/*
 * Called when an inode is released. Note that this is different
 * from ext3_file_open: open gets called at every open, but release
 * gets called only when /all/ the files are closed.
 */
static int ext3_release_file (struct inode * inode, struct file * filp)
{
	/* if we are the last writer on the inode, drop the block reservation */
	if ((filp->f_mode & FMODE_WRITE) &&
			(atomic_read(&inode->i_writecount) == 1))
	{
		mutex_lock(&EXT3_I(inode)->truncate_mutex);
		ext3_discard_reservation(inode);
		mutex_unlock(&EXT3_I(inode)->truncate_mutex);
	}
	if (is_dx(inode) && filp->private_data)
		ext3_htree_free_dir_info(filp->private_data);

	return 0;
}

/*
 * Restart the transaction associated with *handle.  This does a commit,
 * so before we call here everything must be consistently dirtied against
 * this transaction.
 */
static int truncate_restart_transaction(handle_t *handle, struct inode *inode)
{
	int ret;

	jbd_debug(2, "restarting handle %p\n", handle);
	/*
	 * Drop truncate_mutex to avoid deadlock with ext3_get_blocks_handle
	 * At this moment, get_block can be called only for blocks inside
	 * i_size since page cache has been already dropped and writes are
	 * blocked by i_mutex. So we can safely drop the truncate_mutex.
	 */
	mutex_unlock(&EXT3_I(inode)->truncate_mutex);
	ret = ext3_journal_restart(handle, blocks_for_truncate(inode));
	mutex_lock(&EXT3_I(inode)->truncate_mutex);
	return ret;
}

int ext3_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
{
	struct inode *inode = file->f_mapping->host;
	struct ext3_inode_info *ei = EXT3_I(inode);
	journal_t *journal = EXT3_SB(inode->i_sb)->s_journal;
	int ret, needs_barrier = 0;
	tid_t commit_tid;

	trace_ext3_sync_file_enter(file, datasync);

	if (inode->i_sb->s_flags & MS_RDONLY)
		return 0;

	ret = filemap_write_and_wait_range(inode->i_mapping, start, end);
	if (ret)
		goto out;

	J_ASSERT(ext3_journal_current_handle() == NULL);

	/*
	 * data=writeback,ordered:
	 *  The caller's filemap_fdatawrite()/wait will sync the data.
	 *  Metadata is in the journal, we wait for a proper transaction
	 *  to commit here.
	 *
	 * data=journal:
	 *  filemap_fdatawrite won't do anything (the buffers are clean).
	 *  ext3_force_commit will write the file data into the journal and
	 *  will wait on that.
	 *  filemap_fdatawait() will encounter a ton of newly-dirtied pages
	 *  (they were dirtied by commit).  But that's OK - the blocks are
	 *  safe in-journal, which is all fsync() needs to ensure.
	 */
	if (ext3_should_journal_data(inode)) {
		ret = ext3_force_commit(inode->i_sb);
		goto out;
	}

	if (datasync)
		commit_tid = atomic_read(&ei->i_datasync_tid);
	else
		commit_tid = atomic_read(&ei->i_sync_tid);

	if (test_opt(inode->i_sb, BARRIER) &&
	    !journal_trans_will_send_data_barrier(journal, commit_tid))
		needs_barrier = 1;
	log_start_commit(journal, commit_tid);
	ret = log_wait_commit(journal, commit_tid);

	/*
	 * In case we didn't commit a transaction, we have to flush
	 * disk caches manually so that data really is on persistent
	 * storage
	 */
	if (needs_barrier)
		blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL, NULL);
out:
	trace_ext3_sync_file_exit(inode, ret);
	return ret;
}

static int find_group_other(struct super_block *sb, struct inode *parent)
{
	int parent_group = EXT3_I(parent)->i_block_group;
	int ngroups = EXT3_SB(sb)->s_groups_count;
	struct ext3_group_desc *desc;
	struct buffer_head *bh;
	int group, i;

	/*
	 * Try to place the inode in its parent directory
	 */
	group = parent_group;
	desc = ext3_get_group_desc (sb, group, &bh);
	if (desc && le16_to_cpu(desc->bg_free_inodes_count) &&
			le16_to_cpu(desc->bg_free_blocks_count))
		return group;

	/*
	 * We're going to place this inode in a different blockgroup from its
	 * parent.  We want to cause files in a common directory to all land in
	 * the same blockgroup.  But we want files which are in a different
	 * directory which shares a blockgroup with our parent to land in a
	 * different blockgroup.
	 *
	 * So add our directory's i_ino into the starting point for the hash.
	 */
	group = (group + parent->i_ino) % ngroups;

