bitmap.c 53.5 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
/*
 * 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).
 * wait if count gets too high, wake when it drops to half.
 */

19
#include <linux/blkdev.h>
20 21 22 23 24 25 26 27 28 29
#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>
30
#include "md.h"
31
#include "bitmap.h"
32 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 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77

/* 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

//#define DPRINTK PRINTK /* set this NULL to avoid verbose debug output */
#define DPRINTK(x...) do { } while(0)

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

static inline char * bmname(struct bitmap *bitmap)
{
	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;

78
#ifdef INJECT_FAULTS_1
79 80 81 82 83 84 85
	page = NULL;
#else
	page = kmalloc(PAGE_SIZE, GFP_NOIO);
#endif
	if (!page)
		printk("%s: bitmap_alloc_page FAILED\n", bmname(bitmap));
	else
86
		PRINTK("%s: bitmap_alloc_page: allocated page at %p\n",
87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110
			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
 */
static int bitmap_checkpage(struct bitmap *bitmap, unsigned long page, int create)
111 112
__releases(bitmap->lock)
__acquires(bitmap->lock)
113 114 115 116
{
	unsigned char *mappage;

	if (page >= bitmap->pages) {
117 118 119 120
		/* This can happen if bitmap_start_sync goes beyond
		 * End-of-device while looking for a whole page.
		 * It is harmless.
		 */
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
		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;

	spin_unlock_irq(&bitmap->lock);

	/* this page has not been allocated yet */

	if ((mappage = bitmap_alloc_page(bitmap)) == NULL) {
		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 */
		spin_lock_irq(&bitmap->lock);
		if (!bitmap->bp[page].map)
			bitmap->bp[page].hijacked = 1;
		goto out;
	}

	/* got a page */

	spin_lock_irq(&bitmap->lock);

	/* recheck the page */

	if (bitmap->bp[page].map || bitmap->bp[page].hijacked) {
		/* somebody beat us to getting the page */
		bitmap_free_page(bitmap, mappage);
		return 0;
	}

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

	memset(mappage, 0, PAGE_SIZE);
	bitmap->bp[page].map = mappage;
	bitmap->missing_pages--;
out:
	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 */
174
static void bitmap_checkfree(struct bitmap *bitmap, unsigned long page)
175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213
{
	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;
		return;
	}

	/* normal case, free the page */

#if 0
/* actually ... let's not.  We will probably need the page again exactly when
 * memory is tight and we are flusing to disk
 */
	return;
#else
	ptr = bitmap->bp[page].map;
	bitmap->bp[page].map = NULL;
	bitmap->missing_pages++;
	bitmap_free_page(bitmap, ptr);
	return;
#endif
}


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

/*
 * basic page I/O operations
 */

214
/* IO operations when bitmap is stored near all superblocks */
215
static struct page *read_sb_page(mddev_t *mddev, loff_t offset,
216 217
				 struct page *page,
				 unsigned long index, int size)
218 219 220 221 222 223
{
	/* choose a good rdev and read the page from there */

	mdk_rdev_t *rdev;
	sector_t target;

224 225
	if (!page)
		page = alloc_page(GFP_KERNEL);
226 227 228
	if (!page)
		return ERR_PTR(-ENOMEM);

229
	list_for_each_entry(rdev, &mddev->disks, same_set) {
230 231
		if (! test_bit(In_sync, &rdev->flags)
		    || test_bit(Faulty, &rdev->flags))
232 233
			continue;

234
		target = rdev->sb_start + offset + index * (PAGE_SIZE/512);
235

236
		if (sync_page_io(rdev->bdev, target,
237
				 roundup(size, bdev_logical_block_size(rdev->bdev)),
238
				 page, READ)) {
239
			page->index = index;
240 241
			attach_page_buffers(page, NULL); /* so that free_buffer will
							  * quietly no-op */
242 243 244 245
			return page;
		}
	}
	return ERR_PTR(-EIO);
246 247 248

}

249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282
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;
}

283
static int write_sb_page(struct bitmap *bitmap, struct page *page, int wait)
284
{
285
	mdk_rdev_t *rdev = NULL;
286
	mddev_t *mddev = bitmap->mddev;
287

288
	while ((rdev = next_active_rdev(rdev, mddev)) != NULL) {
289
			int size = PAGE_SIZE;
290
			loff_t offset = mddev->bitmap_info.offset;
291 292
			if (page->index == bitmap->file_pages-1)
				size = roundup(bitmap->last_page_size,
293
					       bdev_logical_block_size(rdev->bdev));
294 295 296
			/* Just make sure we aren't corrupting data or
			 * metadata
			 */
297 298 299 300 301 302 303 304 305
			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) {
306
				/* DATA  BITMAP METADATA  */
307
				if (offset
308
				    + (long)(page->index * (PAGE_SIZE/512))
309 310
				    + size/512 > 0)
					/* bitmap runs in to metadata */
311
					goto bad_alignment;
A
Andre Noll 已提交
312
				if (rdev->data_offset + mddev->dev_sectors
313
				    > rdev->sb_start + offset)
314
					/* data runs in to bitmap */
315
					goto bad_alignment;
316
			} else if (rdev->sb_start < rdev->data_offset) {
317
				/* METADATA BITMAP DATA */
318
				if (rdev->sb_start
319
				    + offset
320 321 322
				    + page->index*(PAGE_SIZE/512) + size/512
				    > rdev->data_offset)
					/* bitmap runs in to data */
323
					goto bad_alignment;
324 325 326
			} else {
				/* DATA METADATA BITMAP - no problems */
			}
327
			md_super_write(mddev, rdev,
328
				       rdev->sb_start + offset
329
				       + page->index * (PAGE_SIZE/512),
330
				       size,
331
				       page);
332
	}
333 334

	if (wait)
335
		md_super_wait(mddev);
336
	return 0;
337 338 339

 bad_alignment:
	return -EINVAL;
340 341
}

342
static void bitmap_file_kick(struct bitmap *bitmap);
343
/*
344
 * write out a page to a file
345
 */
346
static void write_page(struct bitmap *bitmap, struct page *page, int wait)
347
{
348
	struct buffer_head *bh;
349

350 351 352 353 354
	if (bitmap->file == NULL) {
		switch (write_sb_page(bitmap, page, wait)) {
		case -EINVAL:
			bitmap->flags |= BITMAP_WRITE_ERROR;
		}
355
	} else {
356

