
Linus Torvalds 已提交
3530 3531 3532 3533 3534 3535 3536 3537

#define SYNC_MARKS	10
#define	SYNC_MARK_STEP	(3*HZ)
static void md_do_sync(mddev_t *mddev)
{
	mddev_t *mddev2;
	unsigned int currspeed = 0,
		 window;
3538
	sector_t max_sectors,j, io_sectors;
L
Linus Torvalds 已提交
3539 3540 3541 3542 3543
	unsigned long mark[SYNC_MARKS];
	sector_t mark_cnt[SYNC_MARKS];
	int last_mark,m;
	struct list_head *tmp;
	sector_t last_check;
3544
	int skipped = 0;
L
Linus Torvalds 已提交
3545 3546 3547 3548 3549 3550 3551 3552 3553 3554 3555 3556 3557 3558 3559 3560 3561 3562 3563 3564 3565 3566 3567 3568 3569 3570 3571 3572 3573 3574 3575 3576 3577 3578 3579 3580 3581 3582 3583 3584 3585 3586 3587 3588 3589 3590 3591 3592 3593 3594 3595 3596 3597 3598 3599 3600 3601 3602 3603 3604 3605 3606 3607 3608

	/* just incase thread restarts... */
	if (test_bit(MD_RECOVERY_DONE, &mddev->recovery))
		return;

	/* we overload curr_resync somewhat here.
	 * 0 == not engaged in resync at all
	 * 2 == checking that there is no conflict with another sync
	 * 1 == like 2, but have yielded to allow conflicting resync to
	 *		commense
	 * other == active in resync - this many blocks
	 *
	 * Before starting a resync we must have set curr_resync to
	 * 2, and then checked that every "conflicting" array has curr_resync
	 * less than ours.  When we find one that is the same or higher
	 * we wait on resync_wait.  To avoid deadlock, we reduce curr_resync
	 * to 1 if we choose to yield (based arbitrarily on address of mddev structure).
	 * This will mean we have to start checking from the beginning again.
	 *
	 */

	do {
		mddev->curr_resync = 2;

	try_again:
		if (signal_pending(current)) {
			flush_signals(current);
			goto skip;
		}
		ITERATE_MDDEV(mddev2,tmp) {
			if (mddev2 == mddev)
				continue;
			if (mddev2->curr_resync && 
			    match_mddev_units(mddev,mddev2)) {
				DEFINE_WAIT(wq);
				if (mddev < mddev2 && mddev->curr_resync == 2) {
					/* arbitrarily yield */
					mddev->curr_resync = 1;
					wake_up(&resync_wait);
				}
				if (mddev > mddev2 && mddev->curr_resync == 1)
					/* no need to wait here, we can wait the next
					 * time 'round when curr_resync == 2
					 */
					continue;
				prepare_to_wait(&resync_wait, &wq, TASK_INTERRUPTIBLE);
				if (!signal_pending(current)
				    && mddev2->curr_resync >= mddev->curr_resync) {
					printk(KERN_INFO "md: delaying resync of %s"
					       " until %s has finished resync (they"
					       " share one or more physical units)\n",
					       mdname(mddev), mdname(mddev2));
					mddev_put(mddev2);
					schedule();
					finish_wait(&resync_wait, &wq);
					goto try_again;
				}
				finish_wait(&resync_wait, &wq);
			}
		}
	} while (mddev->curr_resync < 2);

	if (test_bit(MD_RECOVERY_SYNC, &mddev->recovery))
		/* resync follows the size requested by the personality,
3609
		 * which defaults to physical size, but can be virtual size
L
Linus Torvalds 已提交
3610 3611 3612 3613 3614 3615 3616 3617 3618
		 */
		max_sectors = mddev->resync_max_sectors;
	else
		/* recovery follows the physical size of devices */
		max_sectors = mddev->size << 1;

	printk(KERN_INFO "md: syncing RAID array %s\n", mdname(mddev));
	printk(KERN_INFO "md: minimum _guaranteed_ reconstruction speed:"
		" %d KB/sec/disc.\n", sysctl_speed_limit_min);
3619
	printk(KERN_INFO "md: using maximum available idle IO bandwidth "
L
Linus Torvalds 已提交
3620 3621 3622 3623
	       "(but not more than %d KB/sec) for reconstruction.\n",
	       sysctl_speed_limit_max);