	/*
	 * Use a quadratic hash to find a group with a free inode and some free
	 * blocks.
	 */
	for (i = 1; i < ngroups; i <<= 1) {
		group += i;
		if (group >= ngroups)
			group -= ngroups;
		desc = ext3_get_group_desc (sb, group, &bh);
		if (desc && le16_to_cpu(desc->bg_free_inodes_count) &&
				le16_to_cpu(desc->bg_free_blocks_count))
			return group;
	}

	/*
	 * That failed: try linear search for a free inode, even if that group
	 * has no free blocks.
	 */
	group = parent_group;
	for (i = 0; i < ngroups; i++) {
		if (++group >= ngroups)
			group = 0;
		desc = ext3_get_group_desc (sb, group, &bh);
		if (desc && le16_to_cpu(desc->bg_free_inodes_count))
			return group;
	}

	return -1;
}

/*
 * Called when an inode is released. Note that this is different
 * from ext3_file_open: open gets called at every open, but release
 * gets called only when /all/ the files are closed.
 */
static int ext3_release_file (struct inode * inode, struct file * filp)
{
	if (ext3_test_inode_state(inode, EXT3_STATE_FLUSH_ON_CLOSE)) {
		filemap_flush(inode->i_mapping);
		ext3_clear_inode_state(inode, EXT3_STATE_FLUSH_ON_CLOSE);
	}
	/* if we are the last writer on the inode, drop the block reservation */
	if ((filp->f_mode & FMODE_WRITE) &&
			(atomic_read(&inode->i_writecount) == 1))
	{
		mutex_lock(&EXT3_I(inode)->truncate_mutex);
		ext3_discard_reservation(inode);
		mutex_unlock(&EXT3_I(inode)->truncate_mutex);
	}
	if (is_dx(inode) && filp->private_data)
		ext3_htree_free_dir_info(filp->private_data);

	return 0;
}

static inline int mlowerfs_ext3_should_journal_data(struct inode *inode)
{
	if (!S_ISREG(inode->i_mode))
		return 1;
	if (test_opt(inode->i_sb, DATA_FLAGS) == EXT3_MOUNT_JOURNAL_DATA)
		return 1;
	if (EXT3_I(inode)->i_flags & EXT3_JOURNAL_DATA_FL)
		return 1;
	return 0;
}

/*
 * Inode operation get_posix_acl().
 *
 * inode->i_sem: don't care
 * BKL: held
 */
struct posix_acl *
ext3_get_acl(struct inode *inode, int type)
{
	const size_t max_size = ext3_acl_size(EXT3_ACL_MAX_ENTRIES);
	struct ext3_inode_info *ei = EXT3_I(inode);
	int name_index;
	char *value;
	struct posix_acl *acl;
	int retval;

	if (!IS_POSIXACL(inode))
		return 0;

	switch(type) {
		case ACL_TYPE_ACCESS:
			if (ei->i_acl != EXT3_ACL_NOT_CACHED)
				return posix_acl_dup(ei->i_acl);
			name_index = EXT3_XATTR_INDEX_POSIX_ACL_ACCESS;
			break;

		case ACL_TYPE_DEFAULT:
			if (ei->i_default_acl != EXT3_ACL_NOT_CACHED)
				return posix_acl_dup(ei->i_default_acl);
			name_index = EXT3_XATTR_INDEX_POSIX_ACL_DEFAULT;
			break;

		default:
			return ERR_PTR(-EINVAL);
	}
	value = kmalloc(max_size, GFP_KERNEL);
	if (!value)
		return ERR_PTR(-ENOMEM);

	retval = ext3_xattr_get(inode, name_index, "", value, max_size);
	acl = ERR_PTR(retval);
	if (retval > 0)
		acl = ext3_acl_from_disk(value, retval);
	else if (retval == -ENODATA || retval == -ENOSYS)
		acl = NULL;
	kfree(value);

	if (!IS_ERR(acl)) {
		switch(type) {
			case ACL_TYPE_ACCESS:
				ei->i_acl = posix_acl_dup(acl);
				break;

			case ACL_TYPE_DEFAULT:
				ei->i_default_acl = posix_acl_dup(acl);
				break;
		}
	}
	return acl;
}

/*
 * There are two policies for allocating an inode.  If the new inode is
 * a directory, then a forward search is made for a block group with both
 * free space and a low directory-to-inode ratio; if that fails, then of
 * the groups with above-average free space, that group with the fewest
 * directories already is chosen.
 *
 * For other inodes, search forward from the parent directory's block
 * group to find a free inode.
 */
struct inode *ext3_new_inode(handle_t *handle, struct inode * dir, int mode)
{
	struct super_block *sb;
	struct buffer_head *bitmap_bh = NULL;
	struct buffer_head *bh2;
	int group;
	unsigned long ino = 0;
	struct inode * inode;
	struct ext3_group_desc * gdp = NULL;
	struct ext3_super_block * es;
	struct ext3_inode_info *ei;
	struct ext3_sb_info *sbi;
	int err = 0;
	struct inode *ret;
	int i;

