bitmap.c 54.6 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
/*
 * bitmap.c two-level bitmap (C) Peter T. Breuer (ptb@ot.uc3m.es) 2003
 *
 * bitmap_create  - sets up the bitmap structure
 * bitmap_destroy - destroys the bitmap structure
 *
 * additions, Copyright (C) 2003-2004, Paul Clements, SteelEye Technology, Inc.:
 * - added disk storage for bitmap
 * - changes to allow various bitmap chunk sizes
 */

/*
 * Still to do:
 *
 * flush after percent set rather than just time based. (maybe both).
 */

18
#include <linux/blkdev.h>
19 20 21 22 23 24 25 26 27 28
#include <linux/module.h>
#include <linux/errno.h>
#include <linux/slab.h>
#include <linux/init.h>
#include <linux/timer.h>
#include <linux/sched.h>
#include <linux/list.h>
#include <linux/file.h>
#include <linux/mount.h>
#include <linux/buffer_head.h>
29
#include "md.h"
30
#include "bitmap.h"
31

32
#include <linux/dm-dirty-log.h>
33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61
/* debug macros */

#define DEBUG 0

#if DEBUG
/* these are for debugging purposes only! */

/* define one and only one of these */
#define INJECT_FAULTS_1 0 /* cause bitmap_alloc_page to fail always */
#define INJECT_FAULTS_2 0 /* cause bitmap file to be kicked when first bit set*/
#define INJECT_FAULTS_3 0 /* treat bitmap file as kicked at init time */
#define INJECT_FAULTS_4 0 /* undef */
#define INJECT_FAULTS_5 0 /* undef */
#define INJECT_FAULTS_6 0

/* if these are defined, the driver will fail! debug only */
#define INJECT_FATAL_FAULT_1 0 /* fail kmalloc, causing bitmap_create to fail */
#define INJECT_FATAL_FAULT_2 0 /* undef */
#define INJECT_FATAL_FAULT_3 0 /* undef */
#endif

#ifndef PRINTK
#  if DEBUG > 0
#    define PRINTK(x...) printk(KERN_DEBUG x)
#  else
#    define PRINTK(x...)
#  endif
#endif

62
static inline char *bmname(struct bitmap *bitmap)
63 64 65 66 67 68 69 70 71 72 73
{
	return bitmap->mddev ? mdname(bitmap->mddev) : "mdX";
}

/*
 * just a placeholder - calls kmalloc for bitmap pages
 */
static unsigned char *bitmap_alloc_page(struct bitmap *bitmap)
{
	unsigned char *page;

74
#ifdef INJECT_FAULTS_1
75 76
	page = NULL;
#else
77
	page = kzalloc(PAGE_SIZE, GFP_NOIO);
78 79 80 81
#endif
	if (!page)
		printk("%s: bitmap_alloc_page FAILED\n", bmname(bitmap));
	else
82
		PRINTK("%s: bitmap_alloc_page: allocated page at %p\n",
83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105
			bmname(bitmap), page);
	return page;
}

/*
 * for now just a placeholder -- just calls kfree for bitmap pages
 */
static void bitmap_free_page(struct bitmap *bitmap, unsigned char *page)
{
	PRINTK("%s: bitmap_free_page: free page %p\n", bmname(bitmap), page);
	kfree(page);
}

/*
 * check a page and, if necessary, allocate it (or hijack it if the alloc fails)
 *
 * 1) check to see if this page is allocated, if it's not then try to alloc
 * 2) if the alloc fails, set the page's hijacked flag so we'll use the
 *    page pointer directly as a counter
 *
 * if we find our page, we increment the page's refcount so that it stays
 * allocated while we're using it
 */
106 107
static int bitmap_checkpage(struct bitmap *bitmap,
			    unsigned long page, int create)
108 109
__releases(bitmap->lock)
__acquires(bitmap->lock)
110 111 112 113
{
	unsigned char *mappage;

	if (page >= bitmap->pages) {
114 115 116 117
		/* This can happen if bitmap_start_sync goes beyond
		 * End-of-device while looking for a whole page.
		 * It is harmless.
		 */
118 119 120 121 122 123 124 125 126 127 128 129 130 131
		return -EINVAL;
	}

	if (bitmap->bp[page].hijacked) /* it's hijacked, don't try to alloc */
		return 0;

	if (bitmap->bp[page].map) /* page is already allocated, just return */
		return 0;

	if (!create)
		return -ENOENT;

	/* this page has not been allocated yet */

132 133 134 135 136
	spin_unlock_irq(&bitmap->lock);
	mappage = bitmap_alloc_page(bitmap);
	spin_lock_irq(&bitmap->lock);

	if (mappage == NULL) {
137 138 139 140 141 142
		PRINTK("%s: bitmap map page allocation failed, hijacking\n",
			bmname(bitmap));
		/* failed - set the hijacked flag so that we can use the
		 * pointer as a counter */
		if (!bitmap->bp[page].map)
			bitmap->bp[page].hijacked = 1;
143 144
	} else if (bitmap->bp[page].map ||
		   bitmap->bp[page].hijacked) {
145 146 147
		/* somebody beat us to getting the page */
		bitmap_free_page(bitmap, mappage);
		return 0;
148
	} else {
149

150
		/* no page was in place and we have one, so install it */
151

152 153 154
		bitmap->bp[page].map = mappage;
		bitmap->missing_pages--;
	}
155 156 157 158 159 160
	return 0;
}

/* if page is completely empty, put it back on the free list, or dealloc it */
/* if page was hijacked, unmark the flag so it might get alloced next time */
/* Note: lock should be held when calling this */
161
static void bitmap_checkfree(struct bitmap *bitmap, unsigned long page)
162 163 164 165 166 167 168 169 170 171 172
{
	char *ptr;

	if (bitmap->bp[page].count) /* page is still busy */
		return;

	/* page is no longer in use, it can be released */

	if (bitmap->bp[page].hijacked) { /* page was hijacked, undo this now */
		bitmap->bp[page].hijacked = 0;
		bitmap->bp[page].map = NULL;
173 174 175 176 177 178
	} else {
		/* normal case, free the page */
		ptr = bitmap->bp[page].map;
		bitmap->bp[page].map = NULL;
		bitmap->missing_pages++;
		bitmap_free_page(bitmap, ptr);
179 180 181 182 183 184 185 186 187 188 189
	}
}

/*
 * bitmap file handling - read and write the bitmap file and its superblock
 */

/*
 * basic page I/O operations
 */

190
/* IO operations when bitmap is stored near all superblocks */
191
static struct page *read_sb_page(mddev_t *mddev, loff_t offset,
192 193
				 struct page *page,
				 unsigned long index, int size)
194 195 196 197 198
{
	/* choose a good rdev and read the page from there */

	mdk_rdev_t *rdev;
	sector_t target;
199
	int did_alloc = 0;
200

201
	if (!page) {
202
		page = alloc_page(GFP_KERNEL);
203 204 205 206
		if (!page)
			return ERR_PTR(-ENOMEM);
		did_alloc = 1;
	}
207

208
	list_for_each_entry(rdev, &mddev->disks, same_set) {
209 210
		if (! test_bit(In_sync, &rdev->flags)
		    || test_bit(Faulty, &rdev->flags))
211 212
			continue;

213
		target = rdev->sb_start + offset + index * (PAGE_SIZE/512);
214

215
		if (sync_page_io(rdev, target,
216
				 roundup(size, bdev_logical_block_size(rdev->bdev)),
217
				 page, READ)) {
218
			page->index = index;
219 220
			attach_page_buffers(page, NULL); /* so that free_buffer will
							  * quietly no-op */
221 222 223
			return page;
		}
	}
224 225
	if (did_alloc)
		put_page(page);
226
	return ERR_PTR(-EIO);
227 228 229

}

230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263
static mdk_rdev_t *next_active_rdev(mdk_rdev_t *rdev, mddev_t *mddev)
{
	/* Iterate the disks of an mddev, using rcu to protect access to the
	 * linked list, and raising the refcount of devices we return to ensure
	 * they don't disappear while in use.
	 * As devices are only added or removed when raid_disk is < 0 and
	 * nr_pending is 0 and In_sync is clear, the entries we return will
	 * still be in the same position on the list when we re-enter
	 * list_for_each_continue_rcu.
	 */
	struct list_head *pos;
	rcu_read_lock();
	if (rdev == NULL)
		/* start at the beginning */
		pos = &mddev->disks;
	else {
		/* release the previous rdev and start from there. */
		rdev_dec_pending(rdev, mddev);
		pos = &rdev->same_set;
	}
	list_for_each_continue_rcu(pos, &mddev->disks) {
		rdev = list_entry(pos, mdk_rdev_t, same_set);
		if (rdev->raid_disk >= 0 &&
		    !test_bit(Faulty, &rdev->flags)) {
			/* this is a usable devices */
			atomic_inc(&rdev->nr_pending);
			rcu_read_unlock();
			return rdev;
		}
	}
	rcu_read_unlock();
	return NULL;
}

264
static int write_sb_page(struct bitmap *bitmap, struct page *page, int wait)
265
{
266
	mdk_rdev_t *rdev = NULL;
267
	mddev_t *mddev = bitmap->mddev;
268

269
	while ((rdev = next_active_rdev(rdev, mddev)) != NULL) {
270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314
		int size = PAGE_SIZE;
		loff_t offset = mddev->bitmap_info.offset;
		if (page->index == bitmap->file_pages-1)
			size = roundup(bitmap->last_page_size,
				       bdev_logical_block_size(rdev->bdev));
		/* Just make sure we aren't corrupting data or
		 * metadata
		 */
		if (mddev->external) {
			/* Bitmap could be anywhere. */
			if (rdev->sb_start + offset + (page->index
						       * (PAGE_SIZE/512))
			    > rdev->data_offset
			    &&
			    rdev->sb_start + offset
			    < (rdev->data_offset + mddev->dev_sectors
			     + (PAGE_SIZE/512)))
				goto bad_alignment;
		} else if (offset < 0) {
			/* DATA  BITMAP METADATA  */
			if (offset
			    + (long)(page->index * (PAGE_SIZE/512))
			    + size/512 > 0)
				/* bitmap runs in to metadata */
				goto bad_alignment;
			if (rdev->data_offset + mddev->dev_sectors
			    > rdev->sb_start + offset)
				/* data runs in to bitmap */
				goto bad_alignment;
		} else if (rdev->sb_start < rdev->data_offset) {
			/* METADATA BITMAP DATA */
			if (rdev->sb_start
			    + offset
			    + page->index*(PAGE_SIZE/512) + size/512
			    > rdev->data_offset)
				/* bitmap runs in to data */
				goto bad_alignment;
		} else {
			/* DATA METADATA BITMAP - no problems */
		}
		md_super_write(mddev, rdev,
			       rdev->sb_start + offset
			       + page->index * (PAGE_SIZE/512),
			       size,
			       page);
315
	}
316 317

	if (wait)
318
		md_super_wait(mddev);
319
	return 0;
320 321 322

 bad_alignment:
	return -EINVAL;
323 324
}

325
static void bitmap_file_kick(struct bitmap *bitmap);
326
/*
327
 * write out a page to a file
328
 */
329
static void write_page(struct bitmap *bitmap, struct page *page, int wait)
330
{
331
	struct buffer_head *bh;
332

