md.c

来自「linux 内核源代码」· C语言 代码 · 共 2,595 行 · 第 1/5 页

		MD_BUG();
		return -EINVAL;
	}
	if (rdev->sb_loaded)
		return 0;


	if (!sync_page_io(rdev->bdev, rdev->sb_offset<<1, size, rdev->sb_page, READ))
		goto fail;
	rdev->sb_loaded = 1;
	return 0;

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

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

		return 1;

	return 0;
}


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

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

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

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

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

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

abort:
	kfree(tmp1);
	kfree(tmp2);
	return ret;
}


static u32 md_csum_fold(u32 csum)
{
	csum = (csum & 0xffff) + (csum >> 16);
	return (csum & 0xffff) + (csum >> 16);
}

static unsigned int calc_sb_csum(mdp_super_t * sb)
{
	u64 newcsum = 0;
	u32 *sb32 = (u32*)sb;
	int i;
	unsigned int disk_csum, csum;

	disk_csum = sb->sb_csum;
	sb->sb_csum = 0;

	for (i = 0; i < MD_SB_BYTES/4 ; i++)
		newcsum += sb32[i];
	csum = (newcsum & 0xffffffff) + (newcsum>>32);


#ifdef CONFIG_ALPHA
	/* This used to use csum_partial, which was wrong for several
	 * reasons including that different results are returned on
	 * different architectures.  It isn't critical that we get exactly
	 * the same return value as before (we always csum_fold before
	 * testing, and that removes any differences).  However as we
	 * know that csum_partial always returned a 16bit value on
	 * alphas, do a fold to maximise conformity to previous behaviour.
	 */
	sb->sb_csum = md_csum_fold(disk_csum);
#else
	sb->sb_csum = disk_csum;
#endif
	return csum;
}


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

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

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

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

	ret = read_disk_sb(rdev, MD_SB_BYTES);
	if (ret) return ret;

	ret = -EINVAL;

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

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

	if (sb->major_version != 0 ||
	    sb->minor_version < 90 ||
	    sb->minor_version > 91) {
		printk(KERN_WARNING "Bad version number %d.%d on %s\n",
			sb->major_version, sb->minor_version,
			b);
		goto abort;
	}

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

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

	rdev->preferred_minor = sb->md_minor;
	rdev->data_offset = 0;
	rdev->sb_size = MD_SB_BYTES;

	if (sb->state & (1<<MD_SB_BITMAP_PRESENT)) {
		if (sb->level != 1 && sb->level != 4
		    && sb->level != 5 && sb->level != 6
		    && sb->level != 10) {
			/* FIXME use a better test */
			printk(KERN_WARNING
			       "md: bitmaps not supported for this level.\n");
			goto abort;
		}
	}

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

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

	if (rdev->size < sb->size && sb->level > 1)
		/* "this cannot possibly happen" ... */
		ret = -EINVAL;

 abort:
	return ret;
}

/*
 * validate_super for 0.90.0
 */
static int super_90_validate(mddev_t *mddev, mdk_rdev_t *rdev)
{
	mdp_disk_t *desc;
	mdp_super_t *sb = (mdp_super_t *)page_address(rdev->sb_page);
	__u64 ev1 = md_event(sb);

	rdev->raid_disk = -1;
	rdev->flags = 0;
	if (mddev->raid_disks == 0) {
		mddev->major_version = 0;
		mddev->minor_version = sb->minor_version;
		mddev->patch_version = sb->patch_version;
		mddev->persistent = ! sb->not_persistent;
		mddev->chunk_size = sb->chunk_size;
		mddev->ctime = sb->ctime;
		mddev->utime = sb->utime;
		mddev->level = sb->level;
		mddev->clevel[0] = 0;
		mddev->layout = sb->layout;
		mddev->raid_disks = sb->raid_disks;
		mddev->size = sb->size;
		mddev->events = ev1;
		mddev->bitmap_offset = 0;
		mddev->default_bitmap_offset = MD_SB_BYTES >> 9;

		if (mddev->minor_version >= 91) {
			mddev->reshape_position = sb->reshape_position;
			mddev->delta_disks = sb->delta_disks;
			mddev->new_level = sb->new_level;
			mddev->new_layout = sb->new_layout;
			mddev->new_chunk = sb->new_chunk;
		} else {
			mddev->reshape_position = MaxSector;
			mddev->delta_disks = 0;
			mddev->new_level = mddev->level;
			mddev->new_layout = mddev->layout;
			mddev->new_chunk = mddev->chunk_size;
		}

