Btrfs: Add a write ahead tree log to optimize synchronous operations

File syncs and directory syncs are optimized by copying their
items into a special (copy-on-write) log tree.  There is one log tree per
subvolume and the btrfs super block points to a tree of log tree roots.

After a crash, items are copied out of the log tree and back into the
subvolume.  See tree-log.c for all the details.

Signed-off-by: Chris Mason <chris.mason@oracle.com>

1 files changed, 2804 insertions, 0 deletions

diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c
new file mode 100644
index 00000000000..d1ce8314b94
--- /dev/null
+++ b/fs/btrfs/tree-log.c
@@ -0,0 +1,2804 @@
+/*
+ * Copyright (C) 2008 Oracle.  All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public
+ * License v2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public
+ * License along with this program; if not, write to the
+ * Free Software Foundation, Inc., 59 Temple Place - Suite 330,
+ * Boston, MA 021110-1307, USA.
+ */
+
+#include <linux/sched.h>
+#include "ctree.h"
+#include "transaction.h"
+#include "disk-io.h"
+#include "locking.h"
+#include "print-tree.h"
+#include "compat.h"
+
+/* magic values for the inode_only field in btrfs_log_inode:
+ *
+ * LOG_INODE_ALL means to log everything
+ * LOG_INODE_EXISTS means to log just enough to recreate the inode
+ * during log replay
+ */
+#define LOG_INODE_ALL 0
+#define LOG_INODE_EXISTS 1
+
+/*
+ * stages for the tree walking.  The first
+ * stage (0) is to only pin down the blocks we find
+ * the second stage (1) is to make sure that all the inodes
+ * we find in the log are created in the subvolume.
+ *
+ * The last stage is to deal with directories and links and extents
+ * and all the other fun semantics
+ */
+#define LOG_WALK_PIN_ONLY 0
+#define LOG_WALK_REPLAY_INODES 1
+#define LOG_WALK_REPLAY_ALL 2
+
+static int __btrfs_log_inode(struct btrfs_trans_handle *trans,
+			     struct btrfs_root *root, struct inode *inode,
+			     int inode_only);
+
+/*
+ * tree logging is a special write ahead log used to make sure that
+ * fsyncs and O_SYNCs can happen without doing full tree commits.
+ *
+ * Full tree commits are expensive because they require commonly
+ * modified blocks to be recowed, creating many dirty pages in the
+ * extent tree an 4x-6x higher write load than ext3.
+ *
+ * Instead of doing a tree commit on every fsync, we use the
+ * key ranges and transaction ids to find items for a given file or directory
+ * that have changed in this transaction.  Those items are copied into
+ * a special tree (one per subvolume root), that tree is written to disk
+ * and then the fsync is considered complete.
+ *
+ * After a crash, items are copied out of the log-tree back into the
+ * subvolume tree.  Any file data extents found are recorded in the extent
+ * allocation tree, and the log-tree freed.
+ *
+ * The log tree is read three times, once to pin down all the extents it is
+ * using in ram and once, once to create all the inodes logged in the tree
+ * and once to do all the other items.
+ */
+
+/*
+ * btrfs_add_log_tree adds a new per-subvolume log tree into the
+ * tree of log tree roots.  This must be called with a tree log transaction
+ * running (see start_log_trans).
+ */
+int btrfs_add_log_tree(struct btrfs_trans_handle *trans,
+		      struct btrfs_root *root)
+{
+	struct btrfs_key key;
+	struct btrfs_root_item root_item;
+	struct btrfs_inode_item *inode_item;
+	struct extent_buffer *leaf;
+	struct btrfs_root *new_root = root;
+	int ret;
+	u64 objectid = root->root_key.objectid;
+
+	leaf = btrfs_alloc_free_block(trans, root, root->leafsize,
+				      BTRFS_TREE_LOG_OBJECTID,
+				      0, 0, 0, 0, 0);
+	if (IS_ERR(leaf)) {
+		ret = PTR_ERR(leaf);
+		return ret;
+	}
+
+	btrfs_set_header_nritems(leaf, 0);
+	btrfs_set_header_level(leaf, 0);
+	btrfs_set_header_bytenr(leaf, leaf->start);
+	btrfs_set_header_generation(leaf, trans->transid);
+	btrfs_set_header_owner(leaf, BTRFS_TREE_LOG_OBJECTID);
+
+	write_extent_buffer(leaf, root->fs_info->fsid,
+			    (unsigned long)btrfs_header_fsid(leaf),
+			    BTRFS_FSID_SIZE);
+	btrfs_mark_buffer_dirty(leaf);
+
+	inode_item = &root_item.inode;
+	memset(inode_item, 0, sizeof(*inode_item));
+	inode_item->generation = cpu_to_le64(1);
+	inode_item->size = cpu_to_le64(3);
+	inode_item->nlink = cpu_to_le32(1);
+	inode_item->nblocks = cpu_to_le64(1);
+	inode_item->mode = cpu_to_le32(S_IFDIR | 0755);
+
+	btrfs_set_root_bytenr(&root_item, leaf->start);
+	btrfs_set_root_level(&root_item, 0);
+	btrfs_set_root_refs(&root_item, 0);
+	btrfs_set_root_used(&root_item, 0);
+
+	memset(&root_item.drop_progress, 0, sizeof(root_item.drop_progress));
+	root_item.drop_level = 0;
+
+	btrfs_tree_unlock(leaf);
+	free_extent_buffer(leaf);
+	leaf = NULL;
+
+	btrfs_set_root_dirid(&root_item, 0);
+
+	key.objectid = BTRFS_TREE_LOG_OBJECTID;
+	key.offset = objectid;
+	btrfs_set_key_type(&key, BTRFS_ROOT_ITEM_KEY);
+	ret = btrfs_insert_root(trans, root->fs_info->log_root_tree, &key,
+				&root_item);
+	if (ret)
+		goto fail;
+
+	new_root = btrfs_read_fs_root_no_radix(root->fs_info->log_root_tree,
+					       &key);
+	BUG_ON(!new_root);
+
+	WARN_ON(root->log_root);
+	root->log_root = new_root;
+
+	/*
+	 * log trees do not get reference counted because they go away
+	 * before a real commit is actually done.  They do store pointers
+	 * to file data extents, and those reference counts still get
+	 * updated (along with back refs to the log tree).
