C++ BUFFER_TRACE函数代码示例

int journal_forget (handle_t *handle, struct buffer_head *bh)
{
	transaction_t *transaction = handle->h_transaction;
	journal_t *journal = transaction->t_journal;
	struct journal_head *jh;
	int drop_reserve = 0;
	int err = 0;
	int was_modified = 0;

	BUFFER_TRACE(bh, "entry");

	jbd_lock_bh_state(bh);
	spin_lock(&journal->j_list_lock);

	if (!buffer_jbd(bh))
		goto not_jbd;
	jh = bh2jh(bh);

	/* Critical error: attempting to delete a bitmap buffer, maybe?
	 * Don't do any jbd operations, and return an error. */
	if (!J_EXPECT_JH(jh, !jh->b_committed_data,
			 "inconsistent data on disk")) {
		err = -EIO;
		goto not_jbd;
	}

	/* keep track of wether or not this transaction modified us */
	was_modified = jh->b_modified;

	/*
	 * The buffer's going from the transaction, we must drop
	 * all references -bzzz
	 */
	jh->b_modified = 0;

	if (jh->b_transaction == handle->h_transaction) {
		J_ASSERT_JH(jh, !jh->b_frozen_data);

		/* If we are forgetting a buffer which is already part
		 * of this transaction, then we can just drop it from
		 * the transaction immediately. */
		clear_buffer_dirty(bh);
		clear_buffer_jbddirty(bh);

		JBUFFER_TRACE(jh, "belongs to current transaction: unfile");

		/*
		 * we only want to drop a reference if this transaction
		 * modified the buffer
		 */
		if (was_modified)
			drop_reserve = 1;

		/*
		 * We are no longer going to journal this buffer.
		 * However, the commit of this transaction is still
		 * important to the buffer: the delete that we are now
		 * processing might obsolete an old log entry, so by
		 * committing, we can satisfy the buffer's checkpoint.
		 *
		 * So, if we have a checkpoint on the buffer, we should
		 * now refile the buffer on our BJ_Forget list so that
		 * we know to remove the checkpoint after we commit.
		 */

		if (jh->b_cp_transaction) {
			__journal_temp_unlink_buffer(jh);
			__journal_file_buffer(jh, transaction, BJ_Forget);
		} else {
			__journal_unfile_buffer(jh);
			journal_remove_journal_head(bh);
			__brelse(bh);
			if (!buffer_jbd(bh)) {
				spin_unlock(&journal->j_list_lock);
				jbd_unlock_bh_state(bh);
				__bforget(bh);
				goto drop;
			}
		}
	} else if (jh->b_transaction) {
		J_ASSERT_JH(jh, (jh->b_transaction ==
				 journal->j_committing_transaction));
		/* However, if the buffer is still owned by a prior
		 * (committing) transaction, we can't drop it yet... */
		JBUFFER_TRACE(jh, "belongs to older transaction");
		/* ... but we CAN drop it from the new transaction if we
		 * have also modified it since the original commit. */

		if (jh->b_next_transaction) {
			J_ASSERT(jh->b_next_transaction == transaction);
			jh->b_next_transaction = NULL;

			/*
			 * only drop a reference if this transaction modified
			 * the buffer
			 */
			if (was_modified)
				drop_reserve = 1;
		}
	}
//.........这里部分代码省略.........

static void write_one_revoke_record(transaction_t *transaction,
				    struct list_head *log_bufs,
				    struct buffer_head **descriptorp,
				    int *offsetp,
				    struct jbd2_revoke_record_s *record)
{
	journal_t *journal = transaction->t_journal;
	int csum_size = 0;
	struct buffer_head *descriptor;
	int sz, offset;

	/* If we are already aborting, this all becomes a noop.  We
           still need to go round the loop in
           jbd2_journal_write_revoke_records in order to free all of the
           revoke records: only the IO to the journal is omitted. */
	if (is_journal_aborted(journal))
		return;

	descriptor = *descriptorp;
	offset = *offsetp;

	/* Do we need to leave space at the end for a checksum? */
	if (jbd2_journal_has_csum_v2or3(journal))
		csum_size = sizeof(struct jbd2_journal_block_tail);

	if (jbd2_has_feature_64bit(journal))
		sz = 8;
	else
		sz = 4;

	/* Make sure we have a descriptor with space left for the record */
	if (descriptor) {
		if (offset + sz > journal->j_blocksize - csum_size) {
			flush_descriptor(journal, descriptor, offset);
			descriptor = NULL;
		}
	}

	if (!descriptor) {
		descriptor = jbd2_journal_get_descriptor_buffer(transaction,
							JBD2_REVOKE_BLOCK);
		if (!descriptor)
			return;

		/* Record it so that we can wait for IO completion later */
		BUFFER_TRACE(descriptor, "file in log_bufs");
		jbd2_file_log_bh(log_bufs, descriptor);

		offset = sizeof(jbd2_journal_revoke_header_t);
		*descriptorp = descriptor;
	}

	if (jbd2_has_feature_64bit(journal))
		* ((__be64 *)(&descriptor->b_data[offset])) =
			cpu_to_be64(record->blocknr);
	else
		* ((__be32 *)(&descriptor->b_data[offset])) =
			cpu_to_be32(record->blocknr);
	offset += sz;

