journal.c

linux 内核源代码

static int journal_get_superblock(journal_t *journal)
{
	struct buffer_head *bh;
	journal_superblock_t *sb;
	int err = -EIO;

	bh = journal->j_sb_buffer;

	J_ASSERT(bh != NULL);
	if (!buffer_uptodate(bh)) {
		ll_rw_block(READ, 1, &bh);
		wait_on_buffer(bh);
		if (!buffer_uptodate(bh)) {
			printk (KERN_ERR
				"JBD: IO error reading journal superblock\n");
			goto out;
		}
	}

	sb = journal->j_superblock;

	err = -EINVAL;

	if (sb->s_header.h_magic != cpu_to_be32(JFS_MAGIC_NUMBER) ||
	    sb->s_blocksize != cpu_to_be32(journal->j_blocksize)) {
		printk(KERN_WARNING "JBD: no valid journal superblock found\n");
		goto out;
	}

	switch(be32_to_cpu(sb->s_header.h_blocktype)) {
	case JFS_SUPERBLOCK_V1:
		journal->j_format_version = 1;
		break;
	case JFS_SUPERBLOCK_V2:
		journal->j_format_version = 2;
		break;
	default:
		printk(KERN_WARNING "JBD: unrecognised superblock format ID\n");
		goto out;
	}

	if (be32_to_cpu(sb->s_maxlen) < journal->j_maxlen)
		journal->j_maxlen = be32_to_cpu(sb->s_maxlen);
	else if (be32_to_cpu(sb->s_maxlen) > journal->j_maxlen) {
		printk (KERN_WARNING "JBD: journal file too short\n");
		goto out;
	}

	return 0;

out:
	journal_fail_superblock(journal);
	return err;
}

/*
 * Load the on-disk journal superblock and read the key fields into the
 * journal_t.
 */

static int load_superblock(journal_t *journal)
{
	int err;
	journal_superblock_t *sb;

	err = journal_get_superblock(journal);
	if (err)
		return err;

	sb = journal->j_superblock;

	journal->j_tail_sequence = be32_to_cpu(sb->s_sequence);
	journal->j_tail = be32_to_cpu(sb->s_start);
	journal->j_first = be32_to_cpu(sb->s_first);
	journal->j_last = be32_to_cpu(sb->s_maxlen);
	journal->j_errno = be32_to_cpu(sb->s_errno);

	return 0;
}


/**
 * int journal_load() - Read journal from disk.
 * @journal: Journal to act on.
 *
 * Given a journal_t structure which tells us which disk blocks contain
 * a journal, read the journal from disk to initialise the in-memory
 * structures.
 */
int journal_load(journal_t *journal)
{
	int err;
	journal_superblock_t *sb;

	err = load_superblock(journal);
	if (err)
		return err;

	sb = journal->j_superblock;
	/* If this is a V2 superblock, then we have to check the
	 * features flags on it. */

	if (journal->j_format_version >= 2) {
		if ((sb->s_feature_ro_compat &
		     ~cpu_to_be32(JFS_KNOWN_ROCOMPAT_FEATURES)) ||
		    (sb->s_feature_incompat &
		     ~cpu_to_be32(JFS_KNOWN_INCOMPAT_FEATURES))) {
			printk (KERN_WARNING
				"JBD: Unrecognised features on journal\n");
			return -EINVAL;
		}
	}

	/* Let the recovery code check whether it needs to recover any
	 * data from the journal. */
	if (journal_recover(journal))
		goto recovery_error;

	/* OK, we've finished with the dynamic journal bits:
	 * reinitialise the dynamic contents of the superblock in memory
	 * and reset them on disk. */
	if (journal_reset(journal))
		goto recovery_error;

	journal->j_flags &= ~JFS_ABORT;
	journal->j_flags |= JFS_LOADED;
	return 0;

recovery_error:
	printk (KERN_WARNING "JBD: recovery failed\n");
	return -EIO;
}

/**
 * void journal_destroy() - Release a journal_t structure.
 * @journal: Journal to act on.
 *
 * Release a journal_t structure once it is no longer in use by the
 * journaled object.
 */
void journal_destroy(journal_t *journal)
{
	/* Wait for the commit thread to wake up and die. */
	journal_kill_thread(journal);

	/* Force a final log commit */
	if (journal->j_running_transaction)
		journal_commit_transaction(journal);

	/* Force any old transactions to disk */

	/* Totally anal locking here... */
	spin_lock(&journal->j_list_lock);
	while (journal->j_checkpoint_transactions != NULL) {
		spin_unlock(&journal->j_list_lock);
		log_do_checkpoint(journal);
		spin_lock(&journal->j_list_lock);
	}

	J_ASSERT(journal->j_running_transaction == NULL);
	J_ASSERT(journal->j_committing_transaction == NULL);
	J_ASSERT(journal->j_checkpoint_transactions == NULL);
	spin_unlock(&journal->j_list_lock);