357
		bh = page_buffers(page);
358

359 360 361 362 363 364 365
		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;
		}
366

367 368 369 370
		if (wait) {
			wait_event(bitmap->write_wait,
				   atomic_read(&bitmap->pending_writes)==0);
		}
371
	}
372 373
	if (bitmap->flags & BITMAP_WRITE_ERROR)
		bitmap_file_kick(bitmap);
374 375 376 377 378 379
}

static void end_bitmap_write(struct buffer_head *bh, int uptodate)
{
	struct bitmap *bitmap = bh->b_private;
	unsigned long flags;
380

381 382 383 384
	if (!uptodate) {
		spin_lock_irqsave(&bitmap->lock, flags);
		bitmap->flags |= BITMAP_WRITE_ERROR;
		spin_unlock_irqrestore(&bitmap->lock, flags);
385
	}
386 387 388
	if (atomic_dec_and_test(&bitmap->pending_writes))
		wake_up(&bitmap->write_wait);
}
389

390 391 392 393 394 395 396 397 398 399 400
/* 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);
401

402 403 404 405
	while (bh) {
		struct buffer_head *next = bh->b_this_page;
		free_buffer_head(bh);
		bh = next;
406
	}
407 408
	__clear_page_buffers(page);
	put_page(page);
409 410
}

411 412 413 414 415 416 417
/* 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.
 */
418
static struct page *read_page(struct file *file, unsigned long index,
419 420
			      struct bitmap *bitmap,
			      unsigned long count)
421 422
{
	struct page *page = NULL;
J
Josef Sipek 已提交
423
	struct inode *inode = file->f_path.dentry->d_inode;
424 425
	struct buffer_head *bh;
	sector_t block;
426

427 428
	PRINTK("read bitmap file (%dB @ %Lu)\n", (int)PAGE_SIZE,
			(unsigned long long)index << PAGE_SHIFT);
429

430 431 432
	page = alloc_page(GFP_KERNEL);
	if (!page)
		page = ERR_PTR(-ENOMEM);
433 434
	if (IS_ERR(page))
		goto out;
435 436 437

	bh = alloc_page_buffers(page, 1<<inode->i_blkbits, 0);
	if (!bh) {
438
		put_page(page);
439
		page = ERR_PTR(-ENOMEM);
440 441
		goto out;
	}
442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462
	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;
463 464 465 466
			atomic_inc(&bitmap->pending_writes);
			set_buffer_locked(bh);
			set_buffer_mapped(bh);
			submit_bh(READ, bh);
467 468 469 470 471
		}
		block++;
		bh = bh->b_this_page;
	}
	page->index = index;
472 473 474 475 476 477 478

	wait_event(bitmap->write_wait,
		   atomic_read(&bitmap->pending_writes)==0);
	if (bitmap->flags & BITMAP_WRITE_ERROR) {
		free_buffers(page);
		page = ERR_PTR(-EIO);
	}
479 480 481
out:
	if (IS_ERR(page))
		printk(KERN_ALERT "md: bitmap read error: (%dB @ %Lu): %ld\n",
482 483
			(int)PAGE_SIZE,
			(unsigned long long)index << PAGE_SHIFT,
484 485 486 487 488 489 490 491 492
			PTR_ERR(page));
	return page;
}

/*
 * bitmap file superblock operations
 */

/* update the event counter and sync the superblock to disk */
493
void bitmap_update_sb(struct bitmap *bitmap)
494 495 496 497 498
{
	bitmap_super_t *sb;
	unsigned long flags;

	if (!bitmap || !bitmap->mddev) /* no bitmap for this array */
499
		return;
500 501
	if (bitmap->mddev->bitmap_info.external)
		return;
502 503 504
	spin_lock_irqsave(&bitmap->lock, flags);
	if (!bitmap->sb_page) { /* no superblock */
		spin_unlock_irqrestore(&bitmap->lock, flags);
505
		return;
506 507
	}
	spin_unlock_irqrestore(&bitmap->lock, flags);
508
	sb = kmap_atomic(bitmap->sb_page, KM_USER0);
509
	sb->events = cpu_to_le64(bitmap->mddev->events);
510 511 512 513 514
	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);
	}
515 516 517
	/* 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);
518
	kunmap_atomic(sb, KM_USER0);
519
	write_page(bitmap, bitmap->sb_page, 1);
520 521 522 523 524 525 526 527 528
}

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

	if (!bitmap || !bitmap->sb_page)
		return;
529
	sb = kmap_atomic(bitmap->sb_page, KM_USER0);
530
	printk(KERN_DEBUG "%s: bitmap file superblock:\n", bmname(bitmap));
531 532 533
	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",
534 535 536 537
					*(__u32 *)(sb->uuid+0),
					*(__u32 *)(sb->uuid+4),
					*(__u32 *)(sb->uuid+8),
					*(__u32 *)(sb->uuid+12));
538
	printk(KERN_DEBUG "        events: %llu\n",
539
			(unsigned long long) le64_to_cpu(sb->events));
540
	printk(KERN_DEBUG "events cleared: %llu\n",
541
			(unsigned long long) le64_to_cpu(sb->events_cleared));
542 543 544 545 546
	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);
547
	printk(KERN_DEBUG "max write behind: %d\n", le32_to_cpu(sb->write_behind));
548
	kunmap_atomic(sb, KM_USER0);
549 550 551 552 553 554 555
}

/* 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;
556
	unsigned long chunksize, daemon_sleep, write_behind;
557 558 559 560
	unsigned long long events;
	int err = -EINVAL;

	/* page 0 is the superblock, read it... */
561 562 563 564 565 566
	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 {
567 568
		bitmap->sb_page = read_sb_page(bitmap->mddev,
					       bitmap->mddev->bitmap_info.offset,
569 570
					       NULL,
					       0, sizeof(bitmap_super_t));
571
	}
572 573 574 575 576 577
	if (IS_ERR(bitmap->sb_page)) {
		err = PTR_ERR(bitmap->sb_page);
		bitmap->sb_page = NULL;
		return err;
	}

578
	sb = kmap_atomic(bitmap->sb_page, KM_USER0);
579 580

	chunksize = le32_to_cpu(sb->chunksize);
581
	daemon_sleep = le32_to_cpu(sb->daemon_sleep) * HZ;
582
	write_behind = le32_to_cpu(sb->write_behind);
583 584 585 586