	*offsetp = offset;
}

/*
 * Perform an actual checkpoint. We take the first transaction on the
 * list of transactions to be checkpointed and send all its buffers
 * to disk. We submit larger chunks of data at once.
 *
 * The journal should be locked before calling this function.
 * Called with j_checkpoint_mutex held.
 */
int jbd2_log_do_checkpoint(journal_t *journal)
{
	struct journal_head	*jh;
	struct buffer_head	*bh;
	transaction_t		*transaction;
	tid_t			this_tid;
	int			result, batch_count = 0;

	jbd_debug(1, "Start checkpoint\n");

	/*
	 * First thing: if there are any transactions in the log which
	 * don't need checkpointing, just eliminate them from the
	 * journal straight away.
	 */
	result = jbd2_cleanup_journal_tail(journal);
	trace_jbd2_checkpoint(journal, result);
	jbd_debug(1, "cleanup_journal_tail returned %d\n", result);
	if (result <= 0)
		return result;

	/*
	 * OK, we need to start writing disk blocks.  Take one transaction
	 * and write it.
	 */
	result = 0;
	spin_lock(&journal->j_list_lock);
	if (!journal->j_checkpoint_transactions)
		goto out;
	transaction = journal->j_checkpoint_transactions;
	if (transaction->t_chp_stats.cs_chp_time == 0)
		transaction->t_chp_stats.cs_chp_time = jiffies;
	this_tid = transaction->t_tid;
restart:
	/*
	 * If someone cleaned up this transaction while we slept, we're
	 * done (maybe it's a new transaction, but it fell at the same
	 * address).
	 */
	if (journal->j_checkpoint_transactions != transaction ||
	    transaction->t_tid != this_tid)
		goto out;

	/* checkpoint all of the transaction's buffers */
	while (transaction->t_checkpoint_list) {
		jh = transaction->t_checkpoint_list;
		bh = jh2bh(jh);

		if (buffer_locked(bh)) {
			spin_unlock(&journal->j_list_lock);
			get_bh(bh);
			wait_on_buffer(bh);
			/* the journal_head may have gone by now */
			BUFFER_TRACE(bh, "brelse");
			__brelse(bh);
			goto retry;
		}
		if (jh->b_transaction != NULL) {
			transaction_t *t = jh->b_transaction;
			tid_t tid = t->t_tid;

			transaction->t_chp_stats.cs_forced_to_close++;
			spin_unlock(&journal->j_list_lock);
			if (unlikely(journal->j_flags & JBD2_UNMOUNT))
				/*
				 * The journal thread is dead; so
				 * starting and waiting for a commit
				 * to finish will cause us to wait for
				 * a _very_ long time.
				 */
				printk(KERN_ERR
		"JBD2: %s: Waiting for Godot: block %llu\n",
		journal->j_devname, (unsigned long long) bh->b_blocknr);

			jbd2_log_start_commit(journal, tid);
			jbd2_log_wait_commit(journal, tid);
			goto retry;
		}
		if (!buffer_dirty(bh)) {
			if (unlikely(buffer_write_io_error(bh)) && !result)
				result = -EIO;
			BUFFER_TRACE(bh, "remove from checkpoint");
			if (__jbd2_journal_remove_checkpoint(jh))
				/* The transaction was released; we're done */
				goto out;
			continue;
		}
		/*
		 * Important: we are about to write the buffer, and
		 * possibly block, while still holding the journal
		 * lock.  We cannot afford to let the transaction
		 * logic start messing around with this buffer before
//.........这里部分代码省略.........

/*
 *  Submit all the data buffers to disk
 */
static int journal_submit_data_buffers(journal_t *journal,
                                       transaction_t *commit_transaction,
                                       int write_op)
{
    struct journal_head *jh;
    struct buffer_head *bh;
    int locked;
    int bufs = 0;
    struct buffer_head **wbuf = journal->j_wbuf;
    int err = 0;

    /*
     * Whenever we unlock the journal and sleep, things can get added
     * onto ->t_sync_datalist, so we have to keep looping back to
     * write_out_data until we *know* that the list is empty.
     *
     * Cleanup any flushed data buffers from the data list.  Even in
     * abort mode, we want to flush this out as soon as possible.
     */
write_out_data:
    cond_resched();
    spin_lock(&journal->j_list_lock);

    while (commit_transaction->t_sync_datalist) {
        jh = commit_transaction->t_sync_datalist;
        bh = jh2bh(jh);
        locked = 0;