333 334 335 336 337
	if (bitmap->file == NULL) {
		switch (write_sb_page(bitmap, page, wait)) {
		case -EINVAL:
			bitmap->flags |= BITMAP_WRITE_ERROR;
		}
338
	} else {
339

340
		bh = page_buffers(page);
341

342 343 344 345 346 347 348
		while (bh && bh->b_blocknr) {
			atomic_inc(&bitmap->pending_writes);
			set_buffer_locked(bh);
			set_buffer_mapped(bh);
			submit_bh(WRITE, bh);
			bh = bh->b_this_page;
		}
349

350
		if (wait)
351 352
			wait_event(bitmap->write_wait,
				   atomic_read(&bitmap->pending_writes)==0);
353
	}
354 355
	if (bitmap->flags & BITMAP_WRITE_ERROR)
		bitmap_file_kick(bitmap);
356 357 358 359 360 361
}

static void end_bitmap_write(struct buffer_head *bh, int uptodate)
{
	struct bitmap *bitmap = bh->b_private;
	unsigned long flags;
362

363 364 365 366
	if (!uptodate) {
		spin_lock_irqsave(&bitmap->lock, flags);
		bitmap->flags |= BITMAP_WRITE_ERROR;
		spin_unlock_irqrestore(&bitmap->lock, flags);
367
	}
368 369 370
	if (atomic_dec_and_test(&bitmap->pending_writes))
		wake_up(&bitmap->write_wait);
}
371

372 373 374 375 376 377 378 379 380 381 382
/* copied from buffer.c */
static void
__clear_page_buffers(struct page *page)
{
	ClearPagePrivate(page);
	set_page_private(page, 0);
	page_cache_release(page);
}
static void free_buffers(struct page *page)
{
	struct buffer_head *bh = page_buffers(page);
383

384 385 386 387
	while (bh) {
		struct buffer_head *next = bh->b_this_page;
		free_buffer_head(bh);
		bh = next;
388
	}
389 390
	__clear_page_buffers(page);
	put_page(page);
391 392
}

393 394 395 396 397 398 399
/* read a page from a file.
 * We both read the page, and attach buffers to the page to record the
 * address of each block (using bmap).  These addresses will be used
 * to write the block later, completely bypassing the filesystem.
 * This usage is similar to how swap files are handled, and allows us
 * to write to a file with no concerns of memory allocation failing.
 */
400
static struct page *read_page(struct file *file, unsigned long index,
401 402
			      struct bitmap *bitmap,
			      unsigned long count)
403 404
{
	struct page *page = NULL;
J
Josef Sipek 已提交
405
	struct inode *inode = file->f_path.dentry->d_inode;
406 407
	struct buffer_head *bh;
	sector_t block;
408

409
	PRINTK("read bitmap file (%dB @ %llu)\n", (int)PAGE_SIZE,
410
			(unsigned long long)index << PAGE_SHIFT);
411

412 413 414
	page = alloc_page(GFP_KERNEL);
	if (!page)
		page = ERR_PTR(-ENOMEM);
415 416
	if (IS_ERR(page))
		goto out;
417 418 419

	bh = alloc_page_buffers(page, 1<<inode->i_blkbits, 0);
	if (!bh) {
420
		put_page(page);
421
		page = ERR_PTR(-ENOMEM);
422 423
		goto out;
	}
424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444
	attach_page_buffers(page, bh);
	block = index << (PAGE_SHIFT - inode->i_blkbits);
	while (bh) {
		if (count == 0)
			bh->b_blocknr = 0;
		else {
			bh->b_blocknr = bmap(inode, block);
			if (bh->b_blocknr == 0) {
				/* Cannot use this file! */
				free_buffers(page);
				page = ERR_PTR(-EINVAL);
				goto out;
			}
			bh->b_bdev = inode->i_sb->s_bdev;
			if (count < (1<<inode->i_blkbits))
				count = 0;
			else
				count -= (1<<inode->i_blkbits);

			bh->b_end_io = end_bitmap_write;
			bh->b_private = bitmap;
445 446 447 448
			atomic_inc(&bitmap->pending_writes);
			set_buffer_locked(bh);
			set_buffer_mapped(bh);
			submit_bh(READ, bh);
449 450 451 452 453
		}
		block++;
		bh = bh->b_this_page;
	}
	page->index = index;
454 455 456 457 458 459 460

	wait_event(bitmap->write_wait,
		   atomic_read(&bitmap->pending_writes)==0);
	if (bitmap->flags & BITMAP_WRITE_ERROR) {
		free_buffers(page);
		page = ERR_PTR(-EIO);
	}
461 462
out:
	if (IS_ERR(page))
463
		printk(KERN_ALERT "md: bitmap read error: (%dB @ %llu): %ld\n",
464 465
			(int)PAGE_SIZE,
			(unsigned long long)index << PAGE_SHIFT,
466 467 468 469 470 471 472 473 474
			PTR_ERR(page));
	return page;
}

/*
 * bitmap file superblock operations
 */

/* update the event counter and sync the superblock to disk */
475
void bitmap_update_sb(struct bitmap *bitmap)
476 477 478 479 480
{
	bitmap_super_t *sb;
	unsigned long flags;

	if (!bitmap || !bitmap->mddev) /* no bitmap for this array */
481
		return;
482 483
	if (bitmap->mddev->bitmap_info.external)
		return;
484 485 486
	spin_lock_irqsave(&bitmap->lock, flags);
	if (!bitmap->sb_page) { /* no superblock */
		spin_unlock_irqrestore(&bitmap->lock, flags);
487
		return;
488 489
	}
	spin_unlock_irqrestore(&bitmap->lock, flags);
490
	sb = kmap_atomic(bitmap->sb_page, KM_USER0);
491
	sb->events = cpu_to_le64(bitmap->mddev->events);
492 493 494 495 496
	if (bitmap->mddev->events < bitmap->events_cleared) {
		/* rocking back to read-only */
		bitmap->events_cleared = bitmap->mddev->events;
		sb->events_cleared = cpu_to_le64(bitmap->events_cleared);
	}
497 498 499
	/* Just in case these have been changed via sysfs: */
	sb->daemon_sleep = cpu_to_le32(bitmap->mddev->bitmap_info.daemon_sleep/HZ);
	sb->write_behind = cpu_to_le32(bitmap->mddev->bitmap_info.max_write_behind);
500
	kunmap_atomic(sb, KM_USER0);
501
	write_page(bitmap, bitmap->sb_page, 1);
502 503 504 505 506 507 508 509 510
}

/* print out the bitmap file superblock */
void bitmap_print_sb(struct bitmap *bitmap)
{
	bitmap_super_t *sb;

	if (!bitmap || !bitmap->sb_page)
		return;
511
	sb = kmap_atomic(bitmap->sb_page, KM_USER0);
512
	printk(KERN_DEBUG "%s: bitmap file superblock:\n", bmname(bitmap));
513 514 515
	printk(KERN_DEBUG "         magic: %08x\n", le32_to_cpu(sb->magic));
	printk(KERN_DEBUG "       version: %d\n", le32_to_cpu(sb->version));
	printk(KERN_DEBUG "          uuid: %08x.%08x.%08x.%08x\n",
516 517 518 519
					*(__u32 *)(sb->uuid+0),
					*(__u32 *)(sb->uuid+4),
					*(__u32 *)(sb->uuid+8),
					*(__u32 *)(sb->uuid+12));
520
	printk(KERN_DEBUG "        events: %llu\n",
521
			(unsigned long long) le64_to_cpu(sb->events));
522
	printk(KERN_DEBUG "events cleared: %llu\n",
523
			(unsigned long long) le64_to_cpu(sb->events_cleared));
524 525 526 527 528
	printk(KERN_DEBUG "         state: %08x\n", le32_to_cpu(sb->state));
	printk(KERN_DEBUG "     chunksize: %d B\n", le32_to_cpu(sb->chunksize));
	printk(KERN_DEBUG "  daemon sleep: %ds\n", le32_to_cpu(sb->daemon_sleep));
	printk(KERN_DEBUG "     sync size: %llu KB\n",
			(unsigned long long)le64_to_cpu(sb->sync_size)/2);
529
	printk(KERN_DEBUG "max write behind: %d\n", le32_to_cpu(sb->write_behind));
530
	kunmap_atomic(sb, KM_USER0);
531 532 533 534 535 536 537
}

/* read the superblock from the bitmap file and initialize some bitmap fields */
static int bitmap_read_sb(struct bitmap *bitmap)
{
	char *reason = NULL;
	bitmap_super_t *sb;
538
	unsigned long chunksize, daemon_sleep, write_behind;
539 540 541 542
	unsigned long long events;
	int err = -EINVAL;

	/* page 0 is the superblock, read it... */
543 544 545 546 547 548
	if (bitmap->file) {
		loff_t isize = i_size_read(bitmap->file->f_mapping->host);
		int bytes = isize > PAGE_SIZE ? PAGE_SIZE : isize;

		bitmap->sb_page = read_page(bitmap->file, 0, bitmap, bytes);
	} else {
549 550
		bitmap->sb_page = read_sb_page(bitmap->mddev,
					       bitmap->mddev->bitmap_info.offset,
551 552
					       NULL,
					       0, sizeof(bitmap_super_t));
553
	}
554 555 556 557 558 559
	if (IS_ERR(bitmap->sb_page)) {
		err = PTR_ERR(bitmap->sb_page);
		bitmap->sb_page = NULL;
		return err;
	}