+	 */
+	new_root->ref_cows = 0;
+	new_root->last_trans = trans->transid;
+fail:
+	return ret;
+}
+
+/*
+ * start a sub transaction and setup the log tree
+ * this increments the log tree writer count to make the people
+ * syncing the tree wait for us to finish
+ */
+static int start_log_trans(struct btrfs_trans_handle *trans,
+			   struct btrfs_root *root)
+{
+	int ret;
+	mutex_lock(&root->fs_info->tree_log_mutex);
+	if (!root->fs_info->log_root_tree) {
+		ret = btrfs_init_log_root_tree(trans, root->fs_info);
+		BUG_ON(ret);
+	}
+	if (!root->log_root) {
+		ret = btrfs_add_log_tree(trans, root);
+		BUG_ON(ret);
+	}
+	atomic_inc(&root->fs_info->tree_log_writers);
+	root->fs_info->tree_log_batch++;
+	mutex_unlock(&root->fs_info->tree_log_mutex);
+	return 0;
+}
+
+/*
+ * returns 0 if there was a log transaction running and we were able
+ * to join, or returns -ENOENT if there were not transactions
+ * in progress
+ */
+static int join_running_log_trans(struct btrfs_root *root)
+{
+	int ret = -ENOENT;
+
+	smp_mb();
+	if (!root->log_root)
+		return -ENOENT;
+
+	mutex_lock(&root->fs_info->tree_log_mutex);
+	if (root->log_root) {
+		ret = 0;
+		atomic_inc(&root->fs_info->tree_log_writers);
+		root->fs_info->tree_log_batch++;
+	}
+	mutex_unlock(&root->fs_info->tree_log_mutex);
+	return ret;
+}
+
+/*
+ * indicate we're done making changes to the log tree
+ * and wake up anyone waiting to do a sync
+ */
+static int end_log_trans(struct btrfs_root *root)
+{
+	atomic_dec(&root->fs_info->tree_log_writers);
+	smp_mb();
+	if (waitqueue_active(&root->fs_info->tree_log_wait))
+		wake_up(&root->fs_info->tree_log_wait);
+	return 0;
+}
+
+
+/*
+ * the walk control struct is used to pass state down the chain when
+ * processing the log tree.  The stage field tells us which part
+ * of the log tree processing we are currently doing.  The others
+ * are state fields used for that specific part
+ */
+struct walk_control {
+	/* should we free the extent on disk when done?  This is used
+	 * at transaction commit time while freeing a log tree
+	 */
+	int free;
+
+	/* should we write out the extent buffer?  This is used
+	 * while flushing the log tree to disk during a sync
+	 */
+	int write;
+
+	/* should we wait for the extent buffer io to finish?  Also used
+	 * while flushing the log tree to disk for a sync
+	 */
+	int wait;
+
+	/* pin only walk, we record which extents on disk belong to the
+	 * log trees
+	 */
+	int pin;
+
+	/* what stage of the replay code we're currently in */
+	int stage;
+
+	/* the root we are currently replaying */
+	struct btrfs_root *replay_dest;
+
+	/* the trans handle for the current replay */
+	struct btrfs_trans_handle *trans;
+
+	/* the function that gets used to process blocks we find in the
+	 * tree.  Note the extent_buffer might not be up to date when it is
+	 * passed in, and it must be checked or read if you need the data
+	 * inside it
+	 */
+	int (*process_func)(struct btrfs_root *log, struct extent_buffer *eb,
+			    struct walk_control *wc, u64 gen);
+};
+
+/*
+ * process_func used to pin down extents, write them or wait on them
+ */
+static int process_one_buffer(struct btrfs_root *log,
+			      struct extent_buffer *eb,
+			      struct walk_control *wc, u64 gen)
+{
+	if (wc->pin) {
+		mutex_lock(&log->fs_info->alloc_mutex);
+		btrfs_update_pinned_extents(log->fs_info->extent_root,
+					    eb->start, eb->len, 1);
+		mutex_unlock(&log->fs_info->alloc_mutex);
+	}
+
+	if (btrfs_buffer_uptodate(eb, gen)) {
+		if (wc->write)
+			btrfs_write_tree_block(eb);
+		if (wc->wait)
+			btrfs_wait_tree_block_writeback(eb);
+	}
+	return 0;
+}
+
+/*
+ * Item overwrite used by replay and tree logging.  eb, slot and key all refer
+ * to the src data we are copying out.