	/* Cannot create files in a deleted directory */
	if (!dir || !dir->i_nlink)
		return ERR_PTR(-EPERM);

	sb = dir->i_sb;
	inode = new_inode(sb);
	if (!inode)
		return ERR_PTR(-ENOMEM);
	ei = EXT3_I(inode);

	sbi = EXT3_SB(sb);
	es = sbi->s_es;
	if (S_ISDIR(mode)) {
		if (test_opt (sb, OLDALLOC))
			group = find_group_dir(sb, dir);
		else
			group = find_group_orlov(sb, dir);
	} else 
		group = find_group_other(sb, dir);

	err = -ENOSPC;
	if (group == -1)
		goto out;

	for (i = 0; i < sbi->s_groups_count; i++) {
		err = -EIO;

		gdp = ext3_get_group_desc(sb, group, &bh2);
		if (!gdp)
			goto fail;

		brelse(bitmap_bh);
		bitmap_bh = read_inode_bitmap(sb, group);
		if (!bitmap_bh)
			goto fail;

		ino = 0;

repeat_in_this_group:
		ino = ext3_find_next_zero_bit((unsigned long *)
				bitmap_bh->b_data, EXT3_INODES_PER_GROUP(sb), ino);
		if (ino < EXT3_INODES_PER_GROUP(sb)) {

			BUFFER_TRACE(bitmap_bh, "get_write_access");
			err = ext3_journal_get_write_access(handle, bitmap_bh);
			if (err)
				goto fail;

			if (!ext3_set_bit_atomic(sb_bgl_lock(sbi, group),
						ino, bitmap_bh->b_data)) {
				/* we won it */
				BUFFER_TRACE(bitmap_bh,
					"call ext3_journal_dirty_metadata");
				err = ext3_journal_dirty_metadata(handle,
								bitmap_bh);
				if (err)
					goto fail;
				goto got;
			}
			/* we lost it */
			journal_release_buffer(handle, bitmap_bh);

			if (++ino < EXT3_INODES_PER_GROUP(sb))
				goto repeat_in_this_group;
		}

		/*
		 * This case is possible in concurrent environment.  It is very
		 * rare.  We cannot repeat the find_group_xxx() call because
		 * that will simply return the same blockgroup, because the
		 * group descriptor metadata has not yet been updated.
		 * So we just go onto the next blockgroup.
		 */
		if (++group == sbi->s_groups_count)
//.........这里部分代码省略.........

static int find_group_orlov(struct super_block *sb, struct inode *parent)
{
	int parent_group = EXT3_I(parent)->i_block_group;
	struct ext3_sb_info *sbi = EXT3_SB(sb);
	struct ext3_super_block *es = sbi->s_es;
	int ngroups = sbi->s_groups_count;
	int inodes_per_group = EXT3_INODES_PER_GROUP(sb);
	int freei, avefreei;
	int freeb, avefreeb;
	int blocks_per_dir, ndirs;
	int max_debt, max_dirs, min_blocks, min_inodes;
	int group = -1, i;
	struct ext3_group_desc *desc;
	struct buffer_head *bh;

	freei = percpu_counter_read_positive(&sbi->s_freeinodes_counter);
	avefreei = freei / ngroups;
	freeb = percpu_counter_read_positive(&sbi->s_freeblocks_counter);
	avefreeb = freeb / ngroups;
	ndirs = percpu_counter_read_positive(&sbi->s_dirs_counter);

	if ((parent == sb->s_root->d_inode) ||
	    (EXT3_I(parent)->i_flags & EXT3_TOPDIR_FL)) {
		int best_ndir = inodes_per_group;
		int best_group = -1;

		get_random_bytes(&group, sizeof(group));
		parent_group = (unsigned)group % ngroups;
		for (i = 0; i < ngroups; i++) {
			group = (parent_group + i) % ngroups;
			desc = ext3_get_group_desc (sb, group, &bh);
			if (!desc || !desc->bg_free_inodes_count)
				continue;
			if (le16_to_cpu(desc->bg_used_dirs_count) >= best_ndir)
				continue;
			if (le16_to_cpu(desc->bg_free_inodes_count) < avefreei)
				continue;
			if (le16_to_cpu(desc->bg_free_blocks_count) < avefreeb)
				continue;
			best_group = group;
			best_ndir = le16_to_cpu(desc->bg_used_dirs_count);
		}
		if (best_group >= 0)
			return best_group;
		goto fallback;
	}

	blocks_per_dir = (le32_to_cpu(es->s_blocks_count) - freeb) / ndirs;

	max_dirs = ndirs / ngroups + inodes_per_group / 16;
	min_inodes = avefreei - inodes_per_group / 4;
	min_blocks = avefreeb - EXT3_BLOCKS_PER_GROUP(sb) / 4;