	/* verify that the bitmap-specific fields are valid */
	if (sb->magic != cpu_to_le32(BITMAP_MAGIC))
		reason = "bad magic";
587 588
	else if (le32_to_cpu(sb->version) < BITMAP_MAJOR_LO ||
		 le32_to_cpu(sb->version) > BITMAP_MAJOR_HI)
589
		reason = "unrecognized superblock version";
590
	else if (chunksize < 512)
591
		reason = "bitmap chunksize too small";
592 593
	else if ((1 << ffz(~chunksize)) != chunksize)
		reason = "bitmap chunksize not a power of 2";
594
	else if (daemon_sleep < 1 || daemon_sleep > MAX_SCHEDULE_TIMEOUT)
595
		reason = "daemon sleep period out of range";
596 597
	else if (write_behind > COUNTER_MAX)
		reason = "write-behind limit out of range (0 - 16383)";
598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623
	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);
624
		sb->state |= cpu_to_le32(BITMAP_STALE);
625 626 627
	}
success:
	/* assign fields using values from superblock */
628 629 630
	bitmap->mddev->bitmap_info.chunksize = chunksize;
	bitmap->mddev->bitmap_info.daemon_sleep = daemon_sleep;
	bitmap->mddev->bitmap_info.max_write_behind = write_behind;
631
	bitmap->flags |= le32_to_cpu(sb->state);
632 633
	if (le32_to_cpu(sb->version) == BITMAP_MAJOR_HOSTENDIAN)
		bitmap->flags |= BITMAP_HOSTENDIAN;
634
	bitmap->events_cleared = le64_to_cpu(sb->events_cleared);
635
	if (sb->state & cpu_to_le32(BITMAP_STALE))
636
		bitmap->events_cleared = bitmap->mddev->events;
637 638
	err = 0;
out:
639
	kunmap_atomic(sb, KM_USER0);
640 641 642 643 644 645 646 647 648 649
	if (err)
		bitmap_print_sb(bitmap);
	return err;
}

enum bitmap_mask_op {
	MASK_SET,
	MASK_UNSET
};

650 651 652
/* 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)
653 654 655
{
	bitmap_super_t *sb;
	unsigned long flags;
656
	int old;
657 658

	spin_lock_irqsave(&bitmap->lock, flags);
659
	if (!bitmap->sb_page) { /* can't set the state */
660
		spin_unlock_irqrestore(&bitmap->lock, flags);
661
		return 0;
662 663
	}
	spin_unlock_irqrestore(&bitmap->lock, flags);
664
	sb = kmap_atomic(bitmap->sb_page, KM_USER0);
665
	old = le32_to_cpu(sb->state) & bits;
666
	switch (op) {
667
		case MASK_SET: sb->state |= cpu_to_le32(bits);
668
				break;
669
		case MASK_UNSET: sb->state &= cpu_to_le32(~bits);
670 671 672
				break;
		default: BUG();
	}
673
	kunmap_atomic(sb, KM_USER0);
674
	return old;
675 676 677 678 679 680
}

/*
 * general bitmap file operations
 */

681 682 683 684 685 686
/*
 * 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.
 */
687
/* calculate the index of the page that contains this bit */
688
static inline unsigned long file_page_index(struct bitmap *bitmap, unsigned long chunk)
689
{
690 691 692
	if (!bitmap->mddev->bitmap_info.external)
		chunk += sizeof(bitmap_super_t) << 3;
	return chunk >> PAGE_BIT_SHIFT;
693 694 695
}

/* calculate the (bit) offset of this bit within a page */
696
static inline unsigned long file_page_offset(struct bitmap *bitmap, unsigned long chunk)
697
{
698 699 700
	if (!bitmap->mddev->bitmap_info.external)
		chunk += sizeof(bitmap_super_t) << 3;
	return chunk & (PAGE_BITS - 1);
701 702 703 704 705 706 707 708 709 710 711 712
}

/*
 * 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)
{
713 714 715
	if (file_page_index(bitmap, chunk) >= bitmap->file_pages) return NULL;
	return bitmap->filemap[file_page_index(bitmap, chunk)
			       - file_page_index(bitmap, 0)];
716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737
}


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--)
738
		if (map[pages] != sb_page) /* 0 is sb_page, release it below */
739
			free_buffers(map[pages]);
740 741 742
	kfree(map);
	kfree(attr);

743 744
	if (sb_page)
		free_buffers(sb_page);
745 746 747 748 749 750 751 752 753 754 755 756
}

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);

757 758 759
	if (file)
		wait_event(bitmap->write_wait,
			   atomic_read(&bitmap->pending_writes)==0);
760 761
	bitmap_file_unmap(bitmap);

762
	if (file) {
J
Josef Sipek 已提交
763
		struct inode *inode = file->f_path.dentry->d_inode;
764
		invalidate_mapping_pages(inode->i_mapping, 0, -1);
765
		fput(file);
766
	}
767 768 769 770 771 772 773 774 775 776 777 778
}


/*
 * 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;

779 780
	if (bitmap_mask_state(bitmap, BITMAP_STALE, MASK_SET) == 0) {
		bitmap_update_sb(bitmap);
781

782 783 784
		if (bitmap->file) {
			path = kmalloc(PAGE_SIZE, GFP_KERNEL);
			if (path)
C
Christoph Hellwig 已提交
785 786 787
				ptr = d_path(&bitmap->file->f_path, path,
					     PAGE_SIZE);

788

789 790
			printk(KERN_ALERT
			      "%s: kicking failed bitmap file %s from array!\n",
C
Christoph Hellwig 已提交
791
			      bmname(bitmap), IS_ERR(ptr) ? "" : ptr);
792

793 794 795 796 797
			kfree(path);
		} else
			printk(KERN_ALERT
			       "%s: disabling internal bitmap due to errors\n",
			       bmname(bitmap));
798
	}
799 800 801 802 803 804 805

	bitmap_file_put(bitmap);

	return;
}

enum bitmap_page_attr {
806 807 808
	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
809 810 811 812 813
};

static inline void set_page_attr(struct bitmap *bitmap, struct page *page,
				enum bitmap_page_attr attr)
{
814
	__set_bit((page->index<<2) + attr, bitmap->filemap_attr);
815 816 817 818 819
}

static inline void clear_page_attr(struct bitmap *bitmap, struct page *page,
				enum bitmap_page_attr attr)
{
820
	__clear_bit((page->index<<2) + attr, bitmap->filemap_attr);
821 822
}