+ *
+ * root is the tree we are copying into, and path is a scratch
+ * path for use in this function (it should be released on entry and
+ * will be released on exit).
+ *
+ * If the key is already in the destination tree the existing item is
+ * overwritten.  If the existing item isn't big enough, it is extended.
+ * If it is too large, it is truncated.
+ *
+ * If the key isn't in the destination yet, a new item is inserted.
+ */
+static noinline int overwrite_item(struct btrfs_trans_handle *trans,
+				   struct btrfs_root *root,
+				   struct btrfs_path *path,
+				   struct extent_buffer *eb, int slot,
+				   struct btrfs_key *key)
+{
+	int ret;
+	u32 item_size;
+	u64 saved_i_size = 0;
+	int save_old_i_size = 0;
+	unsigned long src_ptr;
+	unsigned long dst_ptr;
+	int overwrite_root = 0;
+
+	if (root->root_key.objectid != BTRFS_TREE_LOG_OBJECTID)
+		overwrite_root = 1;
+
+	item_size = btrfs_item_size_nr(eb, slot);
+	src_ptr = btrfs_item_ptr_offset(eb, slot);
+
+	/* look for the key in the destination tree */
+	ret = btrfs_search_slot(NULL, root, key, path, 0, 0);
+	if (ret == 0) {
+		char *src_copy;
+		char *dst_copy;
+		u32 dst_size = btrfs_item_size_nr(path->nodes[0],
+						  path->slots[0]);
+		if (dst_size != item_size)
+			goto insert;
+
+		if (item_size == 0) {
+			btrfs_release_path(root, path);
+			return 0;
+		}
+		dst_copy = kmalloc(item_size, GFP_NOFS);
+		src_copy = kmalloc(item_size, GFP_NOFS);
+
+		read_extent_buffer(eb, src_copy, src_ptr, item_size);
+
+		dst_ptr = btrfs_item_ptr_offset(path->nodes[0], path->slots[0]);
+		read_extent_buffer(path->nodes[0], dst_copy, dst_ptr,
+				   item_size);
+		ret = memcmp(dst_copy, src_copy, item_size);
+
+		kfree(dst_copy);
+		kfree(src_copy);
+		/*
+		 * they have the same contents, just return, this saves
+		 * us from cowing blocks in the destination tree and doing
+		 * extra writes that may not have been done by a previous
+		 * sync
+		 */
+		if (ret == 0) {
+			btrfs_release_path(root, path);
+			return 0;
+		}
+
+	}
+insert:
+	btrfs_release_path(root, path);
+	/* try to insert the key into the destination tree */
+	ret = btrfs_insert_empty_item(trans, root, path,
+				      key, item_size);
+
+	/* make sure any existing item is the correct size */
+	if (ret == -EEXIST) {
+		u32 found_size;
+		found_size = btrfs_item_size_nr(path->nodes[0],
+						path->slots[0]);
+		if (found_size > item_size) {
+			btrfs_truncate_item(trans, root, path, item_size, 1);
+		} else if (found_size < item_size) {
+			ret = btrfs_del_item(trans, root,
+					     path);
+			BUG_ON(ret);
+
+			btrfs_release_path(root, path);
+			ret = btrfs_insert_empty_item(trans,
+				  root, path, key, item_size);
+			BUG_ON(ret);
+		}
+	} else if (ret) {
+		BUG();
+	}
+	dst_ptr = btrfs_item_ptr_offset(path->nodes[0],
+					path->slots[0]);
+
+	/* don't overwrite an existing inode if the generation number
+	 * was logged as zero.  This is done when the tree logging code
+	 * is just logging an inode to make sure it exists after recovery.
+	 *
+	 * Also, don't overwrite i_size on directories during replay.