	is_mddev_idle(mddev); /* this also initializes IO event counters */
3624 3625
	/* we don't use the checkpoint if there's a bitmap */
	if (test_bit(MD_RECOVERY_SYNC, &mddev->recovery) && !mddev->bitmap)
L
Linus Torvalds 已提交
3626 3627 3628
		j = mddev->recovery_cp;
	else
		j = 0;
3629
	io_sectors = 0;
L
Linus Torvalds 已提交
3630 3631
	for (m = 0; m < SYNC_MARKS; m++) {
		mark[m] = jiffies;
3632
		mark_cnt[m] = io_sectors;
L
Linus Torvalds 已提交
3633 3634 3635 3636 3637 3638 3639 3640 3641 3642 3643 3644 3645 3646 3647 3648 3649 3650 3651 3652 3653 3654 3655 3656
	}
	last_mark = 0;
	mddev->resync_mark = mark[last_mark];
	mddev->resync_mark_cnt = mark_cnt[last_mark];

	/*
	 * Tune reconstruction:
	 */
	window = 32*(PAGE_SIZE/512);
	printk(KERN_INFO "md: using %dk window, over a total of %llu blocks.\n",
		window/2,(unsigned long long) max_sectors/2);

	atomic_set(&mddev->recovery_active, 0);
	init_waitqueue_head(&mddev->recovery_wait);
	last_check = 0;

	if (j>2) {
		printk(KERN_INFO 
			"md: resuming recovery of %s from checkpoint.\n",
			mdname(mddev));
		mddev->curr_resync = j;
	}

	while (j < max_sectors) {
3657
		sector_t sectors;
L
Linus Torvalds 已提交
3658

3659 3660 3661 3662
		skipped = 0;
		sectors = mddev->pers->sync_request(mddev, j, &skipped,
					    currspeed < sysctl_speed_limit_min);
		if (sectors == 0) {
L
Linus Torvalds 已提交
3663 3664 3665
			set_bit(MD_RECOVERY_ERR, &mddev->recovery);
			goto out;
		}
3666 3667 3668 3669 3670 3671

		if (!skipped) { /* actual IO requested */
			io_sectors += sectors;
			atomic_add(sectors, &mddev->recovery_active);
		}

L
Linus Torvalds 已提交
3672 3673 3674
		j += sectors;
		if (j>1) mddev->curr_resync = j;

3675 3676

		if (last_check + window > io_sectors || j == max_sectors)
L
Linus Torvalds 已提交
3677 3678
			continue;

3679
		last_check = io_sectors;
L
Linus Torvalds 已提交
3680 3681 3682 3683 3684 3685 3686 3687 3688 3689 3690 3691 3692

		if (test_bit(MD_RECOVERY_INTR, &mddev->recovery) ||
		    test_bit(MD_RECOVERY_ERR, &mddev->recovery))
			break;

	repeat:
		if (time_after_eq(jiffies, mark[last_mark] + SYNC_MARK_STEP )) {
			/* step marks */
			int next = (last_mark+1) % SYNC_MARKS;

			mddev->resync_mark = mark[next];
			mddev->resync_mark_cnt = mark_cnt[next];
			mark[next] = jiffies;
3693
			mark_cnt[next] = io_sectors - atomic_read(&mddev->recovery_active);
L
Linus Torvalds 已提交
3694 3695 3696 3697 3698 3699 3700 3701 3702 3703 3704 3705 3706 3707 3708 3709 3710 3711 3712 3713 3714 3715 3716 3717 3718 3719
			last_mark = next;
		}


		if (signal_pending(current)) {
			/*
			 * got a signal, exit.
			 */
			printk(KERN_INFO 
				"md: md_do_sync() got signal ... exiting\n");
			flush_signals(current);
			set_bit(MD_RECOVERY_INTR, &mddev->recovery);
			goto out;
		}

		/*
		 * this loop exits only if either when we are slower than
		 * the 'hard' speed limit, or the system was IO-idle for
		 * a jiffy.
		 * the system might be non-idle CPU-wise, but we only care
		 * about not overloading the IO subsystem. (things like an
		 * e2fsck being done on the RAID array should execute fast)
		 */
		mddev->queue->unplug_fn(mddev->queue);
		cond_resched();