823 824
static inline unsigned long test_page_attr(struct bitmap *bitmap, struct page *page,
					   enum bitmap_page_attr attr)
825
{
826
	return test_bit((page->index<<2) + attr, bitmap->filemap_attr);
827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842
}

/*
 * 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;
	struct page *page;
	void *kaddr;
	unsigned long chunk = block >> CHUNK_BLOCK_SHIFT(bitmap);

843
	if (!bitmap->filemap) {
844 845 846 847
		return;
	}

	page = filemap_get_page(bitmap, chunk);
848
	if (!page) return;
849
	bit = file_page_offset(bitmap, chunk);
850 851 852

 	/* set the bit */
	kaddr = kmap_atomic(page, KM_USER0);
853 854 855 856
	if (bitmap->flags & BITMAP_HOSTENDIAN)
		set_bit(bit, kaddr);
	else
		ext2_set_bit(bit, kaddr);
857 858 859 860 861 862 863 864 865 866 867
	kunmap_atomic(kaddr, KM_USER0);
	PRINTK("set file bit %lu page %lu\n", bit, page->index);

	/* 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 */
868
void bitmap_unplug(struct bitmap *bitmap)
869
{
870 871
	unsigned long i, flags;
	int dirty, need_write;
872 873 874 875
	struct page *page;
	int wait = 0;

	if (!bitmap)
876
		return;
877 878 879 880 881

	/* 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);
882
		if (!bitmap->filemap) {
883
			spin_unlock_irqrestore(&bitmap->lock, flags);
884
			return;
885 886
		}
		page = bitmap->filemap[i];
887 888
		dirty = test_page_attr(bitmap, page, BITMAP_PAGE_DIRTY);
		need_write = test_page_attr(bitmap, page, BITMAP_PAGE_NEEDWRITE);
889 890
		clear_page_attr(bitmap, page, BITMAP_PAGE_DIRTY);
		clear_page_attr(bitmap, page, BITMAP_PAGE_NEEDWRITE);
891
		if (dirty)
892 893 894
			wait = 1;
		spin_unlock_irqrestore(&bitmap->lock, flags);

895
		if (dirty | need_write)
896
			write_page(bitmap, page, 0);
897 898
	}
	if (wait) { /* if any writes were performed, we need to wait on them */
899
		if (bitmap->file)
900 901
			wait_event(bitmap->write_wait,
				   atomic_read(&bitmap->pending_writes)==0);
902
		else
903
			md_super_wait(bitmap->mddev);
904
	}
905 906
	if (bitmap->flags & BITMAP_WRITE_ERROR)
		bitmap_file_kick(bitmap);
907 908
}

909
static void bitmap_set_memory_bits(struct bitmap *bitmap, sector_t offset, int needed);
910 911 912 913 914 915 916
/* * 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.
917 918 919
 *
 * We ignore all bits for sectors that end earlier than 'start'.
 * This is used when reading an out-of-date bitmap...
920
 */
921
static int bitmap_init_from_disk(struct bitmap *bitmap, sector_t start)
922 923 924 925 926
{
	unsigned long i, chunks, index, oldindex, bit;
	struct page *page = NULL, *oldpage = NULL;
	unsigned long num_pages, bit_cnt = 0;
	struct file *file;
927
	unsigned long bytes, offset;
928 929
	int outofdate;
	int ret = -ENOSPC;
930
	void *paddr;
931 932 933 934

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

935
	BUG_ON(!file && !bitmap->mddev->bitmap_info.offset);
936

937
#ifdef INJECT_FAULTS_3
938 939 940 941 942 943 944 945 946
	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));

	bytes = (chunks + 7) / 8;
947 948
	if (!bitmap->mddev->bitmap_info.external)
		bytes += sizeof(bitmap_super_t);
949

950 951
	
	num_pages = (bytes + PAGE_SIZE - 1) / PAGE_SIZE;
952

953
	if (file && i_size_read(file->f_mapping->host) < bytes) {
954 955 956
		printk(KERN_INFO "%s: bitmap file too short %lu < %lu\n",
			bmname(bitmap),
			(unsigned long) i_size_read(file->f_mapping->host),
957
			bytes);
958
		goto err;
959
	}
960 961 962

	ret = -ENOMEM;

963
	bitmap->filemap = kmalloc(sizeof(struct page *) * num_pages, GFP_KERNEL);
964
	if (!bitmap->filemap)
965
		goto err;
966

967 968
	/* We need 4 bits per page, rounded up to a multiple of sizeof(unsigned long) */
	bitmap->filemap_attr = kzalloc(
969
		roundup( DIV_ROUND_UP(num_pages*4, 8), sizeof(unsigned long)),
970
		GFP_KERNEL);
971
	if (!bitmap->filemap_attr)
972
		goto err;
973 974 975 976

	oldindex = ~0L;

	for (i = 0; i < chunks; i++) {
977
		int b;
978 979
		index = file_page_index(bitmap, i);
		bit = file_page_offset(bitmap, i);
980
		if (index != oldindex) { /* this is a new page, read it in */
981
			int count;
982
			/* unmap the old page, we're done with it */
983
			if (index == num_pages-1)
984
				count = bytes - index * PAGE_SIZE;
985 986
			else
				count = PAGE_SIZE;
987
			if (index == 0 && bitmap->sb_page) {
988 989 990 991 992 993 994
				/*
				 * 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
995 996
				if (!file)
					read_sb_page(bitmap->mddev,
997
						     bitmap->mddev->bitmap_info.offset,
N
NeilBrown 已提交
998 999
						     page,
						     index, count);
1000
			} else if (file) {
1001
				page = read_page(file, index, bitmap, count);
1002 1003
				offset = 0;
			} else {
1004 1005
				page = read_sb_page(bitmap->mddev,
						    bitmap->mddev->bitmap_info.offset,
1006 1007
						    NULL,
						    index, count);
1008 1009
				offset = 0;
			}
1010 1011
			if (IS_ERR(page)) { /* read error */
				ret = PTR_ERR(page);
1012
				goto err;
1013 1014
			}

1015 1016 1017
			oldindex = index;
			oldpage = page;

1018 1019 1020
			bitmap->filemap[bitmap->file_pages++] = page;
			bitmap->last_page_size = count;