+	 * log replay inserts and removes directory items based on the
+	 * state of the tree found in the subvolume, and i_size is modified
+	 * as it goes
+	 */
+	if (key->type == BTRFS_INODE_ITEM_KEY && ret == -EEXIST) {
+		struct btrfs_inode_item *src_item;
+		struct btrfs_inode_item *dst_item;
+
+		src_item = (struct btrfs_inode_item *)src_ptr;
+		dst_item = (struct btrfs_inode_item *)dst_ptr;
+
+		if (btrfs_inode_generation(eb, src_item) == 0)
+			goto no_copy;
+
+		if (overwrite_root &&
+		    S_ISDIR(btrfs_inode_mode(eb, src_item)) &&
+		    S_ISDIR(btrfs_inode_mode(path->nodes[0], dst_item))) {
+			save_old_i_size = 1;
+			saved_i_size = btrfs_inode_size(path->nodes[0],
+							dst_item);
+		}
+	}
+
+	copy_extent_buffer(path->nodes[0], eb, dst_ptr,
+			   src_ptr, item_size);
+
+	if (save_old_i_size) {
+		struct btrfs_inode_item *dst_item;
+		dst_item = (struct btrfs_inode_item *)dst_ptr;
+		btrfs_set_inode_size(path->nodes[0], dst_item, saved_i_size);
+	}
+
+	/* make sure the generation is filled in */
+	if (key->type == BTRFS_INODE_ITEM_KEY) {
+		struct btrfs_inode_item *dst_item;
+		dst_item = (struct btrfs_inode_item *)dst_ptr;
+		if (btrfs_inode_generation(path->nodes[0], dst_item) == 0) {
+			btrfs_set_inode_generation(path->nodes[0], dst_item,
+						   trans->transid);
+		}
+	}
+no_copy:
+	btrfs_mark_buffer_dirty(path->nodes[0]);
+	btrfs_release_path(root, path);
+	return 0;
+}
+
+/*
+ * simple helper to read an inode off the disk from a given root
+ * This can only be called for subvolume roots and not for the log
+ */
+static noinline struct inode *read_one_inode(struct btrfs_root *root,
+					     u64 objectid)
+{
+	struct inode *inode;
+	inode = btrfs_iget_locked(root->fs_info->sb, objectid, root);
+	if (inode->i_state & I_NEW) {
+		BTRFS_I(inode)->root = root;
+		BTRFS_I(inode)->location.objectid = objectid;
+		BTRFS_I(inode)->location.type = BTRFS_INODE_ITEM_KEY;
+		BTRFS_I(inode)->location.offset = 0;
+		btrfs_read_locked_inode(inode);
+		unlock_new_inode(inode);
+
+	}
+	if (is_bad_inode(inode)) {
+		iput(inode);
+		inode = NULL;
+	}
+	return inode;
+}
+
+/* replays a single extent in 'eb' at 'slot' with 'key' into the
+ * subvolume 'root'.  path is released on entry and should be released
+ * on exit.
+ *
+ * extents in the log tree have not been allocated out of the extent
+ * tree yet.  So, this completes the allocation, taking a reference
+ * as required if the extent already exists or creating a new extent
+ * if it isn't in the extent allocation tree yet.
+ *
+ * The extent is inserted into the file, dropping any existing extents
+ * from the file that overlap the new one.
+ */
+static noinline int replay_one_extent(struct btrfs_trans_handle *trans,
+				      struct btrfs_root *root,
+				      struct btrfs_path *path,
+				      struct extent_buffer *eb, int slot,
+				      struct btrfs_key *key)
+{
+	int found_type;
+	u64 mask = root->sectorsize - 1;
+	u64 extent_end;
+	u64 alloc_hint;
+	u64 start = key->offset;
+	struct btrfs_file_extent_item *item;
+	struct inode *inode = NULL;
+	unsigned long size;
+	int ret = 0;
+
+	item = btrfs_item_ptr(eb, slot, struct btrfs_file_extent_item);
+	found_type = btrfs_file_extent_type(eb, item);
+
+	if (found_type == BTRFS_FILE_EXTENT_REG)
+		extent_end = start + btrfs_file_extent_num_bytes(eb, item);
+	else if (found_type == BTRFS_FILE_EXTENT_INLINE) {
+		size = btrfs_file_extent_inline_len(eb,
+						    btrfs_item_nr(eb, slot));
+		extent_end = (start + size + mask) & ~mask;
+	} else {
+		ret = 0;
+		goto out;
+	}
+
+	inode = read_one_inode(root, key->objectid);
+	if (!inode) {
+		ret = -EIO;
+		goto out;
+	}
+
+	/*
+	 * first check to see if we already have this extent in the
+	 * file.  This must be done before the btrfs_drop_extents run
+	 * so we don't try to drop this extent.