	max_debt = EXT3_BLOCKS_PER_GROUP(sb) / max(blocks_per_dir, BLOCK_COST);
	if (max_debt * INODE_COST > inodes_per_group)
		max_debt = inodes_per_group / INODE_COST;
	if (max_debt > 255)
		max_debt = 255;
	if (max_debt == 0)
		max_debt = 1;

	for (i = 0; i < ngroups; i++) {
		group = (parent_group + i) % ngroups;
		desc = ext3_get_group_desc (sb, group, &bh);
		if (!desc || !desc->bg_free_inodes_count)
			continue;
		if (le16_to_cpu(desc->bg_used_dirs_count) >= max_dirs)
			continue;
		if (le16_to_cpu(desc->bg_free_inodes_count) < min_inodes)
			continue;
		if (le16_to_cpu(desc->bg_free_blocks_count) < min_blocks)
			continue;
		return group;
	}

fallback:
	for (i = 0; i < ngroups; i++) {
		group = (parent_group + i) % ngroups;
		desc = ext3_get_group_desc (sb, group, &bh);
		if (!desc || !desc->bg_free_inodes_count)
			continue;
		if (le16_to_cpu(desc->bg_free_inodes_count) >= avefreei)
			return group;
	}

	if (avefreei) {
		/*
		 * The free-inodes counter is approximate, and for really small
		 * filesystems the above test can fail to find any blockgroups
		 */
		avefreei = 0;
		goto fallback;
	}

	return -1;
}

long ext3_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
{
	struct inode *inode = filp->f_dentry->d_inode;
	struct ext3_inode_info *ei = EXT3_I(inode);
	unsigned int flags;
	unsigned short rsv_window_size;

	ext3_debug ("cmd = %u, arg = %lu\n", cmd, arg);

	switch (cmd) {
	case EXT3_IOC_GETFLAGS:
		ext3_get_inode_flags(ei);
		flags = ei->i_flags & EXT3_FL_USER_VISIBLE;
		return put_user(flags, (int __user *) arg);
	case EXT3_IOC_SETFLAGS: {
		handle_t *handle = NULL;
		int err;
		struct ext3_iloc iloc;
		unsigned int oldflags;
		unsigned int jflag;

		if (!inode_owner_or_capable(inode))
			return -EACCES;

		if (get_user(flags, (int __user *) arg))
			return -EFAULT;

		err = mnt_want_write_file(filp);
		if (err)
			return err;

		flags = ext3_mask_flags(inode->i_mode, flags);

		mutex_lock(&inode->i_mutex);

		/* Is it quota file? Do not allow user to mess with it */
		err = -EPERM;
		if (IS_NOQUOTA(inode))
			goto flags_out;

		oldflags = ei->i_flags;

		/* The JOURNAL_DATA flag is modifiable only by root */
		jflag = flags & EXT3_JOURNAL_DATA_FL;

		/*
		 * The IMMUTABLE and APPEND_ONLY flags can only be changed by
		 * the relevant capability.
		 *
		 * This test looks nicer. Thanks to Pauline Middelink
		 */
		if ((flags ^ oldflags) & (EXT3_APPEND_FL | EXT3_IMMUTABLE_FL)) {
			if (!capable(CAP_LINUX_IMMUTABLE))
				goto flags_out;
		}

		/*
		 * The JOURNAL_DATA flag can only be changed by
		 * the relevant capability.
		 */
		if ((jflag ^ oldflags) & (EXT3_JOURNAL_DATA_FL)) {
			if (!capable(CAP_SYS_RESOURCE))
				goto flags_out;
		}

		handle = ext3_journal_start(inode, 1);
		if (IS_ERR(handle)) {
			err = PTR_ERR(handle);
			goto flags_out;
		}
		if (IS_SYNC(inode))
			handle->h_sync = 1;
		err = ext3_reserve_inode_write(handle, inode, &iloc);
		if (err)
			goto flags_err;

		flags = flags & EXT3_FL_USER_MODIFIABLE;
		flags |= oldflags & ~EXT3_FL_USER_MODIFIABLE;
		ei->i_flags = flags;

		ext3_set_inode_flags(inode);
		inode->i_ctime = CURRENT_TIME_SEC;

		err = ext3_mark_iloc_dirty(handle, inode, &iloc);
flags_err:
		ext3_journal_stop(handle);
		if (err)
			goto flags_out;

		if ((jflag ^ oldflags) & (EXT3_JOURNAL_DATA_FL))
			err = ext3_change_inode_journal_flag(inode, jflag);
flags_out:
		mutex_unlock(&inode->i_mutex);
		mnt_drop_write_file(filp);
		return err;
	}
	case EXT3_IOC_GETVERSION:
	case EXT3_IOC_GETVERSION_OLD:
		return put_user(inode->i_generation, (int __user *) arg);
	case EXT3_IOC_SETVERSION:
//.........这里部分代码省略.........