1021 1022 1023 1024 1025
			if (outofdate) {
				/*
				 * if bitmap is out of date, dirty the
			 	 * whole page and write it out
				 */
1026 1027
				paddr = kmap_atomic(page, KM_USER0);
				memset(paddr + offset, 0xff,
1028
				       PAGE_SIZE - offset);
1029
				kunmap_atomic(paddr, KM_USER0);
1030 1031 1032
				write_page(bitmap, page, 1);

				ret = -EIO;
1033
				if (bitmap->flags & BITMAP_WRITE_ERROR)
1034
					goto err;
1035 1036
			}
		}
1037
		paddr = kmap_atomic(page, KM_USER0);
1038
		if (bitmap->flags & BITMAP_HOSTENDIAN)
1039
			b = test_bit(bit, paddr);
1040
		else
1041 1042
			b = ext2_test_bit(bit, paddr);
		kunmap_atomic(paddr, KM_USER0);
1043
		if (b) {
1044
			/* if the disk bit is set, set the memory bit */
1045 1046 1047 1048 1049
			int needed = ((sector_t)(i+1) << (CHUNK_BLOCK_SHIFT(bitmap))
				      >= start);
			bitmap_set_memory_bits(bitmap,
					       (sector_t)i << CHUNK_BLOCK_SHIFT(bitmap),
					       needed);
1050
			bit_cnt++;
1051
			set_page_attr(bitmap, page, BITMAP_PAGE_CLEAN);
1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064
		}
	}

 	/* everything went OK */
	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: "
1065 1066 1067 1068
		"read %lu/%lu pages, set %lu bits\n",
		bmname(bitmap), bitmap->file_pages, num_pages, bit_cnt);

	return 0;
1069

1070 1071 1072
 err:
	printk(KERN_INFO "%s: bitmap initialisation failed: %d\n",
	       bmname(bitmap), ret);
1073 1074 1075
	return ret;
}

1076 1077 1078 1079 1080
void bitmap_write_all(struct bitmap *bitmap)
{
	/* We don't actually write all bitmap blocks here,
	 * just flag them as needing to be written
	 */
1081
	int i;
1082

1083 1084 1085
	for (i=0; i < bitmap->file_pages; i++)
		set_page_attr(bitmap, bitmap->filemap[i],
			      BITMAP_PAGE_NEEDWRITE);
1086 1087
}

1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108

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;
/*
	if (page == 0) printk("count page 0, offset %llu: %d gives %d\n",
			      (unsigned long long)offset, inc, bitmap->bp[page].count);
*/
	bitmap_checkfree(bitmap, page);
}
static bitmap_counter_t *bitmap_get_counter(struct bitmap *bitmap,
					    sector_t offset, int *blocks,
					    int create);

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

1109
void bitmap_daemon_work(mddev_t *mddev)
1110
{
1111
	struct bitmap *bitmap;
1112
	unsigned long j;
1113 1114 1115
	unsigned long flags;
	struct page *page = NULL, *lastpage = NULL;
	int blocks;
1116
	void *paddr;
1117

1118 1119 1120
	/* Use a mutex to guard daemon_work against
	 * bitmap_destroy.
	 */
1121
	mutex_lock(&mddev->bitmap_info.mutex);
1122 1123
	bitmap = mddev->bitmap;
	if (bitmap == NULL) {
1124
		mutex_unlock(&mddev->bitmap_info.mutex);
1125
		return;
1126
	}
1127
	if (time_before(jiffies, bitmap->daemon_lastrun
1128
			+ bitmap->mddev->bitmap_info.daemon_sleep))
1129 1130
		goto done;

1131
	bitmap->daemon_lastrun = jiffies;
1132 1133
	if (bitmap->allclean) {
		bitmap->mddev->thread->timeout = MAX_SCHEDULE_TIMEOUT;
1134
		goto done;
1135 1136
	}
	bitmap->allclean = 1;
1137

1138
	spin_lock_irqsave(&bitmap->lock, flags);
1139 1140
	for (j = 0; j < bitmap->chunks; j++) {
		bitmap_counter_t *bmc;
1141
		if (!bitmap->filemap)
1142 1143 1144 1145 1146 1147
			/* error or shutdown */
			break;

		page = filemap_get_page(bitmap, j);

		if (page != lastpage) {
1148
			/* skip this page unless it's marked as needing cleaning */
1149 1150 1151
			if (!test_page_attr(bitmap, page, BITMAP_PAGE_CLEAN)) {
				int need_write = test_page_attr(bitmap, page,
								BITMAP_PAGE_NEEDWRITE);
1152
				if (need_write)
1153
					clear_page_attr(bitmap, page, BITMAP_PAGE_NEEDWRITE);
1154

1155
				spin_unlock_irqrestore(&bitmap->lock, flags);
1156
				if (need_write) {
1157
					write_page(bitmap, page, 0);
1158 1159
					bitmap->allclean = 0;
				}
1160 1161
				spin_lock_irqsave(&bitmap->lock, flags);
				j |= (PAGE_BITS - 1);
1162 1163 1164
				continue;
			}

1165 1166
			/* grab the new page, sync and release the old */
			if (lastpage != NULL) {
1167
				if (test_page_attr(bitmap, lastpage, BITMAP_PAGE_NEEDWRITE)) {
1168 1169
					clear_page_attr(bitmap, lastpage, BITMAP_PAGE_NEEDWRITE);
					spin_unlock_irqrestore(&bitmap->lock, flags);
1170
					write_page(bitmap, lastpage, 0);
1171 1172 1173 1174 1175 1176 1177
				} else {
					set_page_attr(bitmap, lastpage, BITMAP_PAGE_NEEDWRITE);
					spin_unlock_irqrestore(&bitmap->lock, flags);
				}
			} else
				spin_unlock_irqrestore(&bitmap->lock, flags);
			lastpage = page;
1178 1179 1180 1181

			/* We are possibly going to clear some bits, so make
			 * sure that events_cleared is up-to-date.
			 */
1182 1183
			if (bitmap->need_sync &&
			    bitmap->mddev->bitmap_info.external == 0) {
1184 1185 1186 1187 1188 1189 1190 1191
				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);
			}
1192
			spin_lock_irqsave(&bitmap->lock, flags);
1193 1194
			if (!bitmap->need_sync)
				clear_page_attr(bitmap, page, BITMAP_PAGE_CLEAN);
1195
		}
1196 1197 1198
		bmc = bitmap_get_counter(bitmap,
					 (sector_t)j << CHUNK_BLOCK_SHIFT(bitmap),
					 &blocks, 0);
1199 1200 1201 1202
		if (bmc) {
/*
  if (j < 100) printk("bitmap: j=%lu, *bmc = 0x%x\n", j, *bmc);
*/
1203 1204 1205
			if (*bmc)
				bitmap->allclean = 0;