+	 */
+	ret = btrfs_lookup_file_extent(trans, root, path, inode->i_ino,
+				       start, 0);
+
+	if (ret == 0 && found_type == BTRFS_FILE_EXTENT_REG) {
+		struct btrfs_file_extent_item cmp1;
+		struct btrfs_file_extent_item cmp2;
+		struct btrfs_file_extent_item *existing;
+		struct extent_buffer *leaf;
+
+		leaf = path->nodes[0];
+		existing = btrfs_item_ptr(leaf, path->slots[0],
+					  struct btrfs_file_extent_item);
+
+		read_extent_buffer(eb, &cmp1, (unsigned long)item,
+				   sizeof(cmp1));
+		read_extent_buffer(leaf, &cmp2, (unsigned long)existing,
+				   sizeof(cmp2));
+
+		/*
+		 * we already have a pointer to this exact extent,
+		 * we don't have to do anything
+		 */
+		if (memcmp(&cmp1, &cmp2, sizeof(cmp1)) == 0) {
+			btrfs_release_path(root, path);
+			goto out;
+		}
+	}
+	btrfs_release_path(root, path);
+
+	/* drop any overlapping extents */
+	ret = btrfs_drop_extents(trans, root, inode,
+			 start, extent_end, start, &alloc_hint);
+	BUG_ON(ret);
+
+	BUG_ON(ret);
+	if (found_type == BTRFS_FILE_EXTENT_REG) {
+		struct btrfs_key ins;
+
+		ins.objectid = btrfs_file_extent_disk_bytenr(eb, item);
+		ins.offset = btrfs_file_extent_disk_num_bytes(eb, item);
+		ins.type = BTRFS_EXTENT_ITEM_KEY;
+
+		/* insert the extent pointer in the file */
+		ret = overwrite_item(trans, root, path, eb, slot, key);
+		BUG_ON(ret);
+
+		/*
+		 * is this extent already allocated in the extent
+		 * allocation tree?  If so, just add a reference
+		 */
+		ret = btrfs_lookup_extent(root, path, ins.objectid, ins.offset);
+		btrfs_release_path(root, path);
+		if (ret == 0) {
+			ret = btrfs_inc_extent_ref(trans, root,
+				   ins.objectid, ins.offset,
+				   root->root_key.objectid,
+				   trans->transid, key->objectid, start);
+		} else {
+			/*
+			 * insert the extent pointer in the extent
+			 * allocation tree
+			 */
+			ret = btrfs_alloc_logged_extent(trans, root,
+						root->root_key.objectid,
+						trans->transid, key->objectid,
+						start, &ins);
+			BUG_ON(ret);
+		}
+	} else if (found_type == BTRFS_FILE_EXTENT_INLINE) {
+		/* inline extents are easy, we just overwrite them */
+		ret = overwrite_item(trans, root, path, eb, slot, key);
+		BUG_ON(ret);
+	}
+	/* btrfs_drop_extents changes i_blocks, update it here */
+	inode->i_blocks += (extent_end - start) >> 9;
+	btrfs_update_inode(trans, root, inode);
+out:
+	if (inode)
+		iput(inode);
+	return ret;
+}
+
+/*
+ * when cleaning up conflicts between the directory names in the
+ * subvolume, directory names in the log and directory names in the
+ * inode back references, we may have to unlink inodes from directories.
+ *
+ * This is a helper function to do the unlink of a specific directory
+ * item
+ */
+static noinline int drop_one_dir_item(struct btrfs_trans_handle *trans,
+				      struct btrfs_root *root,
+				      struct btrfs_path *path,
+				      struct inode *dir,
+				      struct btrfs_dir_item *di)
+{
+	struct inode *inode;
+	char *name;
+	int name_len;
+	struct extent_buffer *leaf;
+	struct btrfs_key location;
+	int ret;
+
+	leaf = path->nodes[0];
+
+	btrfs_dir_item_key_to_cpu(leaf, di, &location);
+	name_len = btrfs_dir_name_len(leaf, di);
+	name = kmalloc(name_len, GFP_NOFS);
+	read_extent_buffer(leaf, name, (unsigned long)(di + 1), name_len);
+	btrfs_release_path(root, path);
+
+	inode = read_one_inode(root, location.objectid);
+	BUG_ON(!inode);
+
+	btrfs_inc_nlink(inode);
+	ret = btrfs_unlink_inode(trans, root, dir, inode, name, name_len);
+	kfree(name);
+
+	iput(inode);
+	return ret;
+}
+
+/*
+ * helper function to see if a given name and sequence number found
+ * in an inode back reference are already in a directory and correctly
+ * point to this inode
+ */
+static noinline int inode_in_dir(struct btrfs_root *root,
+				 struct btrfs_path *path,
+				 u64 dirid, u64 objectid, u64 index,
+				 const char *name, int name_len)
+{
+	struct btrfs_dir_item *di;
+	struct btrfs_key location;
+	int match = 0;
+
+	di = btrfs_lookup_dir_index_item(NULL, root, path, dirid,
+					 index, name, name_len, 0);
+	if (di && !IS_ERR(di)) {
+		btrfs_dir_item_key_to_cpu(path->nodes[0], di, &location);
+		if (location.objectid != objectid)
+			goto out;
+	} else
+		goto out;
+	btrfs_release_path(root, path);
+
+	di = btrfs_lookup_dir_item(NULL, root, path, dirid, name, name_len, 0);
+	if (di && !IS_ERR(di)) {
+		btrfs_dir_item_key_to_cpu(path->nodes[0], di, &location);
+		if (location.objectid != objectid)
+			goto out;
+	} else
+		goto out;
+	match = 1;
+out:
+	btrfs_release_path(root, path);
+	return match;
+}
+
+/*
+ * helper function to check a log tree for a named back reference in
+ * an inode.  This is used to decide if a back reference that is
+ * found in the subvolume conflicts with what we find in the log.
+ *
+ * inode backreferences may have multiple refs in a single item,
+ * during replay we process one reference at a time, and we don't
+ * want to delete valid links to a file from the subvolume if that
+ * link is also in the log.