	/* We can now mark the journal as empty. */
	journal->j_tail = 0;
	journal->j_tail_sequence = ++journal->j_transaction_sequence;
	if (journal->j_sb_buffer) {
		journal_update_superblock(journal, 1);
		brelse(journal->j_sb_buffer);
	}

	if (journal->j_inode)
		iput(journal->j_inode);
	if (journal->j_revoke)
		journal_destroy_revoke(journal);
	kfree(journal->j_wbuf);
	kfree(journal);
}


/**
 *int journal_check_used_features () - Check if features specified are used.
 * @journal: Journal to check.
 * @compat: bitmask of compatible features
 * @ro: bitmask of features that force read-only mount
 * @incompat: bitmask of incompatible features
 *
 * Check whether the journal uses all of a given set of
 * features.  Return true (non-zero) if it does.
 **/

int journal_check_used_features (journal_t *journal, unsigned long compat,
				 unsigned long ro, unsigned long incompat)
{
	journal_superblock_t *sb;

	if (!compat && !ro && !incompat)
		return 1;
	if (journal->j_format_version == 1)
		return 0;

	sb = journal->j_superblock;

	if (((be32_to_cpu(sb->s_feature_compat) & compat) == compat) &&
	    ((be32_to_cpu(sb->s_feature_ro_compat) & ro) == ro) &&
	    ((be32_to_cpu(sb->s_feature_incompat) & incompat) == incompat))
		return 1;

	return 0;
}

/**
 * int journal_check_available_features() - Check feature set in journalling layer
 * @journal: Journal to check.
 * @compat: bitmask of compatible features
 * @ro: bitmask of features that force read-only mount
 * @incompat: bitmask of incompatible features
 *
 * Check whether the journaling code supports the use of
 * all of a given set of features on this journal.  Return true
 * (non-zero) if it can. */

int journal_check_available_features (journal_t *journal, unsigned long compat,
				      unsigned long ro, unsigned long incompat)
{
	journal_superblock_t *sb;

	if (!compat && !ro && !incompat)
		return 1;

	sb = journal->j_superblock;

	/* We can support any known requested features iff the
	 * superblock is in version 2.  Otherwise we fail to support any
	 * extended sb features. */

	if (journal->j_format_version != 2)
		return 0;

	if ((compat   & JFS_KNOWN_COMPAT_FEATURES) == compat &&
	    (ro       & JFS_KNOWN_ROCOMPAT_FEATURES) == ro &&
	    (incompat & JFS_KNOWN_INCOMPAT_FEATURES) == incompat)
		return 1;

	return 0;
}

/**
 * int journal_set_features () - Mark a given journal feature in the superblock
 * @journal: Journal to act on.
 * @compat: bitmask of compatible features
 * @ro: bitmask of features that force read-only mount
 * @incompat: bitmask of incompatible features
 *
 * Mark a given journal feature as present on the
 * superblock.  Returns true if the requested features could be set.
 *
 */

int journal_set_features (journal_t *journal, unsigned long compat,
			  unsigned long ro, unsigned long incompat)
{
	journal_superblock_t *sb;

	if (journal_check_used_features(journal, compat, ro, incompat))
		return 1;

	if (!journal_check_available_features(journal, compat, ro, incompat))
		return 0;

	jbd_debug(1, "Setting new features 0x%lx/0x%lx/0x%lx\n",
		  compat, ro, incompat);

	sb = journal->j_superblock;

	sb->s_feature_compat    |= cpu_to_be32(compat);
	sb->s_feature_ro_compat |= cpu_to_be32(ro);
	sb->s_feature_incompat  |= cpu_to_be32(incompat);

	return 1;
}


/**
 * int journal_update_format () - Update on-disk journal structure.
 * @journal: Journal to act on.
 *
 * Given an initialised but unloaded journal struct, poke about in the
 * on-disk structure to update it to the most recent supported version.
 */
int journal_update_format (journal_t *journal)
{
	journal_superblock_t *sb;
	int err;

	err = journal_get_superblock(journal);
	if (err)
		return err;

	sb = journal->j_superblock;

	switch (be32_to_cpu(sb->s_header.h_blocktype)) {
	case JFS_SUPERBLOCK_V2:
		return 0;
	case JFS_SUPERBLOCK_V1:
		return journal_convert_superblock_v1(journal, sb);
	default:
		break;
	}
	return -EINVAL;
}

static int journal_convert_superblock_v1(journal_t *journal,
					 journal_superblock_t *sb)
{
	int offset, blocksize;
	struct buffer_head *bh;

	printk(KERN_WARNING
		"JBD: Converting superblock from version 1 to 2.\n");

	/* Pre-initialise new fields to zero */
	offset = ((char *) &(sb->s_feature_compat)) - ((char *) sb);
	blocksize = be32_to_cpu(sb->s_blocksize);
	memset(&sb->s_feature_compat, 0, blocksize-offset);

	sb->s_nr_users = cpu_to_be32(1);
	sb->s_header.h_blocktype = cpu_to_be32(JFS_SUPERBLOCK_V2);
	journal->j_format_version = 2;

	bh = journal->j_sb_buffer;
	BUFFER_TRACE(bh, "marking dirty");
	mark_buffer_dirty(bh);
	sync_dirty_buffer(bh);
	return 0;
}