560
	sb = kmap_atomic(bitmap->sb_page, KM_USER0);
561 562

	chunksize = le32_to_cpu(sb->chunksize);
563
	daemon_sleep = le32_to_cpu(sb->daemon_sleep) * HZ;
564
	write_behind = le32_to_cpu(sb->write_behind);
565 566 567 568

	/* verify that the bitmap-specific fields are valid */
	if (sb->magic != cpu_to_le32(BITMAP_MAGIC))
		reason = "bad magic";
569 570
	else if (le32_to_cpu(sb->version) < BITMAP_MAJOR_LO ||
		 le32_to_cpu(sb->version) > BITMAP_MAJOR_HI)
571
		reason = "unrecognized superblock version";
572
	else if (chunksize < 512)
573
		reason = "bitmap chunksize too small";
574 575
	else if ((1 << ffz(~chunksize)) != chunksize)
		reason = "bitmap chunksize not a power of 2";
576
	else if (daemon_sleep < 1 || daemon_sleep > MAX_SCHEDULE_TIMEOUT)
577
		reason = "daemon sleep period out of range";
578 579
	else if (write_behind > COUNTER_MAX)
		reason = "write-behind limit out of range (0 - 16383)";
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
	if (reason) {
		printk(KERN_INFO "%s: invalid bitmap file superblock: %s\n",
			bmname(bitmap), reason);
		goto out;
	}

	/* keep the array size field of the bitmap superblock up to date */
	sb->sync_size = cpu_to_le64(bitmap->mddev->resync_max_sectors);

	if (!bitmap->mddev->persistent)
		goto success;

	/*
	 * if we have a persistent array superblock, compare the
	 * bitmap's UUID and event counter to the mddev's
	 */
	if (memcmp(sb->uuid, bitmap->mddev->uuid, 16)) {
		printk(KERN_INFO "%s: bitmap superblock UUID mismatch\n",
			bmname(bitmap));
		goto out;
	}
	events = le64_to_cpu(sb->events);
	if (events < bitmap->mddev->events) {
		printk(KERN_INFO "%s: bitmap file is out of date (%llu < %llu) "
			"-- forcing full recovery\n", bmname(bitmap), events,
			(unsigned long long) bitmap->mddev->events);
606
		sb->state |= cpu_to_le32(BITMAP_STALE);
607 608 609
	}
success:
	/* assign fields using values from superblock */
610 611 612
	bitmap->mddev->bitmap_info.chunksize = chunksize;
	bitmap->mddev->bitmap_info.daemon_sleep = daemon_sleep;
	bitmap->mddev->bitmap_info.max_write_behind = write_behind;
613
	bitmap->flags |= le32_to_cpu(sb->state);
614 615
	if (le32_to_cpu(sb->version) == BITMAP_MAJOR_HOSTENDIAN)
		bitmap->flags |= BITMAP_HOSTENDIAN;
616
	bitmap->events_cleared = le64_to_cpu(sb->events_cleared);
617
	if (sb->state & cpu_to_le32(BITMAP_STALE))
618
		bitmap->events_cleared = bitmap->mddev->events;
619 620
	err = 0;
out:
621
	kunmap_atomic(sb, KM_USER0);
622 623 624 625 626 627 628 629 630 631
	if (err)
		bitmap_print_sb(bitmap);
	return err;
}

enum bitmap_mask_op {
	MASK_SET,
	MASK_UNSET
};

632 633 634
/* record the state of the bitmap in the superblock.  Return the old value */
static int bitmap_mask_state(struct bitmap *bitmap, enum bitmap_state bits,
			     enum bitmap_mask_op op)
635 636 637
{
	bitmap_super_t *sb;
	unsigned long flags;
638
	int old;
639 640

	spin_lock_irqsave(&bitmap->lock, flags);
641
	if (!bitmap->sb_page) { /* can't set the state */
642
		spin_unlock_irqrestore(&bitmap->lock, flags);
643
		return 0;
644 645
	}
	spin_unlock_irqrestore(&bitmap->lock, flags);
646
	sb = kmap_atomic(bitmap->sb_page, KM_USER0);
647
	old = le32_to_cpu(sb->state) & bits;
648
	switch (op) {
649 650 651 652 653 654 655 656
	case MASK_SET:
		sb->state |= cpu_to_le32(bits);
		break;
	case MASK_UNSET:
		sb->state &= cpu_to_le32(~bits);
		break;
	default:
		BUG();
657
	}
658
	kunmap_atomic(sb, KM_USER0);
659
	return old;
660 661 662 663 664 665
}

/*
 * general bitmap file operations
 */

666 667 668 669 670 671
/*
 * on-disk bitmap:
 *
 * Use one bit per "chunk" (block set). We do the disk I/O on the bitmap
 * file a page at a time. There's a superblock at the start of the file.
 */
672
/* calculate the index of the page that contains this bit */
673
static inline unsigned long file_page_index(struct bitmap *bitmap, unsigned long chunk)
674
{
675 676 677
	if (!bitmap->mddev->bitmap_info.external)
		chunk += sizeof(bitmap_super_t) << 3;
	return chunk >> PAGE_BIT_SHIFT;
678 679 680
}

/* calculate the (bit) offset of this bit within a page */
681
static inline unsigned long file_page_offset(struct bitmap *bitmap, unsigned long chunk)
682
{
683 684 685
	if (!bitmap->mddev->bitmap_info.external)
		chunk += sizeof(bitmap_super_t) << 3;
	return chunk & (PAGE_BITS - 1);
686 687 688 689 690 691 692 693 694 695 696 697
}

/*
 * return a pointer to the page in the filemap that contains the given bit
 *
 * this lookup is complicated by the fact that the bitmap sb might be exactly
 * 1 page (e.g., x86) or less than 1 page -- so the bitmap might start on page
 * 0 or page 1
 */
static inline struct page *filemap_get_page(struct bitmap *bitmap,
					unsigned long chunk)
{
698 699
	if (bitmap->filemap == NULL)
		return NULL;
700 701
	if (file_page_index(bitmap, chunk) >= bitmap->file_pages)
		return NULL;
702 703
	return bitmap->filemap[file_page_index(bitmap, chunk)
			       - file_page_index(bitmap, 0)];
704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724
}

static void bitmap_file_unmap(struct bitmap *bitmap)
{
	struct page **map, *sb_page;
	unsigned long *attr;
	int pages;
	unsigned long flags;

	spin_lock_irqsave(&bitmap->lock, flags);
	map = bitmap->filemap;
	bitmap->filemap = NULL;
	attr = bitmap->filemap_attr;
	bitmap->filemap_attr = NULL;
	pages = bitmap->file_pages;
	bitmap->file_pages = 0;
	sb_page = bitmap->sb_page;
	bitmap->sb_page = NULL;
	spin_unlock_irqrestore(&bitmap->lock, flags);

	while (pages--)
725
		if (map[pages] != sb_page) /* 0 is sb_page, release it below */
726
			free_buffers(map[pages]);
727 728 729
	kfree(map);
	kfree(attr);

730 731
	if (sb_page)
		free_buffers(sb_page);
732 733 734 735 736 737 738 739 740 741 742 743
}

static void bitmap_file_put(struct bitmap *bitmap)
{
	struct file *file;
	unsigned long flags;

	spin_lock_irqsave(&bitmap->lock, flags);
	file = bitmap->file;
	bitmap->file = NULL;
	spin_unlock_irqrestore(&bitmap->lock, flags);

744 745 746
	if (file)
		wait_event(bitmap->write_wait,
			   atomic_read(&bitmap->pending_writes)==0);
747 748
	bitmap_file_unmap(bitmap);

749
	if (file) {
J
Josef Sipek 已提交
750
		struct inode *inode = file->f_path.dentry->d_inode;
751
		invalidate_mapping_pages(inode->i_mapping, 0, -1);
752
		fput(file);
753
	}
754 755 756 757 758 759 760 761 762 763 764
}

/*
 * bitmap_file_kick - if an error occurs while manipulating the bitmap file
 * then it is no longer reliable, so we stop using it and we mark the file
 * as failed in the superblock
 */
static void bitmap_file_kick(struct bitmap *bitmap)
{
	char *path, *ptr = NULL;

765 766
	if (bitmap_mask_state(bitmap, BITMAP_STALE, MASK_SET) == 0) {
		bitmap_update_sb(bitmap);
767

768 769 770
		if (bitmap->file) {
			path = kmalloc(PAGE_SIZE, GFP_KERNEL);
			if (path)
C
Christoph Hellwig 已提交
771 772 773
				ptr = d_path(&bitmap->file->f_path, path,
					     PAGE_SIZE);

774 775
			printk(KERN_ALERT
			      "%s: kicking failed bitmap file %s from array!\n",
C
Christoph Hellwig 已提交
776
			      bmname(bitmap), IS_ERR(ptr) ? "" : ptr);
777

778 779 780 781 782
			kfree(path);
		} else
			printk(KERN_ALERT
			       "%s: disabling internal bitmap due to errors\n",
			       bmname(bitmap));
783
	}
784 785 786 787 788 789 790

	bitmap_file_put(bitmap);

	return;
}

enum bitmap_page_attr {
791 792 793
	BITMAP_PAGE_DIRTY = 0,     /* there are set bits that need to be synced */
	BITMAP_PAGE_CLEAN = 1,     /* there are bits that might need to be cleared */
	BITMAP_PAGE_NEEDWRITE = 2, /* there are cleared bits that need to be synced */
794 795 796 797 798
};

static inline void set_page_attr(struct bitmap *bitmap, struct page *page,
				enum bitmap_page_attr attr)
{
799 800 801 802
	if (page)
		__set_bit((page->index<<2) + attr, bitmap->filemap_attr);
	else
		__set_bit(attr, &bitmap->logattrs);
803 804 805 806 807
}

static inline void clear_page_attr(struct bitmap *bitmap, struct page *page,
				enum bitmap_page_attr attr)
{
808 809 810 811
	if (page)
		__clear_bit((page->index<<2) + attr, bitmap->filemap_attr);
	else
		__clear_bit(attr, &bitmap->logattrs);
812 813
}