+ */
+static noinline int backref_in_log(struct btrfs_root *log,
+				   struct btrfs_key *key,
+				   char *name, int namelen)
+{
+	struct btrfs_path *path;
+	struct btrfs_inode_ref *ref;
+	unsigned long ptr;
+	unsigned long ptr_end;
+	unsigned long name_ptr;
+	int found_name_len;
+	int item_size;
+	int ret;
+	int match = 0;
+
+	path = btrfs_alloc_path();
+	ret = btrfs_search_slot(NULL, log, key, path, 0, 0);
+	if (ret != 0)
+		goto out;
+
+	item_size = btrfs_item_size_nr(path->nodes[0], path->slots[0]);
+	ptr = btrfs_item_ptr_offset(path->nodes[0], path->slots[0]);
+	ptr_end = ptr + item_size;
+	while (ptr < ptr_end) {
+		ref = (struct btrfs_inode_ref *)ptr;
+		found_name_len = btrfs_inode_ref_name_len(path->nodes[0], ref);
+		if (found_name_len == namelen) {
+			name_ptr = (unsigned long)(ref + 1);
+			ret = memcmp_extent_buffer(path->nodes[0], name,
+						   name_ptr, namelen);
+			if (ret == 0) {
+				match = 1;
+				goto out;
+			}
+		}
+		ptr = (unsigned long)(ref + 1) + found_name_len;
+	}
+out:
+	btrfs_free_path(path);
+	return match;
+}
+
+
+/*
+ * replay one inode back reference item found in the log tree.
+ * eb, slot and key refer to the buffer and key found in the log tree.
+ * root is the destination we are replaying into, and path is for temp
+ * use by this function.  (it should be released on return).
+ */
+static noinline int add_inode_ref(struct btrfs_trans_handle *trans,
+				  struct btrfs_root *root,
+				  struct btrfs_root *log,
+				  struct btrfs_path *path,
+				  struct extent_buffer *eb, int slot,
+				  struct btrfs_key *key)
+{
+	struct inode *dir;
+	int ret;
+	struct btrfs_key location;
+	struct btrfs_inode_ref *ref;
+	struct btrfs_dir_item *di;
+	struct inode *inode;
+	char *name;
+	int namelen;
+	unsigned long ref_ptr;
+	unsigned long ref_end;
+
+	location.objectid = key->objectid;
+	location.type = BTRFS_INODE_ITEM_KEY;
+	location.offset = 0;
+
+	/*
+	 * it is possible that we didn't log all the parent directories
+	 * for a given inode.  If we don't find the dir, just don't
+	 * copy the back ref in.  The link count fixup code will take
+	 * care of the rest
+	 */
+	dir = read_one_inode(root, key->offset);
+	if (!dir)
+		return -ENOENT;
+
+	inode = read_one_inode(root, key->objectid);
+	BUG_ON(!dir);
+
+	ref_ptr = btrfs_item_ptr_offset(eb, slot);
+	ref_end = ref_ptr + btrfs_item_size_nr(eb, slot);
+
+again:
+	ref = (struct btrfs_inode_ref *)ref_ptr;
+
+	namelen = btrfs_inode_ref_name_len(eb, ref);
+	name = kmalloc(namelen, GFP_NOFS);
+	BUG_ON(!name);
+
+	read_extent_buffer(eb, name, (unsigned long)(ref + 1), namelen);
+
+	/* if we already have a perfect match, we're done */
+	if (inode_in_dir(root, path, dir->i_ino, inode->i_ino,
+			 btrfs_inode_ref_index(eb, ref),
+			 name, namelen)) {
+		goto out;
+	}
+
+	/*
+	 * look for a conflicting back reference in the metadata.
+	 * if we find one we have to unlink that name of the file
+	 * before we add our new link.  Later on, we overwrite any
+	 * existing back reference, and we don't want to create
+	 * dangling pointers in the directory.