        /* Get reference just to make sure buffer does not disappear
         * when we are forced to drop various locks */
        get_bh(bh);
        /* If the buffer is dirty, we need to submit IO and hence
         * we need the buffer lock. We try to lock the buffer without
         * blocking. If we fail, we need to drop j_list_lock and do
         * blocking lock_buffer().
         */
        if (buffer_dirty(bh)) {
            if (!trylock_buffer(bh)) {
                BUFFER_TRACE(bh, "needs blocking lock");
                spin_unlock(&journal->j_list_lock);
                trace_jbd_do_submit_data(journal,
                                         commit_transaction);
                /* Write out all data to prevent deadlocks */
                journal_do_submit_data(wbuf, bufs, write_op);
                bufs = 0;
                lock_buffer(bh);
                spin_lock(&journal->j_list_lock);
            }
            locked = 1;
        }
        /* We have to get bh_state lock. Again out of order, sigh. */
        if (!inverted_lock(journal, bh)) {
            jbd_lock_bh_state(bh);
            spin_lock(&journal->j_list_lock);
        }
        /* Someone already cleaned up the buffer? */
        if (!buffer_jbd(bh) || bh2jh(bh) != jh
                || jh->b_transaction != commit_transaction
                || jh->b_jlist != BJ_SyncData) {
            jbd_unlock_bh_state(bh);
            if (locked)
                unlock_buffer(bh);
            BUFFER_TRACE(bh, "already cleaned up");
            release_data_buffer(bh);
            continue;
        }
        if (locked && test_clear_buffer_dirty(bh)) {
            BUFFER_TRACE(bh, "needs writeout, adding to array");
            wbuf[bufs++] = bh;
            __journal_file_buffer(jh, commit_transaction,
                                  BJ_Locked);
            jbd_unlock_bh_state(bh);
            if (bufs == journal->j_wbufsize) {
                spin_unlock(&journal->j_list_lock);
                trace_jbd_do_submit_data(journal,
                                         commit_transaction);
                journal_do_submit_data(wbuf, bufs, write_op);
                bufs = 0;
                goto write_out_data;
            }
        } else if (!locked && buffer_locked(bh)) {
            __journal_file_buffer(jh, commit_transaction,
                                  BJ_Locked);
            jbd_unlock_bh_state(bh);
            put_bh(bh);
        } else {
            BUFFER_TRACE(bh, "writeout complete: unfile");
            if (unlikely(!buffer_uptodate(bh)))
                err = -EIO;
            __journal_unfile_buffer(jh);
            jbd_unlock_bh_state(bh);
            if (locked)
                unlock_buffer(bh);
            release_data_buffer(bh);
        }

        if (need_resched() || spin_needbreak(&journal->j_list_lock)) {
//.........这里部分代码省略.........

/**
 * ext4_add_groupblocks() -- Add given blocks to an existing group
 * @handle:			handle to this transaction
 * @sb:				super block
 * @block:			start physcial block to add to the block group
 * @count:			number of blocks to free
 *
 * This marks the blocks as free in the bitmap. We ask the
 * mballoc to reload the buddy after this by setting group
 * EXT4_GROUP_INFO_NEED_INIT_BIT flag
 */
void ext4_add_groupblocks(handle_t *handle, struct super_block *sb,
			 ext4_fsblk_t block, unsigned long count)
{
	struct buffer_head *bitmap_bh = NULL;
	struct buffer_head *gd_bh;
	ext4_group_t block_group;
	ext4_grpblk_t bit;
	unsigned int i;
	struct ext4_group_desc *desc;
	struct ext4_super_block *es;
	struct ext4_sb_info *sbi;
	int err = 0, ret, blk_free_count;
	ext4_grpblk_t blocks_freed;
	struct ext4_group_info *grp;

	sbi = EXT4_SB(sb);
	es = sbi->s_es;
	ext4_debug("Adding block(s) %llu-%llu\n", block, block + count - 1);

	ext4_get_group_no_and_offset(sb, block, &block_group, &bit);
	grp = ext4_get_group_info(sb, block_group);
	/*
	 * Check to see if we are freeing blocks across a group
	 * boundary.
	 */
	if (bit + count > EXT4_BLOCKS_PER_GROUP(sb)) {
		goto error_return;
	}
	bitmap_bh = ext4_read_block_bitmap(sb, block_group);
	if (!bitmap_bh)
		goto error_return;
	desc = ext4_get_group_desc(sb, block_group, &gd_bh);
	if (!desc)
		goto error_return;

	if (in_range(ext4_block_bitmap(sb, desc), block, count) ||
	    in_range(ext4_inode_bitmap(sb, desc), block, count) ||
	    in_range(block, ext4_inode_table(sb, desc), sbi->s_itb_per_group) ||
	    in_range(block + count - 1, ext4_inode_table(sb, desc),
		     sbi->s_itb_per_group)) {
		ext4_error(sb, __func__,
			   "Adding blocks in system zones - "
			   "Block = %llu, count = %lu",
			   block, count);
		goto error_return;
	}

	/*
	 * We are about to add blocks to the bitmap,
	 * so we need undo access.
	 */
	BUFFER_TRACE(bitmap_bh, "getting undo access");
	err = ext4_journal_get_undo_access(handle, bitmap_bh);
	if (err)
		goto error_return;

	/*
	 * We are about to modify some metadata.  Call the journal APIs
	 * to unshare ->b_data if a currently-committing transaction is
	 * using it
	 */
	BUFFER_TRACE(gd_bh, "get_write_access");
	err = ext4_journal_get_write_access(handle, gd_bh);
	if (err)
		goto error_return;
	/*
	 * make sure we don't allow a parallel init on other groups in the
	 * same buddy cache
	 */
	down_write(&grp->alloc_sem);
	for (i = 0, blocks_freed = 0; i < count; i++) {
		BUFFER_TRACE(bitmap_bh, "clear bit");
		if (!ext4_clear_bit_atomic(ext4_group_lock_ptr(sb, block_group),
						bit + i, bitmap_bh->b_data)) {
			ext4_error(sb, __func__,
				   "bit already cleared for block %llu",
				   (ext4_fsblk_t)(block + i));
			BUFFER_TRACE(bitmap_bh, "bit already cleared");
		} else {
			blocks_freed++;
		}
	}
	ext4_lock_group(sb, block_group);
	blk_free_count = blocks_freed + ext4_free_blks_count(sb, desc);
	ext4_free_blks_set(sb, desc, blk_free_count);
	desc->bg_checksum = ext4_group_desc_csum(sbi, block_group, desc);
	ext4_unlock_group(sb, block_group);
	percpu_counter_add(&sbi->s_freeblocks_counter, blocks_freed);

//.........这里部分代码省略.........