814 815
static inline unsigned long test_page_attr(struct bitmap *bitmap, struct page *page,
					   enum bitmap_page_attr attr)
816
{
817 818 819 820
	if (page)
		return test_bit((page->index<<2) + attr, bitmap->filemap_attr);
	else
		return test_bit(attr, &bitmap->logattrs);
821 822 823 824 825 826 827 828 829 830 831 832
}

/*
 * bitmap_file_set_bit -- called before performing a write to the md device
 * to set (and eventually sync) a particular bit in the bitmap file
 *
 * we set the bit immediately, then we record the page number so that
 * when an unplug occurs, we can flush the dirty pages out to disk
 */
static void bitmap_file_set_bit(struct bitmap *bitmap, sector_t block)
{
	unsigned long bit;
833
	struct page *page = NULL;
834 835 836
	void *kaddr;
	unsigned long chunk = block >> CHUNK_BLOCK_SHIFT(bitmap);

837 838 839 840 841
	if (!bitmap->filemap) {
		struct dm_dirty_log *log = bitmap->mddev->bitmap_info.log;
		if (log)
			log->type->mark_region(log, chunk);
	} else {
842

843 844 845 846
		page = filemap_get_page(bitmap, chunk);
		if (!page)
			return;
		bit = file_page_offset(bitmap, chunk);
847

848 849 850 851 852 853 854 855 856
		/* set the bit */
		kaddr = kmap_atomic(page, KM_USER0);
		if (bitmap->flags & BITMAP_HOSTENDIAN)
			set_bit(bit, kaddr);
		else
			ext2_set_bit(bit, kaddr);
		kunmap_atomic(kaddr, KM_USER0);
		PRINTK("set file bit %lu page %lu\n", bit, page->index);
	}
857 858 859 860 861 862 863
	/* record page number so it gets flushed to disk when unplug occurs */
	set_page_attr(bitmap, page, BITMAP_PAGE_DIRTY);
}

/* this gets called when the md device is ready to unplug its underlying
 * (slave) device queues -- before we let any writes go down, we need to
 * sync the dirty pages of the bitmap file to disk */
864
void bitmap_unplug(struct bitmap *bitmap)
865
{
866 867
	unsigned long i, flags;
	int dirty, need_write;
868 869 870 871
	struct page *page;
	int wait = 0;

	if (!bitmap)
872
		return;
873 874 875 876 877 878 879 880 881 882
	if (!bitmap->filemap) {
		/* Must be using a dirty_log */
		struct dm_dirty_log *log = bitmap->mddev->bitmap_info.log;
		dirty = test_and_clear_bit(BITMAP_PAGE_DIRTY, &bitmap->logattrs);
		need_write = test_and_clear_bit(BITMAP_PAGE_NEEDWRITE, &bitmap->logattrs);
		if (dirty || need_write)
			if (log->type->flush(log))
				bitmap->flags |= BITMAP_WRITE_ERROR;
		goto out;
	}
883 884 885 886 887

	/* look at each page to see if there are any set bits that need to be
	 * flushed out to disk */
	for (i = 0; i < bitmap->file_pages; i++) {
		spin_lock_irqsave(&bitmap->lock, flags);
888
		if (!bitmap->filemap) {
889
			spin_unlock_irqrestore(&bitmap->lock, flags);
890
			return;
891 892
		}
		page = bitmap->filemap[i];
893 894
		dirty = test_page_attr(bitmap, page, BITMAP_PAGE_DIRTY);
		need_write = test_page_attr(bitmap, page, BITMAP_PAGE_NEEDWRITE);
895 896
		clear_page_attr(bitmap, page, BITMAP_PAGE_DIRTY);
		clear_page_attr(bitmap, page, BITMAP_PAGE_NEEDWRITE);
897
		if (dirty)
898 899 900
			wait = 1;
		spin_unlock_irqrestore(&bitmap->lock, flags);

901
		if (dirty || need_write)
902
			write_page(bitmap, page, 0);
903 904
	}
	if (wait) { /* if any writes were performed, we need to wait on them */
905
		if (bitmap->file)
906 907
			wait_event(bitmap->write_wait,
				   atomic_read(&bitmap->pending_writes)==0);
908
		else
909
			md_super_wait(bitmap->mddev);
910
	}
911
out:
912 913
	if (bitmap->flags & BITMAP_WRITE_ERROR)
		bitmap_file_kick(bitmap);
914
}
915
EXPORT_SYMBOL(bitmap_unplug);
916

917
static void bitmap_set_memory_bits(struct bitmap *bitmap, sector_t offset, int needed);
918 919 920 921 922 923 924
/* * bitmap_init_from_disk -- called at bitmap_create time to initialize
 * the in-memory bitmap from the on-disk bitmap -- also, sets up the
 * memory mapping of the bitmap file
 * Special cases:
 *   if there's no bitmap file, or if the bitmap file had been
 *   previously kicked from the array, we mark all the bits as
 *   1's in order to cause a full resync.
925 926 927
 *
 * We ignore all bits for sectors that end earlier than 'start'.
 * This is used when reading an out-of-date bitmap...
928
 */
929
static int bitmap_init_from_disk(struct bitmap *bitmap, sector_t start)
930 931 932 933 934
{
	unsigned long i, chunks, index, oldindex, bit;
	struct page *page = NULL, *oldpage = NULL;
	unsigned long num_pages, bit_cnt = 0;
	struct file *file;
935
	unsigned long bytes, offset;
936 937
	int outofdate;
	int ret = -ENOSPC;
938
	void *paddr;
939 940 941 942

	chunks = bitmap->chunks;
	file = bitmap->file;

943
	BUG_ON(!file && !bitmap->mddev->bitmap_info.offset);
944

945
#ifdef INJECT_FAULTS_3
946 947 948 949 950 951 952 953
	outofdate = 1;
#else
	outofdate = bitmap->flags & BITMAP_STALE;
#endif
	if (outofdate)
		printk(KERN_INFO "%s: bitmap file is out of date, doing full "
			"recovery\n", bmname(bitmap));

954
	bytes = DIV_ROUND_UP(bitmap->chunks, 8);
955 956
	if (!bitmap->mddev->bitmap_info.external)
		bytes += sizeof(bitmap_super_t);
957

958
	num_pages = DIV_ROUND_UP(bytes, PAGE_SIZE);
959

960
	if (file && i_size_read(file->f_mapping->host) < bytes) {
961 962 963
		printk(KERN_INFO "%s: bitmap file too short %lu < %lu\n",
			bmname(bitmap),
			(unsigned long) i_size_read(file->f_mapping->host),
964
			bytes);
965
		goto err;
966
	}
967 968 969

	ret = -ENOMEM;

970
	bitmap->filemap = kmalloc(sizeof(struct page *) * num_pages, GFP_KERNEL);
971
	if (!bitmap->filemap)
972
		goto err;
973

974 975
	/* We need 4 bits per page, rounded up to a multiple of sizeof(unsigned long) */
	bitmap->filemap_attr = kzalloc(
976
		roundup(DIV_ROUND_UP(num_pages*4, 8), sizeof(unsigned long)),
977
		GFP_KERNEL);
978
	if (!bitmap->filemap_attr)
979
		goto err;
980 981 982 983

	oldindex = ~0L;

	for (i = 0; i < chunks; i++) {
984
		int b;
985 986
		index = file_page_index(bitmap, i);
		bit = file_page_offset(bitmap, i);
987
		if (index != oldindex) { /* this is a new page, read it in */
988
			int count;
989
			/* unmap the old page, we're done with it */
990
			if (index == num_pages-1)
991
				count = bytes - index * PAGE_SIZE;
992 993
			else
				count = PAGE_SIZE;
994
			if (index == 0 && bitmap->sb_page) {
995 996 997 998 999 1000 1001
				/*
				 * if we're here then the superblock page
				 * contains some bits (PAGE_SIZE != sizeof sb)
				 * we've already read it in, so just use it
				 */
				page = bitmap->sb_page;
				offset = sizeof(bitmap_super_t);
N
NeilBrown 已提交
1002
				if (!file)
1003 1004 1005 1006 1007
					page = read_sb_page(
						bitmap->mddev,
						bitmap->mddev->bitmap_info.offset,
						page,
						index, count);
1008
			} else if (file) {
1009
				page = read_page(file, index, bitmap, count);
1010 1011
				offset = 0;
			} else {
1012 1013
				page = read_sb_page(bitmap->mddev,
						    bitmap->mddev->bitmap_info.offset,
1014 1015
						    NULL,
						    index, count);
1016 1017
				offset = 0;
			}
1018 1019
			if (IS_ERR(page)) { /* read error */
				ret = PTR_ERR(page);
1020
				goto err;
1021 1022
			}

1023 1024 1025
			oldindex = index;
			oldpage = page;

1026 1027 1028
			bitmap->filemap[bitmap->file_pages++] = page;
			bitmap->last_page_size = count;

1029 1030 1031
			if (outofdate) {
				/*
				 * if bitmap is out of date, dirty the
1032
				 * whole page and write it out
1033
				 */
1034 1035
				paddr = kmap_atomic(page, KM_USER0);
				memset(paddr + offset, 0xff,
1036
				       PAGE_SIZE - offset);
1037
				kunmap_atomic(paddr, KM_USER0);
1038 1039 1040
				write_page(bitmap, page, 1);

				ret = -EIO;
1041
				if (bitmap->flags & BITMAP_WRITE_ERROR)
1042
					goto err;
1043 1044
			}
		}
1045
		paddr = kmap_atomic(page, KM_USER0);
1046
		if (bitmap->flags & BITMAP_HOSTENDIAN)
1047
			b = test_bit(bit, paddr);
1048
		else
1049 1050
			b = ext2_test_bit(bit, paddr);
		kunmap_atomic(paddr, KM_USER0);
1051
		if (b) {
1052
			/* if the disk bit is set, set the memory bit */
1053 1054 1055 1056 1057
			int needed = ((sector_t)(i+1) << (CHUNK_BLOCK_SHIFT(bitmap))
				      >= start);
			bitmap_set_memory_bits(bitmap,
					       (sector_t)i << CHUNK_BLOCK_SHIFT(bitmap),
					       needed);
1058
			bit_cnt++;
1059
			set_page_attr(bitmap, page, BITMAP_PAGE_CLEAN);
1060 1061 1062
		}
	}