+	 */
+conflict_again:
+	ret = btrfs_search_slot(NULL, root, key, path, 0, 0);
+	if (ret == 0) {
+		char *victim_name;
+		int victim_name_len;
+		struct btrfs_inode_ref *victim_ref;
+		unsigned long ptr;
+		unsigned long ptr_end;
+		struct extent_buffer *leaf = path->nodes[0];
+
+		/* are we trying to overwrite a back ref for the root directory
+		 * if so, just jump out, we're done
+		 */
+		if (key->objectid == key->offset)
+			goto out_nowrite;
+
+		/* check all the names in this back reference to see
+		 * if they are in the log.  if so, we allow them to stay
+		 * otherwise they must be unlinked as a conflict
+		 */
+		ptr = btrfs_item_ptr_offset(leaf, path->slots[0]);
+		ptr_end = ptr + btrfs_item_size_nr(leaf, path->slots[0]);
+		while(ptr < ptr_end) {
+			victim_ref = (struct btrfs_inode_ref *)ptr;
+			victim_name_len = btrfs_inode_ref_name_len(leaf,
+								   victim_ref);
+			victim_name = kmalloc(victim_name_len, GFP_NOFS);
+			BUG_ON(!victim_name);
+
+			read_extent_buffer(leaf, victim_name,
+					   (unsigned long)(victim_ref + 1),
+					   victim_name_len);
+
+			if (!backref_in_log(log, key, victim_name,
+					    victim_name_len)) {
+				btrfs_inc_nlink(inode);
+				btrfs_release_path(root, path);
+				ret = btrfs_unlink_inode(trans, root, dir,
+							 inode, victim_name,
+							 victim_name_len);
+				kfree(victim_name);
+				btrfs_release_path(root, path);
+				goto conflict_again;
+			}
+			kfree(victim_name);
+			ptr = (unsigned long)(victim_ref + 1) + victim_name_len;
+		}
+		BUG_ON(ret);
+	}
+	btrfs_release_path(root, path);
+
+	/* look for a conflicting sequence number */
+	di = btrfs_lookup_dir_index_item(trans, root, path, dir->i_ino,
+					 btrfs_inode_ref_index(eb, ref),
+					 name, namelen, 0);
+	if (di && !IS_ERR(di)) {
+		ret = drop_one_dir_item(trans, root, path, dir, di);
+		BUG_ON(ret);
+	}
+	btrfs_release_path(root, path);
+
+
+	/* look for a conflicting name */
+	di = btrfs_lookup_dir_item(trans, root, path, dir->i_ino,
+				   name, namelen, 0);
+	if (di && !IS_ERR(di)) {
+		ret = drop_one_dir_item(trans, root, path, dir, di);
+		BUG_ON(ret);
+	}
+	btrfs_release_path(root, path);
+
+	/* insert our name */
+	ret = btrfs_add_link(trans, dir, inode, name, namelen, 0,
+			     btrfs_inode_ref_index(eb, ref));
+	BUG_ON(ret);
+
+	btrfs_update_inode(trans, root, inode);
+
+out:
+	ref_ptr = (unsigned long)(ref + 1) + namelen;
+	kfree(name);
+	if (ref_ptr < ref_end)
+		goto again;
+
+	/* finally write the back reference in the inode */
+	ret = overwrite_item(trans, root, path, eb, slot, key);
+	BUG_ON(ret);
+
+out_nowrite:
+	btrfs_release_path(root, path);
+	iput(dir);
+	iput(inode);
+	return 0;
+}
+
+/*
+ * replay one csum item from the log tree into the subvolume 'root'
+ * eb, slot and key all refer to the log tree
+ * path is for temp use by this function and should be released on return
+ *
+ * This copies the checksums out of the log tree and inserts them into
+ * the subvolume.  Any existing checksums for this range in the file
+ * are overwritten, and new items are added where required.
+ *
+ * We keep this simple by reusing the btrfs_ordered_sum code from
+ * the data=ordered mode.  This basically means making a copy
+ * of all the checksums in ram, which we have to do anyway for kmap
+ * rules.
+ *
+ * The copy is then sent down to btrfs_csum_file_blocks, which
+ * does all the hard work of finding existing items in the file
+ * or adding new ones.
+ */
+static noinline int replay_one_csum(struct btrfs_trans_handle *trans,
+				      struct btrfs_root *root,
+				      struct btrfs_path *path,
+				      struct extent_buffer *eb, int slot,
+				      struct btrfs_key *key)
+{
+	int ret;
+	u32 item_size = btrfs_item_size_nr(eb, slot);
+	u64 cur_offset;
+	unsigned long file_bytes;
+	struct btrfs_ordered_sum *sums;
+	struct btrfs_sector_sum *sector_sum;
+	struct inode *inode;
+	unsigned long ptr;
+
+	file_bytes = (item_size / BTRFS_CRC32_SIZE) * root->sectorsize;
+	inode = read_one_inode(root, key->objectid);
+	if (!inode) {
+		return -EIO;
+	}
+
+	sums = kzalloc(btrfs_ordered_sum_size(root, file_bytes), GFP_NOFS);
+	if (!sums) {
+		iput(inode);
+		return -ENOMEM;
+	}
+
+	INIT_LIST_HEAD(&sums->list);
+	sums->len = file_bytes;
+	sums->file_offset = key->offset;
+
+	/*
+	 * copy all the sums into the ordered sum struct
+	 */
+	sector_sum = sums->sums;
+	cur_offset = key->offset;
+	ptr = btrfs_item_ptr_offset(eb, slot);
+	while(item_size > 0) {
+		sector_sum->offset = cur_offset;
+		read_extent_buffer(eb, &sector_sum->sum, ptr, BTRFS_CRC32_SIZE);
+		sector_sum++;
+		item_size -= BTRFS_CRC32_SIZE;
+		ptr += BTRFS_CRC32_SIZE;
+		cur_offset += root->sectorsize;
+	}
+
+	/* let btrfs_csum_file_blocks add them into the file */
+	ret = btrfs_csum_file_blocks(trans, root, inode, sums);
+	BUG_ON(ret);
+	kfree(sums);
+	iput(inode);
+
+	return 0;
+}
+/*
+ * There are a few corners where the link count of the file can't
+ * be properly maintained during replay.  So, instead of adding
+ * lots of complexity to the log code, we just scan the backrefs
+ * for any file that has been through replay.
