md.c 94.9 KB
Newer Older
Linus Torvalds's avatar
Linus Torvalds committed
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.

Linus Torvalds's avatar
Linus Torvalds committed
25 26 27 28 29 30 31 32 33 34 35 36 37 38
   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>
#include <linux/linkage.h>
#include <linux/raid/md.h>
39
#include <linux/raid/bitmap.h>
Linus Torvalds's avatar
Linus Torvalds committed
40 41 42 43 44 45 46
#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>

47 48
#include <linux/file.h>

Linus Torvalds's avatar
Linus Torvalds committed
49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197
#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
 * the RAID driver will use the maximum available bandwith if the IO
 * 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);
198
			kfree(new);
Linus Torvalds's avatar
Linus Torvalds committed
199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225
			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);
226
	spin_lock_init(&new->write_lock);
227
	init_waitqueue_head(&new->sb_wait);
Linus Torvalds's avatar
Linus Torvalds committed
228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258

	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);

259
	md_wakeup_thread(mddev->thread);
Linus Torvalds's avatar
Linus Torvalds committed
260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285
}

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;
}

286
static inline sector_t calc_dev_sboffset(struct block_device *bdev)
Linus Torvalds's avatar
Linus Torvalds committed
287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328
{
	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;
	}
}


329 330 331 332 333 334 335 336 337 338 339
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);
Neil Brown's avatar
Neil Brown committed
340
	bio_put(bio);
341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363
	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);
}

Linus Torvalds's avatar
Linus Torvalds committed
364 365 366 367 368 369 370 371 372
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;
}

373
int sync_page_io(struct block_device *bdev, sector_t sector, int size,
Linus Torvalds's avatar
Linus Torvalds committed
374 375
		   struct page *page, int rw)
{
376
	struct bio *bio = bio_alloc(GFP_NOIO, 1);
Linus Torvalds's avatar
Linus Torvalds committed
377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459
	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;
}

static int read_disk_sb(mdk_rdev_t * rdev)
{
	char b[BDEVNAME_SIZE];
	if (!rdev->sb_page) {
		MD_BUG();
		return -EINVAL;
	}
	if (rdev->sb_loaded)
		return 0;


	if (!sync_page_io(rdev->bdev, rdev->sb_offset<<1, MD_SB_BYTES, rdev->sb_page, READ))
		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:
460 461
	kfree(tmp1);
	kfree(tmp2);
Linus Torvalds's avatar
Linus Torvalds committed
462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609
	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;

	ret = read_disk_sb(rdev);
	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;

	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);

610 611
	rdev->raid_disk = -1;
	rdev->in_sync = 0;
Linus Torvalds's avatar
Linus Torvalds committed
612 613 614 615 616 617 618 619 620 621 622 623 624
	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);
625
		mddev->bitmap_offset = 0;
Linus Torvalds's avatar
Linus Torvalds committed
626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642

		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;
643 644 645 646 647 648 649 650 651 652 653

		if (sb->state & (1<<MD_SB_BITMAP_PRESENT) &&
		    mddev->bitmap_file == NULL) {
			if (mddev->level != 1) {
				/* FIXME use a better test */
				printk(KERN_WARNING "md: bitmaps only support for raid1\n");
				return -EINVAL;
			}
			mddev->bitmap_offset = (MD_SB_BYTES >> 9);
		}

654 655 656
	} else if (mddev->pers == NULL) {
		/* Insist on good event counter while assembling */
		__u64 ev1 = md_event(sb);
Linus Torvalds's avatar
Linus Torvalds committed
657 658 659
		++ev1;
		if (ev1 < mddev->events) 
			return -EINVAL;
660 661 662 663 664 665 666 667 668 669
	} 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;