1206 1207 1208
			if (*bmc == 2) {
				*bmc=1; /* maybe clear the bit next time */
				set_page_attr(bitmap, page, BITMAP_PAGE_CLEAN);
1209
			} else if (*bmc == 1 && !bitmap->need_sync) {
1210 1211
				/* we can clear the bit */
				*bmc = 0;
1212 1213
				bitmap_count_page(bitmap,
						  (sector_t)j << CHUNK_BLOCK_SHIFT(bitmap),
1214 1215 1216
						  -1);

				/* clear the bit */
1217
				paddr = kmap_atomic(page, KM_USER0);
1218
				if (bitmap->flags & BITMAP_HOSTENDIAN)
1219 1220
					clear_bit(file_page_offset(bitmap, j),
						  paddr);
1221
				else
1222 1223
					ext2_clear_bit(file_page_offset(bitmap, j),
						       paddr);
1224
				kunmap_atomic(paddr, KM_USER0);
1225
			}
1226 1227
		} else
			j |= PAGE_COUNTER_MASK;
1228
	}
1229
	spin_unlock_irqrestore(&bitmap->lock, flags);
1230 1231 1232 1233

	/* now sync the final page */
	if (lastpage != NULL) {
		spin_lock_irqsave(&bitmap->lock, flags);
1234
		if (test_page_attr(bitmap, lastpage, BITMAP_PAGE_NEEDWRITE)) {
1235 1236
			clear_page_attr(bitmap, lastpage, BITMAP_PAGE_NEEDWRITE);
			spin_unlock_irqrestore(&bitmap->lock, flags);
1237
			write_page(bitmap, lastpage, 0);
1238 1239 1240 1241 1242 1243
		} else {
			set_page_attr(bitmap, lastpage, BITMAP_PAGE_NEEDWRITE);
			spin_unlock_irqrestore(&bitmap->lock, flags);
		}
	}

1244
 done:
1245
	if (bitmap->allclean == 0)
1246
		bitmap->mddev->thread->timeout = 
1247
			bitmap->mddev->bitmap_info.daemon_sleep;
1248
	mutex_unlock(&mddev->bitmap_info.mutex);
1249 1250 1251 1252 1253
}

static bitmap_counter_t *bitmap_get_counter(struct bitmap *bitmap,
					    sector_t offset, int *blocks,
					    int create)
1254 1255
__releases(bitmap->lock)
__acquires(bitmap->lock)
1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289
{
	/* 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;

	if (bitmap_checkpage(bitmap, page, create) < 0) {
		csize = ((sector_t)1) << (CHUNK_BLOCK_SHIFT(bitmap));
		*blocks = csize - (offset & (csize- 1));
		return NULL;
	}
	/* 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);
		csize = ((sector_t)1) << (CHUNK_BLOCK_SHIFT(bitmap) +
					  PAGE_COUNTER_SHIFT - 1);
		*blocks = csize - (offset & (csize- 1));
		return  &((bitmap_counter_t *)
			  &bitmap->bp[page].map)[hi];
	} else { /* page is allocated */
		csize = ((sector_t)1) << (CHUNK_BLOCK_SHIFT(bitmap));
		*blocks = csize - (offset & (csize- 1));
		return (bitmap_counter_t *)
			&(bitmap->bp[page].map[pageoff]);
	}
}

1290
int bitmap_startwrite(struct bitmap *bitmap, sector_t offset, unsigned long sectors, int behind)
1291 1292
{
	if (!bitmap) return 0;
1293 1294

	if (behind) {
1295
		int bw;
1296
		atomic_inc(&bitmap->behind_writes);
1297 1298 1299 1300
		bw = atomic_read(&bitmap->behind_writes);
		if (bw > bitmap->behind_writes_used)
			bitmap->behind_writes_used = bw;

1301
		PRINTK(KERN_DEBUG "inc write-behind count %d/%d\n",
1302
		       bw, bitmap->max_write_behind);
1303 1304
	}

1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315
	while (sectors) {
		int blocks;
		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;
		}

1316 1317 1318 1319 1320 1321 1322 1323 1324
		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);
1325
			blk_unplug(bitmap->mddev->queue);
1326 1327 1328 1329 1330
			schedule();
			finish_wait(&bitmap->overflow_wait, &__wait);
			continue;
		}

1331 1332 1333 1334
		switch(*bmc) {
		case 0:
			bitmap_file_set_bit(bitmap, offset);
			bitmap_count_page(bitmap,offset, 1);
1335
			blk_plug_device_unlocked(bitmap->mddev->queue);
1336 1337 1338 1339
			/* fall through */
		case 1:
			*bmc = 2;
		}
1340

1341 1342 1343 1344 1345 1346 1347 1348 1349
		(*bmc)++;

		spin_unlock_irq(&bitmap->lock);

		offset += blocks;
		if (sectors > blocks)
			sectors -= blocks;
		else sectors = 0;
	}
1350
	bitmap->allclean = 0;
1351 1352 1353 1354
	return 0;
}

void bitmap_endwrite(struct bitmap *bitmap, sector_t offset, unsigned long sectors,
1355
		     int success, int behind)
1356 1357
{
	if (!bitmap) return;
1358
	if (behind) {
1359 1360
		if (atomic_dec_and_test(&bitmap->behind_writes))
			wake_up(&bitmap->behind_wait);
1361 1362 1363
		PRINTK(KERN_DEBUG "dec write-behind count %d/%d\n",
		  atomic_read(&bitmap->behind_writes), bitmap->max_write_behind);
	}
1364 1365 1366
	if (bitmap->mddev->degraded)
		/* Never clear bits or update events_cleared when degraded */
		success = 0;
1367

1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379
	while (sectors) {
		int blocks;
		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;
		}

1380 1381 1382 1383
		if (success &&
		    bitmap->events_cleared < bitmap->mddev->events) {
			bitmap->events_cleared = bitmap->mddev->events;
			bitmap->need_sync = 1;
1384
			sysfs_notify_dirent(bitmap->sysfs_can_clear);
1385 1386
		}

1387 1388 1389
		if (!success && ! (*bmc & NEEDED_MASK))
			*bmc |= NEEDED_MASK;