/*
 * NOTE! When we get the inode, we're the only people
 * that have access to it, and as such there are no
 * race conditions we have to worry about. The inode
 * is not on the hash-lists, and it cannot be reached
 * through the filesystem because the directory entry
 * has been deleted earlier.
 *
 * HOWEVER: we must make sure that we get no aliases,
 * which means that we have to call "clear_inode()"
 * _before_ we mark the inode not in use in the inode
 * bitmaps. Otherwise a newly created file might use
 * the same inode number (not actually the same pointer
 * though), and then we'd have two inodes sharing the
 * same inode number and space on the harddisk.
 */
void ext3_free_inode (handle_t *handle, struct inode * inode)
{
	struct super_block * sb = inode->i_sb;
	int is_directory;
	unsigned long ino;
	struct buffer_head *bitmap_bh = NULL;
	struct buffer_head *bh2;
	unsigned long block_group;
	unsigned long bit;
	struct ext3_group_desc * gdp;
	struct ext3_super_block * es;
	struct ext3_sb_info *sbi;
	int fatal = 0, err;

	if (atomic_read(&inode->i_count) > 1) {
		printk ("ext3_free_inode: inode has count=%d\n",
					atomic_read(&inode->i_count));
		return;
	}
	if (inode->i_nlink) {
		printk ("ext3_free_inode: inode has nlink=%d\n",
			inode->i_nlink);
		return;
	}
	if (!sb) {
		printk("ext3_free_inode: inode on nonexistent device\n");
		return;
	}
	sbi = EXT3_SB(sb);

	ino = inode->i_ino;
	ext3_debug ("freeing inode %lu\n", ino);
	trace_ext3_free_inode(inode);

	is_directory = S_ISDIR(inode->i_mode);

	es = EXT3_SB(sb)->s_es;
	if (ino < EXT3_FIRST_INO(sb) || ino > le32_to_cpu(es->s_inodes_count)) {
		ext3_error (sb, "ext3_free_inode",
			    "reserved or nonexistent inode %lu", ino);
		goto error_return;
	}
	block_group = (ino - 1) / EXT3_INODES_PER_GROUP(sb);
	bit = (ino - 1) % EXT3_INODES_PER_GROUP(sb);
	bitmap_bh = read_inode_bitmap(sb, block_group);
	if (!bitmap_bh)
		goto error_return;

	BUFFER_TRACE(bitmap_bh, "get_write_access");
	fatal = ext3_journal_get_write_access(handle, bitmap_bh);
	if (fatal)
		goto error_return;

	/* Ok, now we can actually update the inode bitmaps.. */
	if (!ext3_clear_bit_atomic(sb_bgl_lock(sbi, block_group),
					bit, bitmap_bh->b_data))
		ext3_error (sb, "ext3_free_inode",
			      "bit already cleared for inode %lu", ino);
	else {
		gdp = ext3_get_group_desc (sb, block_group, &bh2);

		BUFFER_TRACE(bh2, "get_write_access");
		fatal = ext3_journal_get_write_access(handle, bh2);
		if (fatal) goto error_return;

		if (gdp) {
			spin_lock(sb_bgl_lock(sbi, block_group));
			le16_add_cpu(&gdp->bg_free_inodes_count, 1);
			if (is_directory)
				le16_add_cpu(&gdp->bg_used_dirs_count, -1);
			spin_unlock(sb_bgl_lock(sbi, block_group));
			percpu_counter_inc(&sbi->s_freeinodes_counter);
			if (is_directory)
				percpu_counter_dec(&sbi->s_dirs_counter);

		}
		BUFFER_TRACE(bh2, "call ext3_journal_dirty_metadata");
		err = ext3_journal_dirty_metadata(handle, bh2);
		if (!fatal) fatal = err;
	}
	BUFFER_TRACE(bitmap_bh, "call ext3_journal_dirty_metadata");
	err = ext3_journal_dirty_metadata(handle, bitmap_bh);
	if (!fatal)
		fatal = err;
//.........这里部分代码省略.........

static int __process_buffer(journal_t *journal, struct journal_head *jh,
			    int *batch_count, transaction_t *transaction)
{
	struct buffer_head *bh = jh2bh(jh);
	int ret = 0;

	if (buffer_locked(bh)) {
		get_bh(bh);
		spin_unlock(&journal->j_list_lock);
		wait_on_buffer(bh);
		/*                                       */
		BUFFER_TRACE(bh, "brelse");
		__brelse(bh);
		ret = 1;
	} else if (jh->b_transaction != NULL) {
		transaction_t *t = jh->b_transaction;
		tid_t tid = t->t_tid;

		transaction->t_chp_stats.cs_forced_to_close++;
		spin_unlock(&journal->j_list_lock);
		if (unlikely(journal->j_flags & JBD2_UNMOUNT))
			/*
                                                 