/*
 * There are two policies for allocating an inode.  If the new inode is
 * a directory, then a forward search is made for a block group with both
 * free space and a low directory-to-inode ratio; if that fails, then of
 * the groups with above-average free space, that group with the fewest
 * directories already is chosen.
 *
 * For other inodes, search forward from the parent directory's block
 * group to find a free inode.
 */
struct inode *ext3_new_inode(handle_t *handle, struct inode * dir,
			     const struct qstr *qstr, umode_t mode)
{
	struct super_block *sb;
	struct buffer_head *bitmap_bh = NULL;
	struct buffer_head *bh2;
	int group;
	unsigned long ino = 0;
	struct inode * inode;
	struct ext3_group_desc * gdp = NULL;
	struct ext3_super_block * es;
	struct ext3_inode_info *ei;
	struct ext3_sb_info *sbi;
	int err = 0;
	struct inode *ret;
	int i;

	/* Cannot create files in a deleted directory */
	if (!dir || !dir->i_nlink)
		return ERR_PTR(-EPERM);

	sb = dir->i_sb;
	trace_ext3_request_inode(dir, mode);
	inode = new_inode(sb);
	if (!inode)
		return ERR_PTR(-ENOMEM);
	ei = EXT3_I(inode);

	sbi = EXT3_SB(sb);
	es = sbi->s_es;
	if (S_ISDIR(mode))
		group = find_group_orlov(sb, dir);
	else
		group = find_group_other(sb, dir);

	err = -ENOSPC;
	if (group == -1)
		goto out;

	for (i = 0; i < sbi->s_groups_count; i++) {
		err = -EIO;

		gdp = ext3_get_group_desc(sb, group, &bh2);
		if (!gdp)
			goto fail;

		brelse(bitmap_bh);
		bitmap_bh = read_inode_bitmap(sb, group);
		if (!bitmap_bh)
			goto fail;

		ino = 0;

repeat_in_this_group:
		ino = ext3_find_next_zero_bit((unsigned long *)
				bitmap_bh->b_data, EXT3_INODES_PER_GROUP(sb), ino);
		if (ino < EXT3_INODES_PER_GROUP(sb)) {

			BUFFER_TRACE(bitmap_bh, "get_write_access");
			err = ext3_journal_get_write_access(handle, bitmap_bh);
			if (err)
				goto fail;

			if (!ext3_set_bit_atomic(sb_bgl_lock(sbi, group),
						ino, bitmap_bh->b_data)) {
				/* we won it */
				BUFFER_TRACE(bitmap_bh,
					"call ext3_journal_dirty_metadata");
				err = ext3_journal_dirty_metadata(handle,
								bitmap_bh);
				if (err)
					goto fail;
				goto got;
			}
			/* we lost it */
			journal_release_buffer(handle, bitmap_bh);

			if (++ino < EXT3_INODES_PER_GROUP(sb))
				goto repeat_in_this_group;
		}

		/*
		 * This case is possible in concurrent environment.  It is very
		 * rare.  We cannot repeat the find_group_xxx() call because
		 * that will simply return the same blockgroup, because the
		 * group descriptor metadata has not yet been updated.
		 * So we just go onto the next blockgroup.
		 */
		if (++group == sbi->s_groups_count)
			group = 0;
//.........这里部分代码省略.........

int mlowerfs_ext3_write_record(struct file *file, void *buf, int bufsize,
                               loff_t *offs, int force_sync)
{
	struct buffer_head *bh = NULL;
	unsigned long block;
	struct inode *inode = file->f_dentry->d_inode;
	loff_t old_size = i_size_read(inode), offset = *offs;
	loff_t new_size = i_size_read(inode);
	handle_t *handle;
	int err = 0, block_count = 0, blocksize, size, boffs;

	/* Determine how many transaction credits are needed */
	blocksize = 1 << inode->i_blkbits;
	block_count = (*offs & (blocksize - 1)) + bufsize;
	block_count = (block_count + blocksize - 1) >> inode->i_blkbits;

	handle = _mlowerfs_ext3_journal_start(inode,
	                                      block_count * EXT3_DATA_TRANS_BLOCKS(inode->i_sb) + 2);
	if (IS_ERR(handle)) {
		MERROR("can't start transaction for %d blocks (%d bytes)\n",
		       block_count * EXT3_DATA_TRANS_BLOCKS(inode->i_sb) + 2, bufsize);
		return PTR_ERR(handle);
	}

	while (bufsize > 0) {
		if (bh != NULL)
			brelse(bh);

		block = offset >> inode->i_blkbits;
		boffs = offset & (blocksize - 1);
		size = min(blocksize - boffs, bufsize);
		bh = _mlowerfs_ext3_bread(handle, inode, block, 1, &err);
		if (!bh) {
			MERROR("can't read/create block: %d\n", err);
			goto out;
		}