3720 3721
		currspeed = ((unsigned long)(io_sectors-mddev->resync_mark_cnt))/2
			/((jiffies-mddev->resync_mark)/HZ +1) +1;
L
Linus Torvalds 已提交
3722 3723 3724 3725 3726 3727 3728 3729 3730 3731 3732 3733 3734 3735 3736 3737 3738 3739 3740

		if (currspeed > sysctl_speed_limit_min) {
			if ((currspeed > sysctl_speed_limit_max) ||
					!is_mddev_idle(mddev)) {
				msleep_interruptible(250);
				goto repeat;
			}
		}
	}
	printk(KERN_INFO "md: %s: sync done.\n",mdname(mddev));
	/*
	 * this also signals 'finished resyncing' to md_stop
	 */
 out:
	mddev->queue->unplug_fn(mddev->queue);

	wait_event(mddev->recovery_wait, !atomic_read(&mddev->recovery_active));

	/* tell personality that we are finished */
3741
	mddev->pers->sync_request(mddev, max_sectors, &skipped, 1);
L
Linus Torvalds 已提交
3742 3743 3744 3745 3746 3747 3748 3749 3750 3751 3752 3753 3754 3755 3756 3757 3758 3759 3760 3761 3762 3763 3764 3765 3766 3767 3768 3769 3770 3771 3772 3773 3774 3775 3776 3777 3778 3779 3780 3781 3782 3783 3784 3785 3786 3787 3788 3789 3790

	if (!test_bit(MD_RECOVERY_ERR, &mddev->recovery) &&
	    mddev->curr_resync > 2 &&
	    mddev->curr_resync >= mddev->recovery_cp) {
		if (test_bit(MD_RECOVERY_INTR, &mddev->recovery)) {
			printk(KERN_INFO 
				"md: checkpointing recovery of %s.\n",
				mdname(mddev));
			mddev->recovery_cp = mddev->curr_resync;
		} else
			mddev->recovery_cp = MaxSector;
	}

 skip:
	mddev->curr_resync = 0;
	wake_up(&resync_wait);
	set_bit(MD_RECOVERY_DONE, &mddev->recovery);
	md_wakeup_thread(mddev->thread);
}


/*
 * This routine is regularly called by all per-raid-array threads to
 * deal with generic issues like resync and super-block update.
 * Raid personalities that don't have a thread (linear/raid0) do not
 * need this as they never do any recovery or update the superblock.
 *
 * It does not do any resync itself, but rather "forks" off other threads
 * to do that as needed.
 * When it is determined that resync is needed, we set MD_RECOVERY_RUNNING in
 * "->recovery" and create a thread at ->sync_thread.
 * When the thread finishes it sets MD_RECOVERY_DONE (and might set MD_RECOVERY_ERR)
 * and wakeups up this thread which will reap the thread and finish up.
 * This thread also removes any faulty devices (with nr_pending == 0).
 *
 * The overall approach is:
 *  1/ if the superblock needs updating, update it.
 *  2/ If a recovery thread is running, don't do anything else.
 *  3/ If recovery has finished, clean up, possibly marking spares active.
 *  4/ If there are any faulty devices, remove them.
 *  5/ If array is degraded, try to add spares devices
 *  6/ If array has spares or is not in-sync, start a resync thread.
 */
void md_check_recovery(mddev_t *mddev)
{
	mdk_rdev_t *rdev;
	struct list_head *rtmp;


3791 3792
	if (mddev->bitmap)
		bitmap_daemon_work(mddev->bitmap);
L
Linus Torvalds 已提交
3793 3794 3795

	if (mddev->ro)
		return;
3796 3797 3798 3799 3800 3801 3802 3803 3804 3805

	if (signal_pending(current)) {
		if (mddev->pers->sync_request) {
			printk(KERN_INFO "md: %s in immediate safe mode\n",
			       mdname(mddev));
			mddev->safemode = 2;
		}
		flush_signals(current);
	}