                                               
                                        
    */
			printk(KERN_ERR "JBD2: %s: "
			       "Waiting for Godot: block %llu\n",
			       journal->j_devname,
			       (unsigned long long) bh->b_blocknr);
		jbd2_log_start_commit(journal, tid);
		jbd2_log_wait_commit(journal, tid);
		ret = 1;
	} else if (!buffer_dirty(bh)) {
		ret = 1;
		if (unlikely(buffer_write_io_error(bh)))
			ret = -EIO;
		get_bh(bh);
		BUFFER_TRACE(bh, "remove from checkpoint");
		__jbd2_journal_remove_checkpoint(jh);
		spin_unlock(&journal->j_list_lock);
		__brelse(bh);
	} else {
		/*
                                                     
                                                          
                                                        
                                                          
                                              
   */
		BUFFER_TRACE(bh, "queue");
		get_bh(bh);
		J_ASSERT_BH(bh, !buffer_jwrite(bh));
		journal->j_chkpt_bhs[*batch_count] = bh;
		__buffer_relink_io(jh);
		transaction->t_chp_stats.cs_written++;
		(*batch_count)++;
		if (*batch_count == JBD2_NR_BATCH) {
			spin_unlock(&journal->j_list_lock);
			__flush_batch(journal, batch_count);
			ret = 1;
		}
	}
	return ret;
}

static int ext4_update_inline_data(handle_t *handle, struct inode *inode,
				   unsigned int len)
{
	int error;
	void *value = NULL;
	struct ext4_xattr_ibody_find is = {
		.s = { .not_found = -ENODATA, },
	};
	struct ext4_xattr_info i = {
		.name_index = EXT4_XATTR_INDEX_SYSTEM,
		.name = EXT4_XATTR_SYSTEM_DATA,
	};

	/* If the old space is ok, write the data directly. */
	if (len <= EXT4_I(inode)->i_inline_size)
		return 0;

	error = ext4_get_inode_loc(inode, &is.iloc);
	if (error)
		return error;

	error = ext4_xattr_ibody_find(inode, &i, &is);
	if (error)
		goto out;

	BUG_ON(is.s.not_found);

	len -= EXT4_MIN_INLINE_DATA_SIZE;
	value = kzalloc(len, GFP_NOFS);
	if (!value)
		goto out;

	error = ext4_xattr_ibody_get(inode, i.name_index, i.name,
				     value, len);
	if (error == -ENODATA)
		goto out;

	BUFFER_TRACE(is.iloc.bh, "get_write_access");
	error = ext4_journal_get_write_access(handle, is.iloc.bh);
	if (error)
		goto out;

	/* Update the xttr entry. */
	i.value = value;
	i.value_len = len;

	error = ext4_xattr_ibody_inline_set(handle, inode, &i, &is);
	if (error)
		goto out;

	EXT4_I(inode)->i_inline_off = (u16)((void *)is.s.here -
				      (void *)ext4_raw_inode(&is.iloc));
	EXT4_I(inode)->i_inline_size = EXT4_MIN_INLINE_DATA_SIZE +
				le32_to_cpu(is.s.here->e_value_size);
	ext4_set_inode_state(inode, EXT4_STATE_MAY_INLINE_DATA);
	get_bh(is.iloc.bh);
	error = ext4_mark_iloc_dirty(handle, inode, &is.iloc);

out:
	kfree(value);
	brelse(is.iloc.bh);
	return error;
}

/*
 * jbd2_journal_commit_transaction
 *
 * The primary function for committing a transaction to the log.  This
 * function is called by the journal thread to begin a complete commit.
 */
void jbd2_journal_commit_transaction(journal_t *journal)
{
	transaction_t *commit_transaction;
	struct journal_head *jh, *new_jh, *descriptor;
	struct buffer_head **wbuf = journal->j_wbuf;
	int bufs;
	int flags;
	int err;
	unsigned long long blocknr;
	char *tagp = NULL;
	journal_header_t *header;
	journal_block_tag_t *tag = NULL;
	int space_left = 0;
	int first_tag = 0;
	int tag_flag;
	int i;
	int tag_bytes = journal_tag_bytes(journal);

	/*
	 * First job: lock down the current transaction and wait for
	 * all outstanding updates to complete.
	 */

#ifdef COMMIT_STATS
	spin_lock(&journal->j_list_lock);
	summarise_journal_usage(journal);
	spin_unlock(&journal->j_list_lock);
#endif

	/* Do we need to erase the effects of a prior jbd2_journal_flush? */
	if (journal->j_flags & JBD2_FLUSHED) {
		jbd_debug(3, "super block updated\n");
		jbd2_journal_update_superblock(journal, 1);
	} else {
		jbd_debug(3, "superblock not updated\n");
	}