		err = _mlowerfs_ext3_journal_get_write_access(handle, bh);
		if (err) {
			MERROR("journal_get_write_access() returned error %d\n",
			       err);
			goto out;
		}
		MASSERT(bh->b_data + boffs + size <= bh->b_data + bh->b_size);
		memcpy(bh->b_data + boffs, buf, size);
		err = _mlowerfs_ext3_journal_dirty_metadata(handle, bh);
		if (err) {
			MERROR("journal_dirty_metadata() returned error %d\n",
			       err);
			goto out;
		}
		if (offset + size > new_size)
			new_size = offset + size;
		offset += size;
		bufsize -= size;
		buf += size;
	}

	if (force_sync)
		handle->h_sync = 1; /* recovery likes this */
out:
	if (bh)
		brelse(bh);

	/* correct in-core and on-disk sizes */
	if (new_size > i_size_read(inode)) {
		lock_kernel();
		if (new_size > i_size_read(inode))
			i_size_write(inode, new_size);
		if (i_size_read(inode) > EXT3_I(inode)->i_disksize)
			EXT3_I(inode)->i_disksize = i_size_read(inode);
		if (i_size_read(inode) > old_size)
			mark_inode_dirty(inode);
		unlock_kernel();
	}

	_mlowerfs_ext3_journal_stop(handle);

	if (err == 0)
		*offs = offset;
	return err;
}

/*
 * Called at inode eviction from icache
 */
void ext3_evict_inode (struct inode *inode)
{
	struct ext3_inode_info *ei = EXT3_I(inode);
	struct ext3_block_alloc_info *rsv;
	handle_t *handle;
	int want_delete = 0;

	trace_ext3_evict_inode(inode);
	if (!inode->i_nlink && !is_bad_inode(inode)) {
		dquot_initialize(inode);
		want_delete = 1;
	}

	/*
	 * When journalling data dirty buffers are tracked only in the journal.
	 * So although mm thinks everything is clean and ready for reaping the
	 * inode might still have some pages to write in the running
	 * transaction or waiting to be checkpointed. Thus calling
	 * journal_invalidatepage() (via truncate_inode_pages()) to discard
	 * these buffers can cause data loss. Also even if we did not discard
	 * these buffers, we would have no way to find them after the inode
	 * is reaped and thus user could see stale data if he tries to read
	 * them before the transaction is checkpointed. So be careful and
	 * force everything to disk here... We use ei->i_datasync_tid to
	 * store the newest transaction containing inode's data.
	 *
	 * Note that directories do not have this problem because they don't
	 * use page cache.
	 *
	 * The s_journal check handles the case when ext3_get_journal() fails
	 * and puts the journal inode.
	 */
	if (inode->i_nlink && ext3_should_journal_data(inode) &&
	    EXT3_SB(inode->i_sb)->s_journal &&
	    (S_ISLNK(inode->i_mode) || S_ISREG(inode->i_mode)) &&
	    inode->i_ino != EXT3_JOURNAL_INO) {
		tid_t commit_tid = atomic_read(&ei->i_datasync_tid);
		journal_t *journal = EXT3_SB(inode->i_sb)->s_journal;

		log_start_commit(journal, commit_tid);
		log_wait_commit(journal, commit_tid);
		filemap_write_and_wait(&inode->i_data);
	}
	truncate_inode_pages(&inode->i_data, 0);

	ext3_discard_reservation(inode);
	rsv = ei->i_block_alloc_info;
	ei->i_block_alloc_info = NULL;
	if (unlikely(rsv))
		kfree(rsv);

	if (!want_delete)
		goto no_delete;

	handle = start_transaction(inode);
	if (IS_ERR(handle)) {
		/*
		 * If we're going to skip the normal cleanup, we still need to
		 * make sure that the in-core orphan linked list is properly
		 * cleaned up.
		 */
		ext3_orphan_del(NULL, inode);
		goto no_delete;
	}

	if (IS_SYNC(inode))
		handle->h_sync = 1;
	inode->i_size = 0;
	if (inode->i_blocks)
		ext3_truncate(inode);
	/*
	 * Kill off the orphan record created when the inode lost the last
	 * link.  Note that ext3_orphan_del() has to be able to cope with the
	 * deletion of a non-existent orphan - ext3_truncate() could
	 * have removed the record.
	 */
	ext3_orphan_del(handle, inode);
	ei->i_dtime = get_seconds();

	/*
	 * One subtle ordering requirement: if anything has gone wrong
	 * (transaction abort, IO errors, whatever), then we can still
	 * do these next steps (the fs will already have been marked as
	 * having errors), but we can't free the inode if the mark_dirty
	 * fails.
	 */
	if (ext3_mark_inode_dirty(handle, inode)) {
		/* If that failed, just dquot_drop() and be done with that */
		dquot_drop(inode);
		clear_inode(inode);
	} else {
		ext3_xattr_delete_inode(handle, inode);
		dquot_free_inode(inode);
		dquot_drop(inode);
		clear_inode(inode);
		ext3_free_inode(handle, inode);
	}
//.........这里部分代码省略.........