L
Linus Torvalds 已提交
3806 3807 3808
	if ( ! (
		mddev->sb_dirty ||
		test_bit(MD_RECOVERY_NEEDED, &mddev->recovery) ||
3809 3810 3811 3812
		test_bit(MD_RECOVERY_DONE, &mddev->recovery) ||
		(mddev->safemode == 1) ||
		(mddev->safemode == 2 && ! atomic_read(&mddev->writes_pending)
		 && !mddev->in_sync && mddev->recovery_cp == MaxSector)
L
Linus Torvalds 已提交
3813 3814
		))
		return;
3815

L
Linus Torvalds 已提交
3816 3817
	if (mddev_trylock(mddev)==0) {
		int spares =0;
3818

3819
		spin_lock(&mddev->write_lock);
3820 3821 3822 3823 3824 3825 3826
		if (mddev->safemode && !atomic_read(&mddev->writes_pending) &&
		    !mddev->in_sync && mddev->recovery_cp == MaxSector) {
			mddev->in_sync = 1;
			mddev->sb_dirty = 1;
		}
		if (mddev->safemode == 1)
			mddev->safemode = 0;
3827
		spin_unlock(&mddev->write_lock);
3828

L
Linus Torvalds 已提交
3829 3830
		if (mddev->sb_dirty)
			md_update_sb(mddev);
3831 3832


L
Linus Torvalds 已提交
3833 3834 3835 3836 3837 3838 3839 3840 3841 3842 3843 3844 3845 3846 3847 3848 3849
		if (test_bit(MD_RECOVERY_RUNNING, &mddev->recovery) &&
		    !test_bit(MD_RECOVERY_DONE, &mddev->recovery)) {
			/* resync/recovery still happening */
			clear_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
			goto unlock;
		}
		if (mddev->sync_thread) {
			/* resync has finished, collect result */
			md_unregister_thread(mddev->sync_thread);
			mddev->sync_thread = NULL;
			if (!test_bit(MD_RECOVERY_ERR, &mddev->recovery) &&
			    !test_bit(MD_RECOVERY_INTR, &mddev->recovery)) {
				/* success...*/
				/* activate any spares */
				mddev->pers->spare_active(mddev);
			}
			md_update_sb(mddev);
3850 3851 3852 3853 3854 3855 3856 3857

			/* if array is no-longer degraded, then any saved_raid_disk
			 * information must be scrapped
			 */
			if (!mddev->degraded)
				ITERATE_RDEV(mddev,rdev,rtmp)
					rdev->saved_raid_disk = -1;

L
Linus Torvalds 已提交
3858 3859 3860 3861 3862 3863 3864 3865 3866 3867 3868 3869 3870 3871 3872 3873 3874 3875 3876 3877 3878 3879 3880 3881 3882 3883 3884 3885 3886 3887 3888 3889 3890 3891 3892 3893 3894 3895 3896 3897 3898 3899
			mddev->recovery = 0;
			/* flag recovery needed just to double check */
			set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
			goto unlock;
		}
		if (mddev->recovery)
			/* probably just the RECOVERY_NEEDED flag */
			mddev->recovery = 0;

		/* no recovery is running.
		 * remove any failed drives, then
		 * add spares if possible.
		 * Spare are also removed and re-added, to allow
		 * the personality to fail the re-add.
		 */
		ITERATE_RDEV(mddev,rdev,rtmp)
			if (rdev->raid_disk >= 0 &&
			    (rdev->faulty || ! rdev->in_sync) &&
			    atomic_read(&rdev->nr_pending)==0) {
				if (mddev->pers->hot_remove_disk(mddev, rdev->raid_disk)==0)
					rdev->raid_disk = -1;
			}

		if (mddev->degraded) {
			ITERATE_RDEV(mddev,rdev,rtmp)
				if (rdev->raid_disk < 0
				    && !rdev->faulty) {
					if (mddev->pers->hot_add_disk(mddev,rdev))
						spares++;
					else
						break;
				}
		}