1063
	/* everything went OK */
1064 1065 1066 1067 1068 1069 1070 1071 1072
	ret = 0;
	bitmap_mask_state(bitmap, BITMAP_STALE, MASK_UNSET);

	if (bit_cnt) { /* Kick recovery if any bits were set */
		set_bit(MD_RECOVERY_NEEDED, &bitmap->mddev->recovery);
		md_wakeup_thread(bitmap->mddev->thread);
	}

	printk(KERN_INFO "%s: bitmap initialized from disk: "
1073 1074 1075 1076
		"read %lu/%lu pages, set %lu bits\n",
		bmname(bitmap), bitmap->file_pages, num_pages, bit_cnt);

	return 0;
1077

1078 1079 1080
 err:
	printk(KERN_INFO "%s: bitmap initialisation failed: %d\n",
	       bmname(bitmap), ret);
1081 1082 1083
	return ret;
}

1084 1085 1086 1087 1088
void bitmap_write_all(struct bitmap *bitmap)
{
	/* We don't actually write all bitmap blocks here,
	 * just flag them as needing to be written
	 */
1089
	int i;
1090

1091
	for (i = 0; i < bitmap->file_pages; i++)
1092 1093
		set_page_attr(bitmap, bitmap->filemap[i],
			      BITMAP_PAGE_NEEDWRITE);
1094 1095
}

1096 1097 1098 1099 1100 1101 1102 1103
static void bitmap_count_page(struct bitmap *bitmap, sector_t offset, int inc)
{
	sector_t chunk = offset >> CHUNK_BLOCK_SHIFT(bitmap);
	unsigned long page = chunk >> PAGE_COUNTER_SHIFT;
	bitmap->bp[page].count += inc;
	bitmap_checkfree(bitmap, page);
}
static bitmap_counter_t *bitmap_get_counter(struct bitmap *bitmap,
N
NeilBrown 已提交
1104
					    sector_t offset, sector_t *blocks,
1105 1106 1107 1108 1109 1110 1111
					    int create);

/*
 * bitmap daemon -- periodically wakes up to clean bits and flush pages
 *			out to disk
 */

1112
void bitmap_daemon_work(mddev_t *mddev)
1113
{
1114
	struct bitmap *bitmap;
1115
	unsigned long j;
1116 1117
	unsigned long flags;
	struct page *page = NULL, *lastpage = NULL;
N
NeilBrown 已提交
1118
	sector_t blocks;
1119
	void *paddr;
1120
	struct dm_dirty_log *log = mddev->bitmap_info.log;
1121

1122 1123 1124
	/* Use a mutex to guard daemon_work against
	 * bitmap_destroy.
	 */
1125
	mutex_lock(&mddev->bitmap_info.mutex);
1126 1127
	bitmap = mddev->bitmap;
	if (bitmap == NULL) {
1128
		mutex_unlock(&mddev->bitmap_info.mutex);
1129
		return;
1130
	}
1131
	if (time_before(jiffies, bitmap->daemon_lastrun
1132
			+ bitmap->mddev->bitmap_info.daemon_sleep))
1133 1134
		goto done;

1135
	bitmap->daemon_lastrun = jiffies;
1136 1137
	if (bitmap->allclean) {
		bitmap->mddev->thread->timeout = MAX_SCHEDULE_TIMEOUT;
1138
		goto done;
1139 1140
	}
	bitmap->allclean = 1;
1141

1142
	spin_lock_irqsave(&bitmap->lock, flags);
1143 1144
	for (j = 0; j < bitmap->chunks; j++) {
		bitmap_counter_t *bmc;
1145 1146 1147 1148 1149 1150
		if (!bitmap->filemap) {
			if (!log)
				/* error or shutdown */
				break;
		} else
			page = filemap_get_page(bitmap, j);
1151 1152

		if (page != lastpage) {
1153
			/* skip this page unless it's marked as needing cleaning */
1154 1155 1156
			if (!test_page_attr(bitmap, page, BITMAP_PAGE_CLEAN)) {
				int need_write = test_page_attr(bitmap, page,
								BITMAP_PAGE_NEEDWRITE);
1157
				if (need_write)
1158
					clear_page_attr(bitmap, page, BITMAP_PAGE_NEEDWRITE);
1159

1160
				spin_unlock_irqrestore(&bitmap->lock, flags);
1161
				if (need_write) {
1162
					write_page(bitmap, page, 0);
1163 1164
					bitmap->allclean = 0;
				}
1165 1166
				spin_lock_irqsave(&bitmap->lock, flags);
				j |= (PAGE_BITS - 1);
1167 1168 1169
				continue;
			}

1170 1171
			/* grab the new page, sync and release the old */
			if (lastpage != NULL) {
1172
				if (test_page_attr(bitmap, lastpage, BITMAP_PAGE_NEEDWRITE)) {
1173 1174
					clear_page_attr(bitmap, lastpage, BITMAP_PAGE_NEEDWRITE);
					spin_unlock_irqrestore(&bitmap->lock, flags);
1175
					write_page(bitmap, lastpage, 0);
1176 1177 1178 1179 1180 1181 1182
				} else {
					set_page_attr(bitmap, lastpage, BITMAP_PAGE_NEEDWRITE);
					spin_unlock_irqrestore(&bitmap->lock, flags);
				}
			} else
				spin_unlock_irqrestore(&bitmap->lock, flags);
			lastpage = page;
1183 1184 1185 1186

			/* We are possibly going to clear some bits, so make
			 * sure that events_cleared is up-to-date.
			 */
1187 1188
			if (bitmap->need_sync &&
			    bitmap->mddev->bitmap_info.external == 0) {
1189 1190 1191 1192 1193 1194 1195 1196
				bitmap_super_t *sb;
				bitmap->need_sync = 0;
				sb = kmap_atomic(bitmap->sb_page, KM_USER0);
				sb->events_cleared =
					cpu_to_le64(bitmap->events_cleared);
				kunmap_atomic(sb, KM_USER0);
				write_page(bitmap, bitmap->sb_page, 1);
			}
1197
			spin_lock_irqsave(&bitmap->lock, flags);
1198 1199
			if (!bitmap->need_sync)
				clear_page_attr(bitmap, page, BITMAP_PAGE_CLEAN);
1200
		}
1201 1202 1203
		bmc = bitmap_get_counter(bitmap,
					 (sector_t)j << CHUNK_BLOCK_SHIFT(bitmap),
					 &blocks, 0);
1204
		if (bmc) {
1205 1206 1207
			if (*bmc)
				bitmap->allclean = 0;

1208
			if (*bmc == 2) {
1209
				*bmc = 1; /* maybe clear the bit next time */
1210
				set_page_attr(bitmap, page, BITMAP_PAGE_CLEAN);
1211
			} else if (*bmc == 1 && !bitmap->need_sync) {
1212 1213
				/* we can clear the bit */
				*bmc = 0;
1214 1215
				bitmap_count_page(bitmap,
						  (sector_t)j << CHUNK_BLOCK_SHIFT(bitmap),
1216 1217 1218
						  -1);

				/* clear the bit */
1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229
				if (page) {
					paddr = kmap_atomic(page, KM_USER0);
					if (bitmap->flags & BITMAP_HOSTENDIAN)
						clear_bit(file_page_offset(bitmap, j),
							  paddr);
					else
						ext2_clear_bit(file_page_offset(bitmap, j),
							       paddr);
					kunmap_atomic(paddr, KM_USER0);
				} else
					log->type->clear_region(log, j);
1230
			}
1231 1232
		} else
			j |= PAGE_COUNTER_MASK;
1233
	}
1234
	spin_unlock_irqrestore(&bitmap->lock, flags);
1235 1236

	/* now sync the final page */
1237
	if (lastpage != NULL || log != NULL) {
1238
		spin_lock_irqsave(&bitmap->lock, flags);
1239
		if (test_page_attr(bitmap, lastpage, BITMAP_PAGE_NEEDWRITE)) {
1240 1241
			clear_page_attr(bitmap, lastpage, BITMAP_PAGE_NEEDWRITE);
			spin_unlock_irqrestore(&bitmap->lock, flags);
1242 1243 1244 1245 1246
			if (lastpage)
				write_page(bitmap, lastpage, 0);
			else
				if (log->type->flush(log))
					bitmap->flags |= BITMAP_WRITE_ERROR;
1247 1248 1249 1250 1251 1252
		} else {
			set_page_attr(bitmap, lastpage, BITMAP_PAGE_NEEDWRITE);
			spin_unlock_irqrestore(&bitmap->lock, flags);
		}
	}

1253
 done:
1254
	if (bitmap->allclean == 0)
1255
		bitmap->mddev->thread->timeout =
1256
			bitmap->mddev->bitmap_info.daemon_sleep;
1257
	mutex_unlock(&mddev->bitmap_info.mutex);
1258 1259 1260
}

static bitmap_counter_t *bitmap_get_counter(struct bitmap *bitmap,
N
NeilBrown 已提交
1261
					    sector_t offset, sector_t *blocks,
1262
					    int create)
1263 1264
__releases(bitmap->lock)
__acquires(bitmap->lock)
1265 1266 1267 1268 1269 1270 1271 1272 1273
{
	/* If 'create', we might release the lock and reclaim it.
	 * The lock must have been taken with interrupts enabled.
	 * If !create, we don't release the lock.
	 */
	sector_t chunk = offset >> CHUNK_BLOCK_SHIFT(bitmap);
	unsigned long page = chunk >> PAGE_COUNTER_SHIFT;
	unsigned long pageoff = (chunk & PAGE_COUNTER_MASK) << COUNTER_BYTE_SHIFT;
	sector_t csize;
1274
	int err;
1275

1276 1277 1278 1279 1280 1281 1282
	err = bitmap_checkpage(bitmap, page, create);

	if (bitmap->bp[page].hijacked ||
	    bitmap->bp[page].map == NULL)
		csize = ((sector_t)1) << (CHUNK_BLOCK_SHIFT(bitmap) +
					  PAGE_COUNTER_SHIFT - 1);
	else
1283
		csize = ((sector_t)1) << (CHUNK_BLOCK_SHIFT(bitmap));
1284 1285 1286
	*blocks = csize - (offset & (csize - 1));

	if (err < 0)
1287
		return NULL;
1288

1289 1290 1291 1292 1293 1294 1295 1296
	/* now locked ... */

	if (bitmap->bp[page].hijacked) { /* hijacked pointer */
		/* should we use the first or second counter field
		 * of the hijacked pointer? */
		int hi = (pageoff > PAGE_COUNTER_MASK);
		return  &((bitmap_counter_t *)
			  &bitmap->bp[page].map)[hi];
1297
	} else /* page is allocated */
1298 1299 1300 1301
		return (bitmap_counter_t *)
			&(bitmap->bp[page].map[pageoff]);
}