Linus Torvalds's avatar
Linus Torvalds committed
670
	if (mddev->level != LEVEL_MULTIPATH) {
671
		rdev->faulty = 0;
Linus Torvalds's avatar
Linus Torvalds committed
672 673 674 675 676 677 678 679 680
		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;
		}
681 682
	} else /* MULTIPATH are always insync */
		rdev->in_sync = 1;
Linus Torvalds's avatar
Linus Torvalds committed
683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746
	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;

	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;

747 748 749
	if (mddev->bitmap && mddev->bitmap_file == NULL)
		sb->state |= (1<<MD_SB_BITMAP_PRESENT);

Linus Torvalds's avatar
Linus Torvalds committed
750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846
	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++;
		}
	}
	
	/* 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];

	/*
	 * 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;
847
		sb_offset &= ~(sector_t)(4*2-1);
Linus Torvalds's avatar
Linus Torvalds committed
848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926
		/* 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;

	ret = read_disk_sb(rdev);
	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) ||
	    sb->feature_map != 0)
		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);

	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);

927 928
	rdev->raid_disk = -1;
	rdev->in_sync = 0;
Linus Torvalds's avatar
Linus Torvalds committed
929 930 931 932 933 934 935 936 937 938 939 940
	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);
941
		mddev->bitmap_offset = 0;
Linus Torvalds's avatar
Linus Torvalds committed
942 943 944 945 946
		
		mddev->recovery_cp = le64_to_cpu(sb->resync_offset);
		memcpy(mddev->uuid, sb->set_uuid, 16);

		mddev->max_disks =  (4096-256)/2;
947 948 949 950 951 952 953 954 955

		if ((le32_to_cpu(sb->feature_map) & 1) &&
		    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);
		}
956 957 958
	} else if (mddev->pers == NULL) {
		/* Insist of good event counter while assembling */
		__u64 ev1 = le64_to_cpu(sb->events);
Linus Torvalds's avatar
Linus Torvalds committed
959 960 961
		++ev1;
		if (ev1 < mddev->events)
			return -EINVAL;
962 963 964 965 966 967 968 969 970
	} 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;
Linus Torvalds's avatar
Linus Torvalds committed
971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988

	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;
		}
989 990 991
	} else /* MULTIPATH are always insync */
		rdev->in_sync = 1;

Linus Torvalds's avatar
Linus Torvalds committed
992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017
	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);

1018 1019 1020 1021 1022
	if (mddev->bitmap && mddev->bitmap_file == NULL) {
		sb->bitmap_offset = cpu_to_le32((__u32)mddev->bitmap_offset);
		sb->feature_map = cpu_to_le32(1);
	}

Linus Torvalds's avatar
Linus Torvalds committed
1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046
	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);
}


Adrian Bunk's avatar
Adrian Bunk committed
1047
static struct super_type super_types[] = {
Linus Torvalds's avatar
Linus Torvalds committed
1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 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
	[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) {

1287 1288 1289 1290
		if (mddev->bitmap)
			bitmap_print_sb(mddev->bitmap);
		else
			printk("%s: ", mdname(mddev));
Linus Torvalds's avatar
Linus Torvalds committed
1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316
		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)
{
1317
	int err;
Linus Torvalds's avatar
Linus Torvalds committed
1318 1319
	struct list_head *tmp;
	mdk_rdev_t *rdev;
1320
	int sync_req;
Linus Torvalds's avatar
Linus Torvalds committed
1321 1322

repeat:
1323 1324
	spin_lock(&mddev->write_lock);
	sync_req = mddev->in_sync;
Linus Torvalds's avatar
Linus Torvalds committed
1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336
	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 --;
	}
1337
	mddev->sb_dirty = 2;
Linus Torvalds's avatar
Linus Torvalds committed
1338 1339 1340 1341 1342 1343
	sync_sbs(mddev);