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

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

		mddev->max_disks = MD_SB_DISKS;

		if (sb->state & (1<<MD_SB_BITMAP_PRESENT) &&
		    mddev->bitmap_file == NULL)
			mddev->bitmap_offset = mddev->default_bitmap_offset;

	} else if (mddev->pers == NULL) {
		/* Insist on good event counter while assembling */
		++ev1;
		if (ev1 < mddev->events) 
			return -EINVAL;
	} else if (mddev->bitmap) {
		/* if adding to array with a bitmap, then we can accept an
		 * older device ... but not too old.
		 */
		if (ev1 < mddev->bitmap->events_cleared)
			return 0;
	} else {
		if (ev1 < mddev->events)
			/* just a hot-add of a new device, leave raid_disk at -1 */
			return 0;
	}

	if (mddev->level != LEVEL_MULTIPATH) {
		desc = sb->disks + rdev->desc_nr;

		if (desc->state & (1<<MD_DISK_FAULTY))
			set_bit(Faulty, &rdev->flags);
		else if (desc->state & (1<<MD_DISK_SYNC) /* &&
			    desc->raid_disk < mddev->raid_disks */) {
			set_bit(In_sync, &rdev->flags);
			rdev->raid_disk = desc->raid_disk;
		}
		if (desc->state & (1<<MD_DISK_WRITEMOSTLY))
			set_bit(WriteMostly, &rdev->flags);
	} else /* MULTIPATH are always insync */
		set_bit(In_sync, &rdev->flags);
	return 0;
}

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


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

	rdev->sb_size = MD_SB_BYTES;

	sb = (mdp_super_t*)page_address(rdev->sb_page);

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

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

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

	if (mddev->reshape_position == MaxSector)
		sb->minor_version = 90;
	else {
		sb->minor_version = 91;
		sb->reshape_position = mddev->reshape_position;
		sb->new_level = mddev->new_level;
		sb->delta_disks = mddev->delta_disks;
		sb->new_layout = mddev->new_layout;
		sb->new_chunk = mddev->new_chunk;
	}
	mddev->minor_version = sb->minor_version;
	if (mddev->in_sync)
	{
		sb->recovery_cp = mddev->recovery_cp;
		sb->cp_events_hi = (mddev->events>>32);
		sb->cp_events_lo = (u32)mddev->events;
		if (mddev->recovery_cp == MaxSector)
			sb->state = (1<< MD_SB_CLEAN);
	} else
		sb->recovery_cp = 0;

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

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

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

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

/*
 * version 1 superblock
 */

static __le32 calc_sb_1_csum(struct mdp_superblock_1 * sb)
{
	__le32 disk_csum;
	u32 csum;
	unsigned long long newcsum;
	int size = 256 + le32_to_cpu(sb->max_dev)*2;
	__le32 *isuper = (__le32*)sb;
	int i;

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

	if (size == 2)
		newcsum += le16_to_cpu(*(__le16*) isuper);

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

static int super_1_load(mdk_rdev_t *rdev, mdk_rdev_t *refdev, int minor_version)
{
	struct mdp_superblock_1 *sb;
	int ret;
	sector_t sb_offset;
	char b[BDEVNAME_SIZE], b2[BDEVNAME_SIZE];
	int bmask;

	/*
	 * Calculate the position of the superblock.
	 * It is always aligned to a 4K boundary and
	 * depeding on minor_version, it can be:
	 * 0: At least 8K, but less than 12K, from end of device
	 * 1: At start of device
	 * 2: 4K from start of device.
	 */