/**
 * int journal_flush () - Flush journal
 * @journal: Journal to act on.
 *
 * Flush all data for a given journal to disk and empty the journal.
 * Filesystems can use this when remounting readonly to ensure that
 * recovery does not need to happen on remount.
 */

int journal_flush(journal_t *journal)
{
	int err = 0;
	transaction_t *transaction = NULL;
	unsigned long old_tail;

	spin_lock(&journal->j_state_lock);

	/* Force everything buffered to the log... */
	if (journal->j_running_transaction) {
		transaction = journal->j_running_transaction;
		__log_start_commit(journal, transaction->t_tid);
	} else if (journal->j_committing_transaction)
		transaction = journal->j_committing_transaction;

	/* Wait for the log commit to complete... */
	if (transaction) {
		tid_t tid = transaction->t_tid;

		spin_unlock(&journal->j_state_lock);
		log_wait_commit(journal, tid);
	} else {
		spin_unlock(&journal->j_state_lock);
	}

	/* ...and flush everything in the log out to disk. */
	spin_lock(&journal->j_list_lock);
	while (!err && journal->j_checkpoint_transactions != NULL) {
		spin_unlock(&journal->j_list_lock);
		err = log_do_checkpoint(journal);
		spin_lock(&journal->j_list_lock);
	}
	spin_unlock(&journal->j_list_lock);
	cleanup_journal_tail(journal);

	/* Finally, mark the journal as really needing no recovery.
	 * This sets s_start==0 in the underlying superblock, which is
	 * the magic code for a fully-recovered superblock.  Any future
	 * commits of data to the journal will restore the current
	 * s_start value. */
	spin_lock(&journal->j_state_lock);
	old_tail = journal->j_tail;
	journal->j_tail = 0;
	spin_unlock(&journal->j_state_lock);
	journal_update_superblock(journal, 1);
	spin_lock(&journal->j_state_lock);
	journal->j_tail = old_tail;

	J_ASSERT(!journal->j_running_transaction);
	J_ASSERT(!journal->j_committing_transaction);
	J_ASSERT(!journal->j_checkpoint_transactions);
	J_ASSERT(journal->j_head == journal->j_tail);
	J_ASSERT(journal->j_tail_sequence == journal->j_transaction_sequence);
	spin_unlock(&journal->j_state_lock);
	return err;
}

/**
 * int journal_wipe() - Wipe journal contents
 * @journal: Journal to act on.
 * @write: flag (see below)
 *
 * Wipe out all of the contents of a journal, safely.  This will produce
 * a warning if the journal contains any valid recovery information.
 * Must be called between journal_init_*() and journal_load().
 *
 * If 'write' is non-zero, then we wipe out the journal on disk; otherwise
 * we merely suppress recovery.
 */

int journal_wipe(journal_t *journal, int write)
{
	journal_superblock_t *sb;
	int err = 0;

	J_ASSERT (!(journal->j_flags & JFS_LOADED));

	err = load_superblock(journal);
	if (err)
		return err;

	sb = journal->j_superblock;

	if (!journal->j_tail)
		goto no_recovery;

	printk (KERN_WARNING "JBD: %s recovery information on journal\n",
		write ? "Clearing" : "Ignoring");

	err = journal_skip_recovery(journal);
	if (write)
		journal_update_superblock(journal, 1);

 no_recovery:
	return err;
}

/*
 * journal_dev_name: format a character string to describe on what
 * device this journal is present.
 */

static const char *journal_dev_name(journal_t *journal, char *buffer)
{
	struct block_device *bdev;

	if (journal->j_inode)
		bdev = journal->j_inode->i_sb->s_bdev;
	else
		bdev = journal->j_dev;

	return bdevname(bdev, buffer);
}

/*
 * Journal abort has very specific semantics, which we describe
 * for journal abort.
 *
 * Two internal function, which provide abort to te jbd layer
 * itself are here.
 */

/*
 * Quick version for internal journal use (doesn't lock the journal).
 * Aborts hard --- we mark the abort as occurred, but do _nothing_ else,
 * and don't attempt to make any other journal updates.
 */
void __journal_abort_hard(journal_t *journal)
{
	transaction_t *transaction;
	char b[BDEVNAME_SIZE];

	if (journal->j_flags & JFS_ABORT)
		return;

	printk(KERN_ERR "Aborting journal on device %s.\n",
		journal_dev_name(journal, b));

	spin_lock(&journal->j_state_lock);
	journal->j_flags |= JFS_ABORT;
	transaction = journal->j_running_transaction;
	if (transaction)
		__log_start_commit(journal, transaction->t_tid);
	spin_unlock(&journal->j_state_lock);