	/*
	 * do not write anything to disk if using
	 * nonpersistent superblocks
	 */
1344 1345 1346
	if (!mddev->persistent) {
		mddev->sb_dirty = 0;
		spin_unlock(&mddev->write_lock);
1347
		wake_up(&mddev->sb_wait);
Linus Torvalds's avatar
Linus Torvalds committed
1348
		return;
1349 1350
	}
	spin_unlock(&mddev->write_lock);
Linus Torvalds's avatar
Linus Torvalds committed
1351 1352 1353 1354 1355

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

1356
	err = bitmap_update_sb(mddev->bitmap);
Linus Torvalds's avatar
Linus Torvalds committed
1357 1358 1359 1360 1361 1362 1363 1364
	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) {
1365 1366 1367 1368 1369 1370 1371
			md_super_write(mddev,rdev,
				       rdev->sb_offset<<1, MD_SB_BYTES,
				       rdev->sb_page);
			dprintk(KERN_INFO "(write) %s's sb offset: %llu\n",
				bdevname(rdev->bdev,b),
				(unsigned long long)rdev->sb_offset);

Linus Torvalds's avatar
Linus Torvalds committed
1372 1373
		} else
			dprintk(")\n");
1374
		if (mddev->level == LEVEL_MULTIPATH)
Linus Torvalds's avatar
Linus Torvalds committed
1375 1376 1377
			/* only need to write one superblock... */
			break;
	}
1378 1379 1380
	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 */

1381
	spin_lock(&mddev->write_lock);
1382
	if (mddev->in_sync != sync_req|| mddev->sb_dirty == 1) {
1383 1384 1385 1386 1387 1388
		/* have to write it out again */
		spin_unlock(&mddev->write_lock);
		goto repeat;
	}
	mddev->sb_dirty = 0;
	spin_unlock(&mddev->write_lock);
1389
	wake_up(&mddev->sb_wait);
1390

Linus Torvalds's avatar
Linus Torvalds committed
1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473
}

/*
 * 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
 */


1474
static void analyze_sbs(mddev_t * mddev)
Linus Torvalds's avatar
Linus Torvalds committed
1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593
{
	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];

1594 1595
	if (list_empty(&mddev->disks))
		/* cannot run an array with no devices.. */
Linus Torvalds's avatar
Linus Torvalds committed
1596 1597 1598 1599 1600 1601 1602 1603
		return -EINVAL;

	if (mddev->pers)
		return -EBUSY;

	/*
	 * Analyze all RAID superblock(s)
	 */
1604 1605
	if (!mddev->raid_disks)
		analyze_sbs(mddev);
Linus Torvalds's avatar
Linus Torvalds committed
1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630

	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) {
1631
			printk(KERN_ERR "chunk_size of %d not valid\n", chunk_size);
Linus Torvalds's avatar
Linus Torvalds committed
1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691
			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);

1692
	mddev->recovery = 0;
Linus Torvalds's avatar
Linus Torvalds committed
1693 1694
	mddev->resync_max_sectors = mddev->size << 1; /* may be over-ridden by personality */

1695 1696 1697 1698 1699 1700 1701
	/* 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);
Linus Torvalds's avatar
Linus Torvalds committed
1702 1703 1704 1705
	if (err) {
		printk(KERN_ERR "md: pers->run() failed ...\n");
		module_put(mddev->pers->owner);
		mddev->pers = NULL;
1706 1707
		bitmap_destroy(mddev);
		return err;
Linus Torvalds's avatar
Linus Torvalds committed
1708 1709 1710 1711 1712 1713 1714 1715 1716
	}
 	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);
1717
	md_wakeup_thread(mddev->thread);
Linus Torvalds's avatar
Linus Torvalds committed
1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803
	
	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 {
1804 1805
			bitmap_flush(mddev);
			wait_event(mddev->sb_wait, atomic_read(&mddev->pending_writes)==0);
Linus Torvalds's avatar
Linus Torvalds committed
1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822
			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);
	}
1823 1824 1825 1826 1827 1828 1829

	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;
	}
1830
	mddev->bitmap_offset = 0;