1302
int bitmap_startwrite(struct bitmap *bitmap, sector_t offset, unsigned long sectors, int behind)
1303
{
1304 1305
	if (!bitmap)
		return 0;
1306 1307

	if (behind) {
1308
		int bw;
1309
		atomic_inc(&bitmap->behind_writes);
1310 1311 1312 1313
		bw = atomic_read(&bitmap->behind_writes);
		if (bw > bitmap->behind_writes_used)
			bitmap->behind_writes_used = bw;

1314
		PRINTK(KERN_DEBUG "inc write-behind count %d/%d\n",
1315
		       bw, bitmap->max_write_behind);
1316 1317
	}

1318
	while (sectors) {
N
NeilBrown 已提交
1319
		sector_t blocks;
1320 1321 1322 1323 1324 1325 1326 1327 1328
		bitmap_counter_t *bmc;

		spin_lock_irq(&bitmap->lock);
		bmc = bitmap_get_counter(bitmap, offset, &blocks, 1);
		if (!bmc) {
			spin_unlock_irq(&bitmap->lock);
			return 0;
		}

1329 1330 1331 1332 1333 1334 1335 1336 1337
		if (unlikely((*bmc & COUNTER_MAX) == COUNTER_MAX)) {
			DEFINE_WAIT(__wait);
			/* note that it is safe to do the prepare_to_wait
			 * after the test as long as we do it before dropping
			 * the spinlock.
			 */
			prepare_to_wait(&bitmap->overflow_wait, &__wait,
					TASK_UNINTERRUPTIBLE);
			spin_unlock_irq(&bitmap->lock);
N
NeilBrown 已提交
1338
			md_unplug(bitmap->mddev);
1339 1340 1341 1342 1343
			schedule();
			finish_wait(&bitmap->overflow_wait, &__wait);
			continue;
		}

1344
		switch (*bmc) {
1345 1346
		case 0:
			bitmap_file_set_bit(bitmap, offset);
1347
			bitmap_count_page(bitmap, offset, 1);
1348 1349 1350 1351
			/* fall through */
		case 1:
			*bmc = 2;
		}
1352

1353 1354 1355 1356 1357 1358 1359
		(*bmc)++;

		spin_unlock_irq(&bitmap->lock);

		offset += blocks;
		if (sectors > blocks)
			sectors -= blocks;
1360 1361
		else
			sectors = 0;
1362
	}
1363
	bitmap->allclean = 0;
1364 1365
	return 0;
}
1366
EXPORT_SYMBOL(bitmap_startwrite);
1367 1368

void bitmap_endwrite(struct bitmap *bitmap, sector_t offset, unsigned long sectors,
1369
		     int success, int behind)
1370
{
1371 1372
	if (!bitmap)
		return;
1373
	if (behind) {
1374 1375
		if (atomic_dec_and_test(&bitmap->behind_writes))
			wake_up(&bitmap->behind_wait);
1376 1377 1378
		PRINTK(KERN_DEBUG "dec write-behind count %d/%d\n",
		  atomic_read(&bitmap->behind_writes), bitmap->max_write_behind);
	}
1379 1380 1381
	if (bitmap->mddev->degraded)
		/* Never clear bits or update events_cleared when degraded */
		success = 0;
1382

1383
	while (sectors) {
N
NeilBrown 已提交
1384
		sector_t blocks;
1385 1386 1387 1388 1389 1390 1391 1392 1393 1394
		unsigned long flags;
		bitmap_counter_t *bmc;

		spin_lock_irqsave(&bitmap->lock, flags);
		bmc = bitmap_get_counter(bitmap, offset, &blocks, 0);
		if (!bmc) {
			spin_unlock_irqrestore(&bitmap->lock, flags);
			return;
		}

1395 1396 1397 1398
		if (success &&
		    bitmap->events_cleared < bitmap->mddev->events) {
			bitmap->events_cleared = bitmap->mddev->events;
			bitmap->need_sync = 1;
1399
			sysfs_notify_dirent_safe(bitmap->sysfs_can_clear);
1400 1401
		}

1402 1403 1404
		if (!success && ! (*bmc & NEEDED_MASK))
			*bmc |= NEEDED_MASK;

1405 1406 1407
		if ((*bmc & COUNTER_MAX) == COUNTER_MAX)
			wake_up(&bitmap->overflow_wait);

1408
		(*bmc)--;
1409
		if (*bmc <= 2)
1410
			set_page_attr(bitmap,
1411 1412 1413
				      filemap_get_page(
					      bitmap,
					      offset >> CHUNK_BLOCK_SHIFT(bitmap)),
1414
				      BITMAP_PAGE_CLEAN);
1415

1416 1417 1418 1419
		spin_unlock_irqrestore(&bitmap->lock, flags);
		offset += blocks;
		if (sectors > blocks)
			sectors -= blocks;
1420 1421
		else
			sectors = 0;
1422 1423
	}
}
1424
EXPORT_SYMBOL(bitmap_endwrite);
1425

N
NeilBrown 已提交
1426
static int __bitmap_start_sync(struct bitmap *bitmap, sector_t offset, sector_t *blocks,
1427
			       int degraded)
1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443
{
	bitmap_counter_t *bmc;
	int rv;
	if (bitmap == NULL) {/* FIXME or bitmap set as 'failed' */
		*blocks = 1024;
		return 1; /* always resync if no bitmap */
	}
	spin_lock_irq(&bitmap->lock);
	bmc = bitmap_get_counter(bitmap, offset, blocks, 0);
	rv = 0;
	if (bmc) {
		/* locked */
		if (RESYNC(*bmc))
			rv = 1;
		else if (NEEDED(*bmc)) {
			rv = 1;
1444 1445 1446 1447
			if (!degraded) { /* don't set/clear bits if degraded */
				*bmc |= RESYNC_MASK;
				*bmc &= ~NEEDED_MASK;
			}
1448 1449 1450
		}
	}
	spin_unlock_irq(&bitmap->lock);
1451
	bitmap->allclean = 0;
1452 1453 1454
	return rv;
}

N
NeilBrown 已提交
1455
int bitmap_start_sync(struct bitmap *bitmap, sector_t offset, sector_t *blocks,
1456 1457 1458 1459 1460 1461 1462 1463 1464 1465
		      int degraded)
{
	/* bitmap_start_sync must always report on multiples of whole
	 * pages, otherwise resync (which is very PAGE_SIZE based) will
	 * get confused.
	 * So call __bitmap_start_sync repeatedly (if needed) until
	 * At least PAGE_SIZE>>9 blocks are covered.
	 * Return the 'or' of the result.
	 */
	int rv = 0;
N
NeilBrown 已提交
1466
	sector_t blocks1;
1467 1468 1469 1470 1471 1472 1473 1474 1475 1476

	*blocks = 0;
	while (*blocks < (PAGE_SIZE>>9)) {
		rv |= __bitmap_start_sync(bitmap, offset,
					  &blocks1, degraded);
		offset += blocks1;
		*blocks += blocks1;
	}
	return rv;
}
1477
EXPORT_SYMBOL(bitmap_start_sync);
1478

N
NeilBrown 已提交
1479
void bitmap_end_sync(struct bitmap *bitmap, sector_t offset, sector_t *blocks, int aborted)
1480 1481 1482
{
	bitmap_counter_t *bmc;
	unsigned long flags;
1483 1484

	if (bitmap == NULL) {
1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498
		*blocks = 1024;
		return;
	}
	spin_lock_irqsave(&bitmap->lock, flags);
	bmc = bitmap_get_counter(bitmap, offset, blocks, 0);
	if (bmc == NULL)
		goto unlock;
	/* locked */
	if (RESYNC(*bmc)) {
		*bmc &= ~RESYNC_MASK;

		if (!NEEDED(*bmc) && aborted)
			*bmc |= NEEDED_MASK;
		else {
1499
			if (*bmc <= 2)
1500 1501 1502 1503 1504 1505 1506
				set_page_attr(bitmap,
					      filemap_get_page(bitmap, offset >> CHUNK_BLOCK_SHIFT(bitmap)),
					      BITMAP_PAGE_CLEAN);
		}
	}
 unlock:
	spin_unlock_irqrestore(&bitmap->lock, flags);
1507
	bitmap->allclean = 0;
1508
}
1509
EXPORT_SYMBOL(bitmap_end_sync);
1510 1511 1512 1513 1514 1515 1516 1517

void bitmap_close_sync(struct bitmap *bitmap)
{
	/* Sync has finished, and any bitmap chunks that weren't synced
	 * properly have been aborted.  It remains to us to clear the
	 * RESYNC bit wherever it is still on
	 */
	sector_t sector = 0;
N
NeilBrown 已提交
1518
	sector_t blocks;
N
NeilBrown 已提交
1519 1520
	if (!bitmap)
		return;
1521 1522
	while (sector < bitmap->mddev->resync_max_sectors) {
		bitmap_end_sync(bitmap, sector, &blocks, 0);
N
NeilBrown 已提交
1523 1524 1525
		sector += blocks;
	}
}
1526
EXPORT_SYMBOL(bitmap_close_sync);
N
NeilBrown 已提交
1527 1528 1529 1530

void bitmap_cond_end_sync(struct bitmap *bitmap, sector_t sector)
{
	sector_t s = 0;
N
NeilBrown 已提交
1531
	sector_t blocks;
N
NeilBrown 已提交
1532 1533 1534 1535 1536 1537 1538 1539

	if (!bitmap)
		return;
	if (sector == 0) {
		bitmap->last_end_sync = jiffies;
		return;
	}
	if (time_before(jiffies, (bitmap->last_end_sync
1540
				  + bitmap->mddev->bitmap_info.daemon_sleep)))
N
NeilBrown 已提交
1541 1542 1543 1544
		return;
	wait_event(bitmap->mddev->recovery_wait,
		   atomic_read(&bitmap->mddev->recovery_active) == 0);