		if (!spares && (mddev->recovery_cp == MaxSector )) {
			/* nothing we can do ... */
			goto unlock;
		}
		if (mddev->pers->sync_request) {
			set_bit(MD_RECOVERY_RUNNING, &mddev->recovery);
			if (!spares)
				set_bit(MD_RECOVERY_SYNC, &mddev->recovery);
3900 3901 3902 3903 3904 3905 3906
			if (spares && mddev->bitmap && ! mddev->bitmap->file) {
				/* We are adding a device or devices to an array
				 * which has the bitmap stored on all devices.
				 * So make sure all bitmap pages get written
				 */
				bitmap_write_all(mddev->bitmap);
			}
L
Linus Torvalds 已提交
3907 3908 3909 3910 3911 3912 3913 3914 3915 3916 3917 3918 3919 3920 3921 3922 3923 3924
			mddev->sync_thread = md_register_thread(md_do_sync,
								mddev,
								"%s_resync");
			if (!mddev->sync_thread) {
				printk(KERN_ERR "%s: could not start resync"
					" thread...\n", 
					mdname(mddev));
				/* leave the spares where they are, it shouldn't hurt */
				mddev->recovery = 0;
			} else {
				md_wakeup_thread(mddev->sync_thread);
			}
		}
	unlock:
		mddev_unlock(mddev);
	}
}

A
Adrian Bunk 已提交
3925 3926
static int md_notify_reboot(struct notifier_block *this,
			    unsigned long code, void *x)
L
Linus Torvalds 已提交
3927 3928 3929 3930 3931 3932 3933 3934 3935 3936 3937 3938 3939 3940 3941 3942 3943 3944 3945 3946 3947 3948
{
	struct list_head *tmp;
	mddev_t *mddev;

	if ((code == SYS_DOWN) || (code == SYS_HALT) || (code == SYS_POWER_OFF)) {

		printk(KERN_INFO "md: stopping all md devices.\n");

		ITERATE_MDDEV(mddev,tmp)
			if (mddev_trylock(mddev)==0)
				do_md_stop (mddev, 1);
		/*
		 * certain more exotic SCSI devices are known to be
		 * volatile wrt too early system reboots. While the
		 * right place to handle this issue is the given
		 * driver, we do want to have a safe RAID driver ...
		 */
		mdelay(1000*1);
	}
	return NOTIFY_DONE;
}

A
Adrian Bunk 已提交
3949
static struct notifier_block md_notifier = {
L
Linus Torvalds 已提交
3950 3951 3952 3953 3954 3955 3956 3957 3958 3959 3960 3961 3962 3963 3964 3965
	.notifier_call	= md_notify_reboot,
	.next		= NULL,
	.priority	= INT_MAX, /* before any real devices */
};

static void md_geninit(void)
{
	struct proc_dir_entry *p;

	dprintk("md: sizeof(mdp_super_t) = %d\n", (int)sizeof(mdp_super_t));

	p = create_proc_entry("mdstat", S_IRUGO, NULL);
	if (p)
		p->proc_fops = &md_seq_fops;
}

A
Adrian Bunk 已提交
3966
static int __init md_init(void)
L
Linus Torvalds 已提交
3967 3968 3969 3970 3971 3972 3973
{
	int minor;

	printk(KERN_INFO "md: md driver %d.%d.%d MAX_MD_DEVS=%d,"
			" MD_SB_DISKS=%d\n",
			MD_MAJOR_VERSION, MD_MINOR_VERSION,
			MD_PATCHLEVEL_VERSION, MAX_MD_DEVS, MD_SB_DISKS);
3974 3975
	printk(KERN_INFO "md: bitmap version %d.%d\n", BITMAP_MAJOR,
			BITMAP_MINOR);
L
Linus Torvalds 已提交
3976 3977 3978 3979 3980 3981 3982 3983 3984 3985 3986 3987 3988 3989 3990 3991 3992 3993 3994 3995 3996 3997 3998 3999 4000 4001 4002 4003 4004 4005 4006 4007 4008 4009 4010 4011 4012 4013 4014 4015 4016 4017 4018 4019 4020 4021 4022 4023 4024 4025 4026 4027 4028 4029 4030 4031 4032 4033 4034 4035 4036 4037 4038 4039 4040 4041 4042 4043 4044 4045 4046 4047 4048 4049 4050 4051 4052 4053 4054 4055 4056 4057 4058 4059 4060 4061 4062 4063 4064 4065 4066 4067 4068 4069 4070 4071 4072 4073 4074 4075 4076 4077 4078 4079 4080 4081 4082 4083 4084 4085 4086 4087 4088 4089 4090 4091 4092 4093 4094 4095 4096