	J_ASSERT(journal->j_running_transaction != NULL);
	J_ASSERT(journal->j_committing_transaction == NULL);

	commit_transaction = journal->j_running_transaction;
	J_ASSERT(commit_transaction->t_state == T_RUNNING);

	jbd_debug(1, "JBD: starting commit of transaction %d\n",
			commit_transaction->t_tid);

	spin_lock(&journal->j_state_lock);
	commit_transaction->t_state = T_LOCKED;

	spin_lock(&commit_transaction->t_handle_lock);
	while (commit_transaction->t_updates) {
		DEFINE_WAIT(wait);

		prepare_to_wait(&journal->j_wait_updates, &wait,
					TASK_UNINTERRUPTIBLE);
		if (commit_transaction->t_updates) {
			spin_unlock(&commit_transaction->t_handle_lock);
			spin_unlock(&journal->j_state_lock);
			schedule();
			spin_lock(&journal->j_state_lock);
			spin_lock(&commit_transaction->t_handle_lock);
		}
		finish_wait(&journal->j_wait_updates, &wait);
	}
	spin_unlock(&commit_transaction->t_handle_lock);

	J_ASSERT (commit_transaction->t_outstanding_credits <=
			journal->j_max_transaction_buffers);

	/*
	 * First thing we are allowed to do is to discard any remaining
	 * BJ_Reserved buffers.  Note, it is _not_ permissible to assume
	 * that there are no such buffers: if a large filesystem
	 * operation like a truncate needs to split itself over multiple
	 * transactions, then it may try to do a jbd2_journal_restart() while
	 * there are still BJ_Reserved buffers outstanding.  These must
	 * be released cleanly from the current transaction.
	 *
	 * In this case, the filesystem must still reserve write access
	 * again before modifying the buffer in the new transaction, but
	 * we do not require it to remember exactly which old buffers it
	 * has reserved.  This is consistent with the existing behaviour
	 * that multiple jbd2_journal_get_write_access() calls to the same
	 * buffer are perfectly permissable.
	 */
	while (commit_transaction->t_reserved_list) {
		jh = commit_transaction->t_reserved_list;
		JBUFFER_TRACE(jh, "reserved, unused: refile");
		/*
		 * A jbd2_journal_get_undo_access()+jbd2_journal_release_buffer() may
		 * leave undo-committed data.
		 */
		if (jh->b_committed_data) {
			struct buffer_head *bh = jh2bh(jh);
//.........这里部分代码省略.........

/*
 * Clean up a transaction's checkpoint list.  
 *
 * We wait for any pending IO to complete and make sure any clean
 * buffers are removed from the transaction. 
 *
 * Return 1 if we performed any actions which might have destroyed the
 * checkpoint.  (journal_remove_checkpoint() deletes the transaction when
 * the last checkpoint buffer is cleansed)
 *
 * Called with the journal locked.
 * Called with journal_datalist_lock held.
 */
static int __cleanup_transaction(journal_t *journal, transaction_t *transaction)
{
	struct journal_head *jh, *next_jh, *last_jh;
	struct buffer_head *bh;
	int ret = 0;

	assert_spin_locked(&journal_datalist_lock);
	jh = transaction->t_checkpoint_list;
	if (!jh)
		return 0;

	last_jh = jh->b_cpprev;
	next_jh = jh;
	do {
		jh = next_jh;
		bh = jh2bh(jh);
		if (buffer_locked(bh)) {
			atomic_inc(&bh->b_count);
			spin_unlock(&journal_datalist_lock);
			unlock_journal(journal);
			wait_on_buffer(bh);
			/* the journal_head may have gone by now */
			BUFFER_TRACE(bh, "brelse");
			__brelse(bh);
			goto out_return_1;
		}
		
		if (jh->b_transaction != NULL) {
			transaction_t *transaction = jh->b_transaction;
			tid_t tid = transaction->t_tid;

			spin_unlock(&journal_datalist_lock);
			log_start_commit(journal, transaction);
			unlock_journal(journal);
			log_wait_commit(journal, tid);
			goto out_return_1;
		}

		/*
		 * We used to test for (jh->b_list != BUF_CLEAN) here.
		 * But unmap_underlying_metadata() can place buffer onto
		 * BUF_CLEAN. Since refile_buffer() no longer takes buffers
		 * off checkpoint lists, we cope with it here
		 */
		/*
		 * AKPM: I think the buffer_jdirty test is redundant - it
		 * shouldn't have NULL b_transaction?
		 */
		next_jh = jh->b_cpnext;
		if (!buffer_dirty(bh) && !buffer_jdirty(bh)) {
			BUFFER_TRACE(bh, "remove from checkpoint");
			__journal_remove_checkpoint(jh);
			__journal_remove_journal_head(bh);
			refile_buffer(bh);
			__brelse(bh);
			ret = 1;
		}
		
		jh = next_jh;
	} while (jh != last_jh);

	return ret;
out_return_1:
	lock_journal(journal);
	spin_lock(&journal_datalist_lock);
	return 1;
}

/*
 * Clean up a transaction's checkpoint list.
 *
 * We wait for any pending IO to complete and make sure any clean
 * buffers are removed from the transaction.
 *
 * Return 1 if we performed any actions which might have destroyed the
 * checkpoint.  (journal_remove_checkpoint() deletes the transaction when
 * the last checkpoint buffer is cleansed)
 *
 * Called with j_list_lock held.
 */
static int __cleanup_transaction(journal_t *journal, transaction_t *transaction)
{
	struct journal_head *jh, *next_jh, *last_jh;
	struct buffer_head *bh;
	int ret = 0;

	assert_spin_locked(&journal->j_list_lock);
	jh = transaction->t_checkpoint_list;
	if (!jh)
		return 0;

	last_jh = jh->b_cpprev;
	next_jh = jh;
	do {
		jh = next_jh;
		bh = jh2bh(jh);
		if (buffer_locked(bh)) {
			atomic_inc(&bh->b_count);
			spin_unlock(&journal->j_list_lock);
			wait_on_buffer(bh);
			/* the journal_head may have gone by now */
			BUFFER_TRACE(bh, "brelse");
			__brelse(bh);
			goto out_return_1;
		}