1545
	bitmap->mddev->curr_resync_completed = bitmap->mddev->curr_resync;
1546
	set_bit(MD_CHANGE_CLEAN, &bitmap->mddev->flags);
N
NeilBrown 已提交
1547 1548 1549 1550 1551
	sector &= ~((1ULL << CHUNK_BLOCK_SHIFT(bitmap)) - 1);
	s = 0;
	while (s < sector && s < bitmap->mddev->resync_max_sectors) {
		bitmap_end_sync(bitmap, s, &blocks, 0);
		s += blocks;
1552
	}
N
NeilBrown 已提交
1553
	bitmap->last_end_sync = jiffies;
1554
	sysfs_notify(&bitmap->mddev->kobj, NULL, "sync_completed");
1555
}
1556
EXPORT_SYMBOL(bitmap_cond_end_sync);
1557

1558
static void bitmap_set_memory_bits(struct bitmap *bitmap, sector_t offset, int needed)
1559 1560
{
	/* For each chunk covered by any of these sectors, set the
1561
	 * counter to 1 and set resync_needed.  They should all
1562 1563
	 * be 0 at this point
	 */
1564

N
NeilBrown 已提交
1565
	sector_t secs;
1566 1567 1568 1569
	bitmap_counter_t *bmc;
	spin_lock_irq(&bitmap->lock);
	bmc = bitmap_get_counter(bitmap, offset, &secs, 1);
	if (!bmc) {
1570
		spin_unlock_irq(&bitmap->lock);
1571
		return;
1572
	}
1573
	if (!*bmc) {
1574
		struct page *page;
1575
		*bmc = 1 | (needed ? NEEDED_MASK : 0);
1576 1577 1578 1579 1580
		bitmap_count_page(bitmap, offset, 1);
		page = filemap_get_page(bitmap, offset >> CHUNK_BLOCK_SHIFT(bitmap));
		set_page_attr(bitmap, page, BITMAP_PAGE_CLEAN);
	}
	spin_unlock_irq(&bitmap->lock);
1581
	bitmap->allclean = 0;
1582 1583
}

1584 1585 1586 1587 1588 1589
/* dirty the memory and file bits for bitmap chunks "s" to "e" */
void bitmap_dirty_bits(struct bitmap *bitmap, unsigned long s, unsigned long e)
{
	unsigned long chunk;

	for (chunk = s; chunk <= e; chunk++) {
1590
		sector_t sec = (sector_t)chunk << CHUNK_BLOCK_SHIFT(bitmap);
1591 1592
		bitmap_set_memory_bits(bitmap, sec, 1);
		bitmap_file_set_bit(bitmap, sec);
1593 1594 1595 1596 1597 1598
		if (sec < bitmap->mddev->recovery_cp)
			/* We are asserting that the array is dirty,
			 * so move the recovery_cp address back so
			 * that it is obvious that it is dirty
			 */
			bitmap->mddev->recovery_cp = sec;
1599 1600 1601
	}
}

1602 1603 1604 1605 1606 1607
/*
 * flush out any pending updates
 */
void bitmap_flush(mddev_t *mddev)
{
	struct bitmap *bitmap = mddev->bitmap;
1608
	long sleep;
1609 1610 1611 1612 1613 1614 1615

	if (!bitmap) /* there was no bitmap */
		return;

	/* run the daemon_work three time to ensure everything is flushed
	 * that can be
	 */
1616
	sleep = mddev->bitmap_info.daemon_sleep * 2;
1617
	bitmap->daemon_lastrun -= sleep;
1618
	bitmap_daemon_work(mddev);
1619
	bitmap->daemon_lastrun -= sleep;
1620
	bitmap_daemon_work(mddev);
1621
	bitmap->daemon_lastrun -= sleep;
1622
	bitmap_daemon_work(mddev);
1623 1624 1625
	bitmap_update_sb(bitmap);
}

1626 1627 1628
/*
 * free memory that was allocated
 */
1629
static void bitmap_free(struct bitmap *bitmap)
1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651
{
	unsigned long k, pages;
	struct bitmap_page *bp;

	if (!bitmap) /* there was no bitmap */
		return;

	/* release the bitmap file and kill the daemon */
	bitmap_file_put(bitmap);

	bp = bitmap->bp;
	pages = bitmap->pages;

	/* free all allocated memory */

	if (bp) /* deallocate the page memory */
		for (k = 0; k < pages; k++)
			if (bp[k].map && !bp[k].hijacked)
				kfree(bp[k].map);
	kfree(bp);
	kfree(bitmap);
}
1652

1653 1654 1655 1656 1657 1658 1659
void bitmap_destroy(mddev_t *mddev)
{
	struct bitmap *bitmap = mddev->bitmap;

	if (!bitmap) /* there was no bitmap */
		return;

1660
	mutex_lock(&mddev->bitmap_info.mutex);
1661
	mddev->bitmap = NULL; /* disconnect from the md device */
1662
	mutex_unlock(&mddev->bitmap_info.mutex);
1663 1664
	if (mddev->thread)
		mddev->thread->timeout = MAX_SCHEDULE_TIMEOUT;
1665

1666 1667 1668
	if (bitmap->sysfs_can_clear)
		sysfs_put(bitmap->sysfs_can_clear);

1669 1670
	bitmap_free(bitmap);
}
1671 1672 1673 1674 1675 1676 1677 1678

/*
 * initialize the bitmap structure
 * if this returns an error, bitmap_destroy must be called to do clean up
 */
int bitmap_create(mddev_t *mddev)
{
	struct bitmap *bitmap;
1679
	sector_t blocks = mddev->resync_max_sectors;
1680 1681
	unsigned long chunks;
	unsigned long pages;
1682
	struct file *file = mddev->bitmap_info.file;
1683
	int err;
1684
	struct sysfs_dirent *bm = NULL;
1685

A
Alexey Dobriyan 已提交
1686
	BUILD_BUG_ON(sizeof(bitmap_super_t) != 256);
1687

1688 1689 1690
	if (!file
	    && !mddev->bitmap_info.offset
	    && !mddev->bitmap_info.log) /* bitmap disabled, nothing to do */
1691 1692
		return 0;

1693
	BUG_ON(file && mddev->bitmap_info.offset);
1694
	BUG_ON(mddev->bitmap_info.offset && mddev->bitmap_info.log);
1695

1696
	bitmap = kzalloc(sizeof(*bitmap), GFP_KERNEL);
1697 1698 1699 1700
	if (!bitmap)
		return -ENOMEM;

	spin_lock_init(&bitmap->lock);
1701 1702
	atomic_set(&bitmap->pending_writes, 0);
	init_waitqueue_head(&bitmap->write_wait);
1703
	init_waitqueue_head(&bitmap->overflow_wait);
1704
	init_waitqueue_head(&bitmap->behind_wait);
1705

1706 1707
	bitmap->mddev = mddev;

1708 1709
	if (mddev->kobj.sd)
		bm = sysfs_get_dirent(mddev->kobj.sd, NULL, "bitmap");
1710
	if (bm) {
1711
		bitmap->sysfs_can_clear = sysfs_get_dirent(bm, NULL, "can_clear");
1712 1713 1714 1715
		sysfs_put(bm);
	} else
		bitmap->sysfs_can_clear = NULL;

1716
	bitmap->file = file;
1717 1718
	if (file) {
		get_file(file);
1719 1720 1721 1722
		/* As future accesses to this file will use bmap,
		 * and bypass the page cache, we must sync the file
		 * first.
		 */
1723
		vfs_fsync(file, 1);
1724
	}
1725
	/* read superblock from bitmap file (this sets mddev->bitmap_info.chunksize) */
1726 1727 1728 1729 1730 1731 1732 1733 1734 1735
	if (!mddev->bitmap_info.external)
		err = bitmap_read_sb(bitmap);
	else {
		err = 0;
		if (mddev->bitmap_info.chunksize == 0 ||
		    mddev->bitmap_info.daemon_sleep == 0)
			/* chunksize and time_base need to be
			 * set first. */
			err = -EINVAL;
	}
1736
	if (err)
1737
		goto error;
1738

1739
	bitmap->daemon_lastrun = jiffies;
1740
	bitmap->chunkshift = ffz(~mddev->bitmap_info.chunksize);
1741 1742

	/* now that chunksize and chunkshift are set, we can use these macros */
1743
	chunks = (blocks + CHUNK_BLOCK_RATIO(bitmap) - 1) >>
1744
			CHUNK_BLOCK_SHIFT(bitmap);
1745
	pages = (chunks + PAGE_COUNTER_RATIO - 1) / PAGE_COUNTER_RATIO;
1746 1747 1748 1749 1750 1751 1752 1753 1754 1755

	BUG_ON(!pages);

	bitmap->chunks = chunks;
	bitmap->pages = pages;
	bitmap->missing_pages = pages;
	bitmap->counter_bits = COUNTER_BITS;

	bitmap->syncchunk = ~0UL;

1756
#ifdef INJECT_FATAL_FAULT_1
1757 1758
	bitmap->bp = NULL;
#else
1759
	bitmap->bp = kzalloc(pages * sizeof(*bitmap->bp), GFP_KERNEL);
1760
#endif
1761
	err = -ENOMEM;
1762
	if (!bitmap->bp)
1763
		goto error;
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
	printk(KERN_INFO "created bitmap (%lu pages) for device %s\n",
		pages, bmname(bitmap));

	mddev->bitmap = bitmap;


	return (bitmap->flags & BITMAP_WRITE_ERROR) ? -EIO : 0;

 error:
	bitmap_free(bitmap);
	return err;
}

int bitmap_load(mddev_t *mddev)
{
	int err = 0;
	sector_t sector = 0;
	struct bitmap *bitmap = mddev->bitmap;

	if (!bitmap)
		goto out;

	/* Clear out old bitmap info first:  Either there is none, or we
	 * are resuming after someone else has possibly changed things,
	 * so we should forget old cached info.
	 * All chunks should be clean, but some might need_sync.
	 */
	while (sector < mddev->resync_max_sectors) {
N
NeilBrown 已提交
1793
		sector_t blocks;
1794 1795 1796 1797 1798
		bitmap_start_sync(bitmap, sector, &blocks, 0);
		sector += blocks;
	}
	bitmap_close_sync(bitmap);