	if (register_blkdev(MAJOR_NR, "md"))
		return -1;
	if ((mdp_major=register_blkdev(0, "mdp"))<=0) {
		unregister_blkdev(MAJOR_NR, "md");
		return -1;
	}
	devfs_mk_dir("md");
	blk_register_region(MKDEV(MAJOR_NR, 0), MAX_MD_DEVS, THIS_MODULE,
				md_probe, NULL, NULL);
	blk_register_region(MKDEV(mdp_major, 0), MAX_MD_DEVS<<MdpMinorShift, THIS_MODULE,
			    md_probe, NULL, NULL);

	for (minor=0; minor < MAX_MD_DEVS; ++minor)
		devfs_mk_bdev(MKDEV(MAJOR_NR, minor),
				S_IFBLK|S_IRUSR|S_IWUSR,
				"md/%d", minor);

	for (minor=0; minor < MAX_MD_DEVS; ++minor)
		devfs_mk_bdev(MKDEV(mdp_major, minor<<MdpMinorShift),
			      S_IFBLK|S_IRUSR|S_IWUSR,
			      "md/mdp%d", minor);


	register_reboot_notifier(&md_notifier);
	raid_table_header = register_sysctl_table(raid_root_table, 1);

	md_geninit();
	return (0);
}


#ifndef MODULE

/*
 * Searches all registered partitions for autorun RAID arrays
 * at boot time.
 */
static dev_t detected_devices[128];
static int dev_cnt;

void md_autodetect_dev(dev_t dev)
{
	if (dev_cnt >= 0 && dev_cnt < 127)
		detected_devices[dev_cnt++] = dev;
}


static void autostart_arrays(int part)
{
	mdk_rdev_t *rdev;
	int i;

	printk(KERN_INFO "md: Autodetecting RAID arrays.\n");

	for (i = 0; i < dev_cnt; i++) {
		dev_t dev = detected_devices[i];

		rdev = md_import_device(dev,0, 0);
		if (IS_ERR(rdev))
			continue;

		if (rdev->faulty) {
			MD_BUG();
			continue;
		}
		list_add(&rdev->same_set, &pending_raid_disks);
	}
	dev_cnt = 0;

	autorun_devices(part);
}

#endif

static __exit void md_exit(void)
{
	mddev_t *mddev;
	struct list_head *tmp;
	int i;
	blk_unregister_region(MKDEV(MAJOR_NR,0), MAX_MD_DEVS);
	blk_unregister_region(MKDEV(mdp_major,0), MAX_MD_DEVS << MdpMinorShift);
	for (i=0; i < MAX_MD_DEVS; i++)
		devfs_remove("md/%d", i);
	for (i=0; i < MAX_MD_DEVS; i++)
		devfs_remove("md/d%d", i);

	devfs_remove("md");

	unregister_blkdev(MAJOR_NR,"md");
	unregister_blkdev(mdp_major, "mdp");
	unregister_reboot_notifier(&md_notifier);
	unregister_sysctl_table(raid_table_header);
	remove_proc_entry("mdstat", NULL);
	ITERATE_MDDEV(mddev,tmp) {
		struct gendisk *disk = mddev->gendisk;
		if (!disk)
			continue;
		export_array(mddev);
		del_gendisk(disk);
		put_disk(disk);
		mddev->gendisk = NULL;
		mddev_put(mddev);
	}
}

module_init(md_init)
module_exit(md_exit)

EXPORT_SYMBOL(register_md_personality);
EXPORT_SYMBOL(unregister_md_personality);
EXPORT_SYMBOL(md_error);
EXPORT_SYMBOL(md_done_sync);
EXPORT_SYMBOL(md_write_start);
EXPORT_SYMBOL(md_write_end);
EXPORT_SYMBOL(md_register_thread);
EXPORT_SYMBOL(md_unregister_thread);
EXPORT_SYMBOL(md_wakeup_thread);
EXPORT_SYMBOL(md_print_devices);
EXPORT_SYMBOL(md_check_recovery);
MODULE_LICENSE("GPL");
4097
MODULE_ALIAS("md");
4098
MODULE_ALIAS_BLOCKDEV_MAJOR(MD_MAJOR);