		/*
		 * This is foul
		 */
		if (!jbd_trylock_bh_state(bh)) {
			jbd_sync_bh(journal, bh);
			goto out_return_1;
		}

		if (jh->b_transaction != NULL) {
			transaction_t *t = jh->b_transaction;
			tid_t tid = t->t_tid;

			spin_unlock(&journal->j_list_lock);
			jbd_unlock_bh_state(bh);
			log_start_commit(journal, tid);
			log_wait_commit(journal, tid);
			goto out_return_1;
		}

		/*
		 * AKPM: I think the buffer_jbddirty test is redundant - it
		 * shouldn't have NULL b_transaction?
		 */
		next_jh = jh->b_cpnext;
		if (!buffer_dirty(bh) && !buffer_jbddirty(bh)) {
			BUFFER_TRACE(bh, "remove from checkpoint");
			__journal_remove_checkpoint(jh);
			jbd_unlock_bh_state(bh);
			journal_remove_journal_head(bh);
			__brelse(bh);
			ret = 1;
		} else {
			jbd_unlock_bh_state(bh);
		}
	} while (jh != last_jh);

	return ret;
out_return_1:
	spin_lock(&journal->j_list_lock);
	return 1;
}

/*
 * Try to add the new entry to the inline data.
 * If succeeds, return 0. If not, extended the inline dir and copied data to
 * the new created block.
 */
int ext4_try_add_inline_entry(handle_t *handle, struct dentry *dentry,
			      struct inode *inode)
{
	int ret, inline_size;
	void *inline_start, *backup_buf = NULL;
	struct buffer_head *dir_block = NULL;
	struct ext4_iloc iloc;
	int blocksize = inode->i_sb->s_blocksize;
	struct inode *dir = dentry->d_parent->d_inode;

	ret = ext4_get_inode_loc(dir, &iloc);
	if (ret)
		return ret;

	down_write(&EXT4_I(dir)->xattr_sem);
	if (!ext4_has_inline_data(dir))
		goto out;

	inline_start = ext4_raw_inode(&iloc)->i_block;
	inline_size = EXT4_MIN_INLINE_DATA_SIZE;

	ret = ext4_add_dirent_to_inline(handle, dentry, inode, &iloc,
					inline_start, inline_size);
	if (ret != -ENOSPC)
		goto out;

	/* check whether it can be inserted to inline xattr space. */
	inline_size = EXT4_I(dir)->i_inline_size -
			EXT4_MIN_INLINE_DATA_SIZE;
	if (inline_size > 0) {
		inline_start = ext4_get_inline_xattr_pos(dir, &iloc);

		ret = ext4_add_dirent_to_inline(handle, dentry, inode, &iloc,
						inline_start, inline_size);
		if (ret != -ENOSPC)
			goto out;
	}

	/* Try to add more xattr space.*/
	ret = ext4_update_inline_dir(handle, dentry, dir, &iloc);
	if (ret && ret != -ENOSPC)
		goto out;
	else if (!ret) {
		inline_size = EXT4_I(dir)->i_inline_size -
				EXT4_MIN_INLINE_DATA_SIZE;
		inline_start = ext4_get_inline_xattr_pos(dir, &iloc);

		ret = ext4_add_dirent_to_inline(handle, dentry, inode, &iloc,
						inline_start, inline_size);
		if (ret != -ENOSPC)
			goto out;
	}

	/*
	 * The inline space is filled up, so create a new block for it.
	 * As the extent tree will be created, we have to save the inline
	 * dir first.
	 */
	inline_size = EXT4_I(dir)->i_inline_size;
	backup_buf = kmalloc(inline_size, GFP_NOFS);
	if (!backup_buf) {
		ret = -ENOMEM;
		goto out;
	}

	memcpy(backup_buf, (void *)ext4_raw_inode(&iloc)->i_block,
	       EXT4_MIN_INLINE_DATA_SIZE);
	if (inline_size > EXT4_MIN_INLINE_DATA_SIZE)
		memcpy(backup_buf + EXT4_MIN_INLINE_DATA_SIZE,
		       ext4_get_inline_xattr_pos(dir, &iloc),
		       inline_size - EXT4_MIN_INLINE_DATA_SIZE);

	/* clear the entry and the flag in dir now. */
	ret = ext4_destroy_inline_data_nolock(handle, dir);
	if (ret)
		goto out;

	dir->i_size = EXT4_I(dir)->i_disksize = blocksize;
	dir_block = ext4_bread(handle, dir, 0, 1, &ret);
	if (!dir_block)
		goto out;

	BUFFER_TRACE(dir_block, "get_write_access");
	ret = ext4_journal_get_write_access(handle, dir_block);
	if (ret)
		goto out;
	memcpy(dir_block->b_data, backup_buf, inline_size);

	/* Set the final de to cover the whole block. */
	ext4_update_final_de(dir_block->b_data, inline_size,
			     blocksize);

	BUFFER_TRACE(dir_block, "call ext4_handle_dirty_metadata");
	ret = ext4_handle_dirty_metadata(handle, dir, dir_block);

//.........这里部分代码省略.........