+ *
+ * The scan will update the link count on the inode to reflect the
+ * number of back refs found.  If it goes down to zero, the iput
+ * will free the inode.
+ */
+static noinline int fixup_inode_link_count(struct btrfs_trans_handle *trans,
+					   struct btrfs_root *root,
+					   struct inode *inode)
+{
+	struct btrfs_path *path;
+	int ret;
+	struct btrfs_key key;
+	u64 nlink = 0;
+	unsigned long ptr;
+	unsigned long ptr_end;
+	int name_len;
+
+	key.objectid = inode->i_ino;
+	key.type = BTRFS_INODE_REF_KEY;
+	key.offset = (u64)-1;
+
+	path = btrfs_alloc_path();
+
+	while(1) {
+		ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
+		if (ret < 0)
+			break;
+		if (ret > 0) {
+			if (path->slots[0] == 0)
+				break;
+			path->slots[0]--;
+		}
+		btrfs_item_key_to_cpu(path->nodes[0], &key,
+				      path->slots[0]);
+		if (key.objectid != inode->i_ino ||
+		    key.type != BTRFS_INODE_REF_KEY)
+			break;
+		ptr = btrfs_item_ptr_offset(path->nodes[0], path->slots[0]);
+		ptr_end = ptr + btrfs_item_size_nr(path->nodes[0],
+						   path->slots[0]);
+		while(ptr < ptr_end) {
+			struct btrfs_inode_ref *ref;
+
+			ref = (struct btrfs_inode_ref *)ptr;
+			name_len = btrfs_inode_ref_name_len(path->nodes[0],
+							    ref);
+			ptr = (unsigned long)(ref + 1) + name_len;
+			nlink++;
+		}
+
+		if (key.offset == 0)
+			break;
+		key.offset--;
+		btrfs_release_path(root, path);
+	}
+	btrfs_free_path(path);
+	if (nlink != inode->i_nlink) {
+		inode->i_nlink = nlink;
+		btrfs_update_inode(trans, root, inode);
+	}
+
+	return 0;
+}
+
+static noinline int fixup_inode_link_counts(struct btrfs_trans_handle *trans,
+					    struct btrfs_root *root,
+					    struct btrfs_path *path)
+{
+	int ret;
+	struct btrfs_key key;
+	struct inode *inode;
+
+	key.objectid = BTRFS_TREE_LOG_FIXUP_OBJECTID;
+	key.type = BTRFS_ORPHAN_ITEM_KEY;
+	key.offset = (u64)-1;
+	while(1) {
+		ret = btrfs_search_slot(trans, root, &key, path, -1, 1);
+		if (ret < 0)
+			break;
+
+		if (ret == 1) {
+			if (path->slots[0] == 0)
+				break;
+			path->slots[0]--;
+		}
+
+		btrfs_item_key_to_cpu(path->nodes[0], &key, path->slots[0]);
+		if (key.objectid != BTRFS_TREE_LOG_FIXUP_OBJECTID ||
+		    key.type != BTRFS_ORPHAN_ITEM_KEY)
+			break;
+
+		ret = btrfs_del_item(trans, root, path);
+		BUG_ON(ret);
+
+		btrfs_release_path(root, path);
+		inode = read_one_inode(root, key.offset);
+		BUG_ON(!inode);
+
+		ret = fixup_inode_link_count(trans, root, inode);
+		BUG_ON(ret);
+
+		iput(inode);
+
+		if (key.offset == 0)
+			break;
+		key.offset--;
+	}
+	btrfs_release_path(root, path);
+	return 0;
+}
+
+
+/*
+ * record a given inode in the fixup dir so we can check its link
+ * count when replay is done.  The link count is incremented here
+ * so the inode won't go away until we check it
+ */
+static noinline int link_to_fixup_dir(struct btrfs_trans_handle *trans,
+				      struct btrfs_root *root,
+				      struct btrfs_path *path,
+				      u64 objectid)
+{
+	struct btrfs_key key;
+	int ret = 0;
+	struct inode *inode;
+
+	inode = read_one_inode(root, objectid);
+	BUG_ON(!inode);
+
+	key.objectid = BTRFS_TREE_LOG_FIXUP_OBJECTID;
+	btrfs_set_key_type(&key, BTRFS_ORPHAN_ITEM_KEY);
+	key.offset = objectid;
+
+	ret = btrfs_insert_empty_item(trans, root, path, &key, 0);
+
+	btrfs_release_path(root, path);
+	if (ret == 0) {
+		btrfs_inc_nlink(inode);
+		btrfs_update_inode(trans, root, inode);
+	} else if (ret == -EEXIST) {
+		ret = 0;
+	} else {
+		BUG();
+	}
+	iput(inode);
+
+	return ret;
+}
+
+/*
+ * when replaying the log for a directory, we only insert names
+ * for inodes that actually exist.  This means an fsync on a directory
+ * does not implicitly fsync all the new files in it
+ */
+static noinline int insert_one_name(struct btrfs_tra