1799 1800 1801 1802 1803 1804 1805 1806
	if (mddev->bitmap_info.log) {
		unsigned long i;
		struct dm_dirty_log *log = mddev->bitmap_info.log;
		for (i = 0; i < bitmap->chunks; i++)
			if (!log->type->in_sync(log, i, 1))
				bitmap_set_memory_bits(bitmap,
						       (sector_t)i << CHUNK_BLOCK_SHIFT(bitmap),
						       1);
1807 1808 1809 1810 1811 1812 1813
	} else {
		sector_t start = 0;
		if (mddev->degraded == 0
		    || bitmap->events_cleared == mddev->events)
			/* no need to keep dirty bits to optimise a
			 * re-add of a missing device */
			start = mddev->recovery_cp;
1814

1815 1816
		err = bitmap_init_from_disk(bitmap, start);
	}
1817
	if (err)
1818
		goto out;
1819

1820
	mddev->thread->timeout = mddev->bitmap_info.daemon_sleep;
1821
	md_wakeup_thread(mddev->thread);
1822

1823 1824
	bitmap_update_sb(bitmap);

1825 1826 1827
	if (bitmap->flags & BITMAP_WRITE_ERROR)
		err = -EIO;
out:
1828
	return err;
1829
}
1830
EXPORT_SYMBOL_GPL(bitmap_load);
1831

1832 1833 1834 1835
static ssize_t
location_show(mddev_t *mddev, char *page)
{
	ssize_t len;
1836
	if (mddev->bitmap_info.file)
1837
		len = sprintf(page, "file");
1838
	else if (mddev->bitmap_info.offset)
1839
		len = sprintf(page, "%+lld", (long long)mddev->bitmap_info.offset);
1840
	else
1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891
		len = sprintf(page, "none");
	len += sprintf(page+len, "\n");
	return len;
}

static ssize_t
location_store(mddev_t *mddev, const char *buf, size_t len)
{

	if (mddev->pers) {
		if (!mddev->pers->quiesce)
			return -EBUSY;
		if (mddev->recovery || mddev->sync_thread)
			return -EBUSY;
	}

	if (mddev->bitmap || mddev->bitmap_info.file ||
	    mddev->bitmap_info.offset) {
		/* bitmap already configured.  Only option is to clear it */
		if (strncmp(buf, "none", 4) != 0)
			return -EBUSY;
		if (mddev->pers) {
			mddev->pers->quiesce(mddev, 1);
			bitmap_destroy(mddev);
			mddev->pers->quiesce(mddev, 0);
		}
		mddev->bitmap_info.offset = 0;
		if (mddev->bitmap_info.file) {
			struct file *f = mddev->bitmap_info.file;
			mddev->bitmap_info.file = NULL;
			restore_bitmap_write_access(f);
			fput(f);
		}
	} else {
		/* No bitmap, OK to set a location */
		long long offset;
		if (strncmp(buf, "none", 4) == 0)
			/* nothing to be done */;
		else if (strncmp(buf, "file:", 5) == 0) {
			/* Not supported yet */
			return -EINVAL;
		} else {
			int rv;
			if (buf[0] == '+')
				rv = strict_strtoll(buf+1, 10, &offset);
			else
				rv = strict_strtoll(buf, 10, &offset);
			if (rv)
				return rv;
			if (offset == 0)
				return -EINVAL;
1892 1893
			if (mddev->bitmap_info.external == 0 &&
			    mddev->major_version == 0 &&
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
			    offset != mddev->bitmap_info.default_offset)
				return -EINVAL;
			mddev->bitmap_info.offset = offset;
			if (mddev->pers) {
				mddev->pers->quiesce(mddev, 1);
				rv = bitmap_create(mddev);
				if (rv) {
					bitmap_destroy(mddev);
					mddev->bitmap_info.offset = 0;
				}
				mddev->pers->quiesce(mddev, 0);
				if (rv)
					return rv;
			}
		}
	}
	if (!mddev->external) {
		/* Ensure new bitmap info is stored in
		 * metadata promptly.
		 */
		set_bit(MD_CHANGE_DEVS, &mddev->flags);
		md_wakeup_thread(mddev->thread);
	}
	return len;
}

static struct md_sysfs_entry bitmap_location =
__ATTR(location, S_IRUGO|S_IWUSR, location_show, location_store);

static ssize_t
timeout_show(mddev_t *mddev, char *page)
{
	ssize_t len;
	unsigned long secs = mddev->bitmap_info.daemon_sleep / HZ;
	unsigned long jifs = mddev->bitmap_info.daemon_sleep % HZ;
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
	len = sprintf(page, "%lu", secs);
	if (jifs)
		len += sprintf(page+len, ".%03u", jiffies_to_msecs(jifs));
	len += sprintf(page+len, "\n");
	return len;
}

static ssize_t
timeout_store(mddev_t *mddev, const char *buf, size_t len)
{
	/* timeout can be set at any time */
	unsigned long timeout;
	int rv = strict_strtoul_scaled(buf, &timeout, 4);
	if (rv)
		return rv;

	/* just to make sure we don't overflow... */
	if (timeout >= LONG_MAX / HZ)
		return -EINVAL;

	timeout = timeout * HZ / 10000;

	if (timeout >= MAX_SCHEDULE_TIMEOUT)
		timeout = MAX_SCHEDULE_TIMEOUT-1;
	if (timeout < 1)
		timeout = 1;
	mddev->bitmap_info.daemon_sleep = timeout;
	if (mddev->thread) {
		/* if thread->timeout is MAX_SCHEDULE_TIMEOUT, then
		 * the bitmap is all clean and we don't need to
		 * adjust the timeout right now
		 */
		if (mddev->thread->timeout < MAX_SCHEDULE_TIMEOUT) {
			mddev->thread->timeout = timeout;
			md_wakeup_thread(mddev->thread);
		}
	}
	return len;
}

static struct md_sysfs_entry bitmap_timeout =
__ATTR(time_base, S_IRUGO|S_IWUSR, timeout_show, timeout_store);

static ssize_t
backlog_show(mddev_t *mddev, char *page)
{
	return sprintf(page, "%lu\n", mddev->bitmap_info.max_write_behind);
}

static ssize_t
backlog_store(mddev_t *mddev, const char *buf, size_t len)
{
	unsigned long backlog;
	int rv = strict_strtoul(buf, 10, &backlog);
	if (rv)
		return rv;
	if (backlog > COUNTER_MAX)
		return -EINVAL;
	mddev->bitmap_info.max_write_behind = backlog;
	return len;
}

static struct md_sysfs_entry bitmap_backlog =
__ATTR(backlog, S_IRUGO|S_IWUSR, backlog_show, backlog_store);

static ssize_t
chunksize_show(mddev_t *mddev, char *page)
{
	return sprintf(page, "%lu\n", mddev->bitmap_info.chunksize);
}

static ssize_t
chunksize_store(mddev_t *mddev, const char *buf, size_t len)
{
	/* Can only be changed when no bitmap is active */
	int rv;
	unsigned long csize;
	if (mddev->bitmap)
		return -EBUSY;
	rv = strict_strtoul(buf, 10, &csize);
	if (rv)
		return rv;
	if (csize < 512 ||
	    !is_power_of_2(csize))
		return -EINVAL;
	mddev->bitmap_info.chunksize = csize;
	return len;
}

static struct md_sysfs_entry bitmap_chunksize =
__ATTR(chunksize, S_IRUGO|S_IWUSR, chunksize_show, chunksize_store);

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
static ssize_t metadata_show(mddev_t *mddev, char *page)
{
	return sprintf(page, "%s\n", (mddev->bitmap_info.external
				      ? "external" : "internal"));
}

static ssize_t metadata_store(mddev_t *mddev, const char *buf, size_t len)
{
	if (mddev->bitmap ||
	    mddev->bitmap_info.file ||
	    mddev->bitmap_info.offset)
		return -EBUSY;
	if (strncmp(buf, "external", 8) == 0)
		mddev->bitmap_info.external = 1;
	else if (strncmp(buf, "internal", 8) == 0)
		mddev->bitmap_info.external = 0;
	else
		return -EINVAL;
	return len;
}

static struct md_sysfs_entry bitmap_metadata =
__ATTR(metadata, S_IRUGO|S_IWUSR, metadata_show, metadata_store);

static ssize_t can_clear_show(mddev_t *mddev, char *page)
{
	int len;
	if (mddev->bitmap)
		len = sprintf(page, "%s\n", (mddev->bitmap->need_sync ?
					     "false" : "true"));
	else
		len = sprintf(page, "\n");
	return len;
}

static ssize_t can_clear_store(mddev_t *mddev, const char *buf, size_t len)
{
	if (mddev->bitmap == NULL)
		return -ENOENT;
	if (strncmp(buf, "false", 5) == 0)
		mddev->bitmap->need_sync = 1;
	else if (strncmp(buf, "true", 4) == 0) {
		if (mddev->degraded)
			return -EBUSY;
		mddev->bitmap->need_sync = 0;
	} else
		return -EINVAL;
	return len;
}

static struct md_sysfs_entry bitmap_can_clear =
__ATTR(can_clear, S_IRUGO|S_IWUSR, can_clear_show, can_clear_store);

2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095
static ssize_t
behind_writes_used_show(mddev_t *mddev, char *page)
{
	if (mddev->bitmap == NULL)
		return sprintf(page, "0\n");
	return sprintf(page, "%lu\n",
		       mddev->bitmap->behind_writes_used);
}

static ssize_t
behind_writes_used_reset(mddev_t *mddev, const char *buf, size_t len)
{
	if (mddev->bitmap)
		mddev->bitmap->behind_writes_used = 0;
	return len;
}

static struct md_sysfs_entry max_backlog_used =
__ATTR(max_backlog_used, S_IRUGO | S_IWUSR,
       behind_writes_used_show, behind_writes_used_reset);

2096 2097 2098 2099 2100
static struct attribute *md_bitmap_attrs[] = {
	&bitmap_location.attr,
	&bitmap_timeout.attr,
	&bitmap_backlog.attr,
	&bitmap_chunksize.attr,
2101 2102
	&bitmap_metadata.attr,
	&bitmap_can_clear.attr,
2103
	&max_backlog_used.attr,
2104 2105 2106 2107 2108 2109 2110
	NULL
};
struct attribute_group md_bitmap_group = {
	.name = "bitmap",
	.attrs = md_bitmap_attrs,
};