1390 1391 1392
		if ((*bmc & COUNTER_MAX) == COUNTER_MAX)
			wake_up(&bitmap->overflow_wait);

1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406
		(*bmc)--;
		if (*bmc <= 2) {
			set_page_attr(bitmap,
				      filemap_get_page(bitmap, offset >> CHUNK_BLOCK_SHIFT(bitmap)),
				      BITMAP_PAGE_CLEAN);
		}
		spin_unlock_irqrestore(&bitmap->lock, flags);
		offset += blocks;
		if (sectors > blocks)
			sectors -= blocks;
		else sectors = 0;
	}
}

1407 1408
static int __bitmap_start_sync(struct bitmap *bitmap, sector_t offset, int *blocks,
			       int degraded)
1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424
{
	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;
1425 1426 1427 1428
			if (!degraded) { /* don't set/clear bits if degraded */
				*bmc |= RESYNC_MASK;
				*bmc &= ~NEEDED_MASK;
			}
1429 1430 1431
		}
	}
	spin_unlock_irq(&bitmap->lock);
1432
	bitmap->allclean = 0;
1433 1434 1435
	return rv;
}

1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458
int bitmap_start_sync(struct bitmap *bitmap, sector_t offset, int *blocks,
		      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;
	int blocks1;

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

1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491
void bitmap_end_sync(struct bitmap *bitmap, sector_t offset, int *blocks, int aborted)
{
	bitmap_counter_t *bmc;
	unsigned long flags;
/*
	if (offset == 0) printk("bitmap_end_sync 0 (%d)\n", aborted);
*/	if (bitmap == NULL) {
		*blocks = 1024;
		return;
	}
	spin_lock_irqsave(&bitmap->lock, flags);
	bmc = bitmap_get_counter(bitmap, offset, blocks, 0);
	if (bmc == NULL)
		goto unlock;
	/* locked */
/*
	if (offset == 0) printk("bitmap_end sync found 0x%x, blocks %d\n", *bmc, *blocks);
*/
	if (RESYNC(*bmc)) {
		*bmc &= ~RESYNC_MASK;

		if (!NEEDED(*bmc) && aborted)
			*bmc |= NEEDED_MASK;
		else {
			if (*bmc <= 2) {
				set_page_attr(bitmap,
					      filemap_get_page(bitmap, offset >> CHUNK_BLOCK_SHIFT(bitmap)),
					      BITMAP_PAGE_CLEAN);
			}
		}
	}
 unlock:
	spin_unlock_irqrestore(&bitmap->lock, flags);
1492
	bitmap->allclean = 0;
1493 1494 1495 1496 1497 1498 1499 1500 1501 1502
}

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;
	int blocks;
N
NeilBrown 已提交
1503 1504
	if (!bitmap)
		return;
1505 1506
	while (sector < bitmap->mddev->resync_max_sectors) {
		bitmap_end_sync(bitmap, sector, &blocks, 0);
N
NeilBrown 已提交
1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522
		sector += blocks;
	}
}

void bitmap_cond_end_sync(struct bitmap *bitmap, sector_t sector)
{
	sector_t s = 0;
	int blocks;

	if (!bitmap)
		return;
	if (sector == 0) {
		bitmap->last_end_sync = jiffies;
		return;
	}
	if (time_before(jiffies, (bitmap->last_end_sync
1523
				  + bitmap->mddev->bitmap_info.daemon_sleep)))
N
NeilBrown 已提交
1524 1525 1526 1527
		return;
	wait_event(bitmap->mddev->recovery_wait,
		   atomic_read(&bitmap->mddev->recovery_active) == 0);

1528 1529
	bitmap->mddev->curr_resync_completed = bitmap->mddev->curr_resync;
	set_bit(MD_CHANGE_CLEAN, &bitmap->mddev->flags);
N
NeilBrown 已提交
1530 1531 1532 1533 1534
	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;
1535
	}
N
NeilBrown 已提交
1536
	bitmap->last_end_sync = jiffies;
1537
	sysfs_notify(&bitmap->mddev->kobj, NULL, "sync_completed");
1538 1539
}

1540
static void bitmap_set_memory_bits(struct bitmap *bitmap, sector_t offset, int needed)
1541 1542
{
	/* For each chunk covered by any of these sectors, set the
1543
	 * counter to 1 and set resync_needed.  They should all
1544 1545
	 * be 0 at this point
	 */
1546 1547 1548 1549 1550 1551

	int secs;
	bitmap_counter_t *bmc;
	spin_lock_irq(&bitmap->lock);
	bmc = bitmap_get_counter(bitmap, offset, &secs, 1);
	if (!bmc) {
1552
		spin_unlock_irq(&bitmap->lock);
1553
		return;
1554
	}
1555 1556
	if (! *bmc) {
		struct page *page;
1557
		*bmc = 1 | (needed?NEEDED_MASK:0);
1558 1559 1560 1561 1562
		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);
1563
	bitmap->allclean = 0;
1564 1565
}

1566 1567 1568 1569 1570 1571
/* 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++) {
1572
		sector_t sec = (sector_t)chunk << CHUNK_BLOCK_SHIFT(bitmap);
1573 1574
		bitmap_set_memory_bits(bitmap, sec, 1);
		bitmap_file_set_bit(bitmap, sec);
1575 1576 1577 1578 1579 1580
		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;
1581 1582 1583
	}
}

1584 1585 1586 1587 1588 1589
/*
 * flush out any pending updates
 */
void bitmap_flush(mddev_t *mddev)
{
	struct bitmap *bitmap = mddev->bitmap;
1590
	long sleep;
1591 1592 1593 1594 1595 1596 1597

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

	/* run the daemon_work three time to ensure everything is flushed
	 * that can be
	 */
1598
	sleep = mddev->bitmap_info.daemon_sleep * 2;
1599
	bitmap->daemon_lastrun -= sleep;
1600
	bitmap_daemon_work(mddev);
1601
	bitmap->daemon_lastrun -= sleep;
1602
	bitmap_daemon_work(mddev);
1603
	bitmap->daemon_lastrun -= sleep;
1604
	bitmap_daemon_work(mddev);
1605 1606 1607
	bitmap_update_sb(bitmap);
}

1608 1609 1610
/*
 * free memory that was allocated
 */
1611
static void bitmap_free(struct bitmap *bitmap)
1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633
{
	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);
}
1634