/*
 * inode->i_sem: down, or inode is just being initialized
 * BKL: held
 */
static int
ext3_do_set_acl(handle_t *handle, struct inode *inode, int type,
		struct posix_acl *acl)
{
	struct ext3_inode_info *ei = EXT3_I(inode);
	int name_index;
	void *value = NULL;
	size_t size;
	int error;

	if (S_ISLNK(inode->i_mode))
		return -ENODATA;

	switch(type) {
		case ACL_TYPE_ACCESS:
			name_index = EXT3_XATTR_INDEX_POSIX_ACL_ACCESS;
			if (acl) {
				mode_t mode = inode->i_mode;
				error = posix_acl_equiv_mode(acl, &mode);
				if (error < 0)
					return error;
				else {
					inode->i_mode = mode;
					ext3_mark_inode_dirty(handle, inode);
					if (error == 0)
						acl = NULL;
				}
			}
			break;

		case ACL_TYPE_DEFAULT:
			name_index = EXT3_XATTR_INDEX_POSIX_ACL_DEFAULT;
			if (!S_ISDIR(inode->i_mode))
				return acl ? -EACCES : 0;
			break;

		default:
			return -EINVAL;
	}
 	if (acl) {
		if (acl->a_count > EXT3_ACL_MAX_ENTRIES)
			return -EINVAL;
		value = ext3_acl_to_disk(acl, &size);
		if (IS_ERR(value))
			return (int)PTR_ERR(value);
	}

	error = ext3_xattr_set_handle(handle, inode, name_index, "",
				      value, size, 0);

	if (value)
		kfree(value);
	if (!error) {
		switch(type) {
			case ACL_TYPE_ACCESS:
				if (ei->i_acl != EXT3_ACL_NOT_CACHED)
					posix_acl_release(ei->i_acl);
				ei->i_acl = posix_acl_dup(acl);
				break;

			case ACL_TYPE_DEFAULT:
				if (ei->i_default_acl != EXT3_ACL_NOT_CACHED)
					posix_acl_release(ei->i_default_acl);
				ei->i_default_acl = posix_acl_dup(acl);
				break;
		}
	}
	return error;
}

int ext3_ioctl (struct inode * inode, struct file * filp, unsigned int cmd,
		unsigned long arg)
{
	struct ext3_inode_info *ei = EXT3_I(inode);
	unsigned int flags;
	unsigned short rsv_window_size;

	ext3_debug ("cmd = %u, arg = %lu\n", cmd, arg);

	switch (cmd) {
	case EXT3_IOC_GETFLAGS:
		flags = ei->i_flags & EXT3_FL_USER_VISIBLE;
		return put_user(flags, (int __user *) arg);
	case EXT3_IOC_SETFLAGS: {
		handle_t *handle = NULL;
		int err;
		struct ext3_iloc iloc;
		unsigned int oldflags;
		unsigned int jflag;

		if (IS_RDONLY(inode))
			return -EROFS;

		if ((current->fsuid != inode->i_uid) && !capable(CAP_FOWNER))
			return -EACCES;

		if (get_user(flags, (int __user *) arg))
			return -EFAULT;

		if (!S_ISDIR(inode->i_mode))
			flags &= ~EXT3_DIRSYNC_FL;

		mutex_lock(&inode->i_mutex);
		oldflags = ei->i_flags;

		/* The JOURNAL_DATA flag is modifiable only by root */
		jflag = flags & EXT3_JOURNAL_DATA_FL;

		/*
		 * The IMMUTABLE and APPEND_ONLY flags can only be changed by
		 * the relevant capability.
		 *
		 * This test looks nicer. Thanks to Pauline Middelink
		 */
		if ((flags ^ oldflags) & (EXT3_APPEND_FL | EXT3_IMMUTABLE_FL)) {
			if (!capable(CAP_LINUX_IMMUTABLE)) {
				mutex_unlock(&inode->i_mutex);
				return -EPERM;
			}
		}