//.........这里部分代码省略.........
		}
		if (!gdp) {
			/*
			 * That failed: try linear search for a free inode
			 */
			i = dir->u.ext3_i.i_block_group + 1;
			for (j = 2; j < sb->u.ext3_sb.s_groups_count; j++) {
				if (++i >= sb->u.ext3_sb.s_groups_count)
					i = 0;
				tmp = ext3_get_group_desc (sb, i, &bh2);
				if (tmp &&
				    le16_to_cpu(tmp->bg_free_inodes_count)) {
					gdp = tmp;
					break;
				}
			}
		}
	}

	err = -ENOSPC;
	if (!gdp)
		goto fail;

	err = -EIO;
	bitmap_nr = load_inode_bitmap (sb, i);
	if (bitmap_nr < 0)
		goto fail;

	bh = sb->u.ext3_sb.s_inode_bitmap[bitmap_nr];

	if ((j = ext3_find_first_zero_bit ((unsigned long *) bh->b_data,
				      EXT3_INODES_PER_GROUP(sb))) <
	    EXT3_INODES_PER_GROUP(sb)) {
		BUFFER_TRACE(bh, "get_write_access");
		err = ext3_journal_get_write_access(handle, bh);
		if (err) goto fail;
		
		if (ext3_set_bit (j, bh->b_data)) {
			ext3_error (sb, "ext3_new_inode",
				      "bit already set for inode %d", j);
			goto repeat;
		}
		BUFFER_TRACE(bh, "call ext3_journal_dirty_metadata");
		err = ext3_journal_dirty_metadata(handle, bh);
		if (err) goto fail;
	} else {
		if (le16_to_cpu(gdp->bg_free_inodes_count) != 0) {
			ext3_error (sb, "ext3_new_inode",
				    "Free inodes count corrupted in group %d",
				    i);
			/* Is it really ENOSPC? */
			err = -ENOSPC;
			if (sb->s_flags & MS_RDONLY)
				goto fail;

			BUFFER_TRACE(bh2, "get_write_access");
			err = ext3_journal_get_write_access(handle, bh2);
			if (err) goto fail;
			gdp->bg_free_inodes_count = 0;
			BUFFER_TRACE(bh2, "call ext3_journal_dirty_metadata");
			err = ext3_journal_dirty_metadata(handle, bh2);
			if (err) goto fail;
		}
		goto repeat;
	}
	j += i * EXT3_INODES_PER_GROUP(sb) + 1;

static int ext4_create_inline_data(handle_t *handle,
				   struct inode *inode, unsigned len)
{
	int error;
	void *value = NULL;
	struct ext4_xattr_ibody_find is = {
		.s = { .not_found = -ENODATA, },
	};
	struct ext4_xattr_info i = {
		.name_index = EXT4_XATTR_INDEX_SYSTEM,
		.name = EXT4_XATTR_SYSTEM_DATA,
	};

	error = ext4_get_inode_loc(inode, &is.iloc);
	if (error)
		return error;

	BUFFER_TRACE(is.iloc.bh, "get_write_access");
	error = ext4_journal_get_write_access(handle, is.iloc.bh);
	if (error)
		goto out;

	if (len > EXT4_MIN_INLINE_DATA_SIZE) {
		value = EXT4_ZERO_XATTR_VALUE;
		len -= EXT4_MIN_INLINE_DATA_SIZE;
	} else {
		value = "";
		len = 0;
	}

	/* Insert the the xttr entry. */
	i.value = value;
	i.value_len = len;

	error = ext4_xattr_ibody_find(inode, &i, &is);
	if (error)
		goto out;

	BUG_ON(!is.s.not_found);

	error = ext4_xattr_ibody_inline_set(handle, inode, &i, &is);
	if (error) {
		if (error == -ENOSPC)
			ext4_clear_inode_state(inode,
					       EXT4_STATE_MAY_INLINE_DATA);
		goto out;
	}

	memset((void *)ext4_raw_inode(&is.iloc)->i_block,
		0, EXT4_MIN_INLINE_DATA_SIZE);

	EXT4_I(inode)->i_inline_off = (u16)((void *)is.s.here -
				      (void *)ext4_raw_inode(&is.iloc));
	EXT4_I(inode)->i_inline_size = len + EXT4_MIN_INLINE_DATA_SIZE;
	ext4_clear_inode_flag(inode, EXT4_INODE_EXTENTS);
	ext4_set_inode_flag(inode, EXT4_INODE_INLINE_DATA);
	get_bh(is.iloc.bh);
	error = ext4_mark_iloc_dirty(handle, inode, &is.iloc);

out:
	brelse(is.iloc.bh);
	return error;
}

/*
 * 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++) {
		gdp = ext3_get_group_desc(sb, group, &bh2);

		err = -EIO;
		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)) {
			int credits = 0;

			BUFFER_TRACE(bitmap_bh, "get_write_access");
			err = ext3_journal_get_write_access_credits(handle,
							bitmap_bh, &credits);
			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, credits);

			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.........