1635 1636 1637 1638 1639 1640 1641
void bitmap_destroy(mddev_t *mddev)
{
	struct bitmap *bitmap = mddev->bitmap;

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

1642
	mutex_lock(&mddev->bitmap_info.mutex);
1643
	mddev->bitmap = NULL; /* disconnect from the md device */
1644
	mutex_unlock(&mddev->bitmap_info.mutex);
1645 1646
	if (mddev->thread)
		mddev->thread->timeout = MAX_SCHEDULE_TIMEOUT;
1647

1648 1649 1650
	if (bitmap->sysfs_can_clear)
		sysfs_put(bitmap->sysfs_can_clear);

1651 1652
	bitmap_free(bitmap);
}
1653 1654 1655 1656 1657 1658 1659 1660

/*
 * 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;
1661
	sector_t blocks = mddev->resync_max_sectors;
1662 1663
	unsigned long chunks;
	unsigned long pages;
1664
	struct file *file = mddev->bitmap_info.file;
1665
	int err;
1666
	sector_t start;
1667
	struct sysfs_dirent *bm;
1668

A
Alexey Dobriyan 已提交
1669
	BUILD_BUG_ON(sizeof(bitmap_super_t) != 256);
1670

1671
	if (!file && !mddev->bitmap_info.offset) /* bitmap disabled, nothing to do */
1672 1673
		return 0;

1674
	BUG_ON(file && mddev->bitmap_info.offset);
1675

1676
	bitmap = kzalloc(sizeof(*bitmap), GFP_KERNEL);
1677 1678 1679 1680
	if (!bitmap)
		return -ENOMEM;

	spin_lock_init(&bitmap->lock);
1681 1682
	atomic_set(&bitmap->pending_writes, 0);
	init_waitqueue_head(&bitmap->write_wait);
1683
	init_waitqueue_head(&bitmap->overflow_wait);
1684
	init_waitqueue_head(&bitmap->behind_wait);
1685

1686 1687
	bitmap->mddev = mddev;

1688
	bm = sysfs_get_dirent(mddev->kobj.sd, NULL, "bitmap");
1689
	if (bm) {
1690
		bitmap->sysfs_can_clear = sysfs_get_dirent(bm, NULL, "can_clear");
1691 1692 1693 1694
		sysfs_put(bm);
	} else
		bitmap->sysfs_can_clear = NULL;

1695
	bitmap->file = file;
1696 1697
	if (file) {
		get_file(file);
1698 1699 1700 1701
		/* As future accesses to this file will use bmap,
		 * and bypass the page cache, we must sync the file
		 * first.
		 */
1702
		vfs_fsync(file, 1);
1703
	}
1704
	/* read superblock from bitmap file (this sets mddev->bitmap_info.chunksize) */
1705 1706 1707 1708 1709 1710 1711 1712 1713 1714
	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;
	}
1715
	if (err)
1716
		goto error;
1717

1718
	bitmap->daemon_lastrun = jiffies;
1719
	bitmap->chunkshift = ffz(~mddev->bitmap_info.chunksize);
1720 1721

	/* now that chunksize and chunkshift are set, we can use these macros */
1722 1723
 	chunks = (blocks + CHUNK_BLOCK_RATIO(bitmap) - 1) >>
			CHUNK_BLOCK_SHIFT(bitmap);
1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734
 	pages = (chunks + PAGE_COUNTER_RATIO - 1) / PAGE_COUNTER_RATIO;

	BUG_ON(!pages);

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

	bitmap->syncchunk = ~0UL;

1735
#ifdef INJECT_FATAL_FAULT_1
1736 1737
	bitmap->bp = NULL;
#else
1738
	bitmap->bp = kzalloc(pages * sizeof(*bitmap->bp), GFP_KERNEL);
1739
#endif
1740
	err = -ENOMEM;
1741
	if (!bitmap->bp)
1742
		goto error;
1743 1744 1745

	/* now that we have some pages available, initialize the in-memory
	 * bitmap from the on-disk bitmap */
1746 1747 1748 1749 1750 1751
	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;
	err = bitmap_init_from_disk(bitmap, start);
1752

1753
	if (err)
1754
		goto error;
1755 1756 1757 1758

	printk(KERN_INFO "created bitmap (%lu pages) for device %s\n",
		pages, bmname(bitmap));

1759 1760
	mddev->bitmap = bitmap;

1761
	mddev->thread->timeout = mddev->bitmap_info.daemon_sleep;
1762
	md_wakeup_thread(mddev->thread);
1763

1764 1765 1766
	bitmap_update_sb(bitmap);

	return (bitmap->flags & BITMAP_WRITE_ERROR) ? -EIO : 0;
1767 1768 1769 1770

 error:
	bitmap_free(bitmap);
	return err;
1771 1772
}

1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832
static ssize_t
location_show(mddev_t *mddev, char *page)
{
	ssize_t len;
	if (mddev->bitmap_info.file) {
		len = sprintf(page, "file");
	} else if (mddev->bitmap_info.offset) {
		len = sprintf(page, "%+lld", (long long)mddev->bitmap_info.offset);
	} else
		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;
1833 1834
			if (mddev->bitmap_info.external == 0 &&
			    mddev->major_version == 0 &&
1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962
			    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;
	
	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);

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
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);

2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036
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);

2037 2038 2039 2040 2041
static struct attribute *md_bitmap_attrs[] = {
	&bitmap_location.attr,
	&bitmap_timeout.attr,
	&bitmap_backlog.attr,
	&bitmap_chunksize.attr,
2042 2043
	&bitmap_metadata.attr,
	&bitmap_can_clear.attr,
2044
	&max_backlog_used.attr,
2045 2046 2047 2048 2049 2050 2051 2052
	NULL
};
struct attribute_group md_bitmap_group = {
	.name = "bitmap",
	.attrs = md_bitmap_attrs,
};


2053 2054 2055 2056 2057 2058 2059
/* the bitmap API -- for raid personalities */
EXPORT_SYMBOL(bitmap_startwrite);
EXPORT_SYMBOL(bitmap_endwrite);
EXPORT_SYMBOL(bitmap_start_sync);
EXPORT_SYMBOL(bitmap_end_sync);
EXPORT_SYMBOL(bitmap_unplug);
EXPORT_SYMBOL(bitmap_close_sync);
N
NeilBrown 已提交
2060
EXPORT_SYMBOL(bitmap_cond_end_sync);