		/*
		 * The JOURNAL_DATA flag can only be changed by
		 * the relevant capability.
		 */
		if ((jflag ^ oldflags) & (EXT3_JOURNAL_DATA_FL)) {
			if (!capable(CAP_SYS_RESOURCE)) {
				mutex_unlock(&inode->i_mutex);
				return -EPERM;
			}
		}


		handle = ext3_journal_start(inode, 1);
		if (IS_ERR(handle)) {
			mutex_unlock(&inode->i_mutex);
			return PTR_ERR(handle);
		}
		if (IS_SYNC(inode))
			handle->h_sync = 1;
		err = ext3_reserve_inode_write(handle, inode, &iloc);
		if (err)
			goto flags_err;

		flags = flags & EXT3_FL_USER_MODIFIABLE;
		flags |= oldflags & ~EXT3_FL_USER_MODIFIABLE;
		ei->i_flags = flags;

		ext3_set_inode_flags(inode);
		inode->i_ctime = CURRENT_TIME_SEC;

		err = ext3_mark_iloc_dirty(handle, inode, &iloc);
flags_err:
		ext3_journal_stop(handle);
		if (err) {
			mutex_unlock(&inode->i_mutex);
			return err;
		}

		if ((jflag ^ oldflags) & (EXT3_JOURNAL_DATA_FL))
			err = ext3_change_inode_journal_flag(inode, jflag);
		mutex_unlock(&inode->i_mutex);
		return err;
	}
	case EXT3_IOC_GETVERSION:
	case EXT3_IOC_GETVERSION_OLD:
		return put_user(inode->i_generation, (int __user *) arg);
	case EXT3_IOC_SETVERSION:
	case EXT3_IOC_SETVERSION_OLD: {
		handle_t *handle;
//.........这里部分代码省略.........

static void * ext3_follow_link(struct dentry *dentry, struct nameidata *nd)
{
	struct ext3_inode_info *ei = EXT3_I(dentry->d_inode);
	nd_set_link(nd, (char*)ei->i_data);
	return NULL;
}

static int find_group_orlov(struct super_block *sb, struct inode *parent)
{
	int parent_group = EXT3_I(parent)->i_block_group;
	struct ext3_sb_info *sbi = EXT3_SB(sb);
	int ngroups = sbi->s_groups_count;
	int inodes_per_group = EXT3_INODES_PER_GROUP(sb);
	unsigned int freei, avefreei;
	ext3_fsblk_t freeb, avefreeb;
	unsigned int ndirs;
	int max_dirs, min_inodes;
	ext3_grpblk_t min_blocks;
	int group = -1, i;
	struct ext3_group_desc *desc;

	freei = percpu_counter_read_positive(&sbi->s_freeinodes_counter);
	avefreei = freei / ngroups;
	freeb = percpu_counter_read_positive(&sbi->s_freeblocks_counter);
	avefreeb = freeb / ngroups;
	ndirs = percpu_counter_read_positive(&sbi->s_dirs_counter);

	if ((parent == sb->s_root->d_inode) ||
	    (EXT3_I(parent)->i_flags & EXT3_TOPDIR_FL)) {
		int best_ndir = inodes_per_group;
		int best_group = -1;

		group = prandom_u32();
		parent_group = (unsigned)group % ngroups;
		for (i = 0; i < ngroups; i++) {
			group = (parent_group + i) % ngroups;
			desc = ext3_get_group_desc (sb, group, NULL);
			if (!desc || !desc->bg_free_inodes_count)
				continue;
			if (le16_to_cpu(desc->bg_used_dirs_count) >= best_ndir)
				continue;
			if (le16_to_cpu(desc->bg_free_inodes_count) < avefreei)
				continue;
			if (le16_to_cpu(desc->bg_free_blocks_count) < avefreeb)
				continue;
			best_group = group;
			best_ndir = le16_to_cpu(desc->bg_used_dirs_count);
		}
		if (best_group >= 0)
			return best_group;
		goto fallback;
	}

	max_dirs = ndirs / ngroups + inodes_per_group / 16;
	min_inodes = avefreei - inodes_per_group / 4;
	min_blocks = avefreeb - EXT3_BLOCKS_PER_GROUP(sb) / 4;

	for (i = 0; i < ngroups; i++) {
		group = (parent_group + i) % ngroups;
		desc = ext3_get_group_desc (sb, group, NULL);
		if (!desc || !desc->bg_free_inodes_count)
			continue;
		if (le16_to_cpu(desc->bg_used_dirs_count) >= max_dirs)
			continue;
		if (le16_to_cpu(desc->bg_free_inodes_count) < min_inodes)
			continue;
		if (le16_to_cpu(desc->bg_free_blocks_count) < min_blocks)
			continue;
		return group;
	}

fallback:
	for (i = 0; i < ngroups; i++) {
		group = (parent_group + i) % ngroups;
		desc = ext3_get_group_desc (sb, group, NULL);
		if (!desc || !desc->bg_free_inodes_count)
			continue;
		if (le16_to_cpu(desc->bg_free_inodes_count) >= avefreei)
			return group;
	}

	if (avefreei) {
		/*
		 * The free-inodes counter is approximate, and for really small
		 * filesystems the above test can fail to find any blockgroups
		 */
		avefreei = 0;
		goto fallback;
	}

	return -1;
}