Merge branch 'delayed_inode' into inode_numbers

Conflicts:
	fs/btrfs/inode.c
	fs/btrfs/ioctl.c
	fs/btrfs/transaction.c
Signed-off-by: Chris Mason <chris.mason@oracle.com>

fs/btrfs/Makefile

@@ -7,4 +7,4 @@ btrfs-y += super.o ctree.o extent-tree.o print-tree.o root-tree.o dir-item.o \
 	   extent_map.o sysfs.o struct-funcs.o xattr.o ordered-data.o \
 	   extent_io.o volumes.o async-thread.o ioctl.o locking.o orphan.o \
 	   export.o tree-log.o acl.o free-space-cache.o zlib.o lzo.o \
-	   compression.o delayed-ref.o relocation.o
+	   compression.o delayed-ref.o relocation.o delayed-inode.o

fs/btrfs/btrfs_inode.h

@@ -22,6 +22,7 @@
 #include "extent_map.h"
 #include "extent_io.h"
 #include "ordered-data.h"
+#include "delayed-inode.h"
 
 /* in memory btrfs inode */
 struct btrfs_inode {
@@ -158,9 +159,13 @@ struct btrfs_inode {
 	 */
 	unsigned force_compress:4;
 
+	struct btrfs_delayed_node *delayed_node;
+
 	struct inode vfs_inode;
 };
 
+extern unsigned char btrfs_filetype_table[];
+
 static inline struct btrfs_inode *BTRFS_I(struct inode *inode)
 {
 	return container_of(inode, struct btrfs_inode, vfs_inode);

fs/btrfs/ctree.c

@@ -38,11 +38,6 @@ static int balance_node_right(struct btrfs_trans_handle *trans,
 			      struct extent_buffer *src_buf);
 static int del_ptr(struct btrfs_trans_handle *trans, struct btrfs_root *root,
 		   struct btrfs_path *path, int level, int slot);
-static int setup_items_for_insert(struct btrfs_trans_handle *trans,
-			struct btrfs_root *root, struct btrfs_path *path,
-			struct btrfs_key *cpu_key, u32 *data_size,
-			u32 total_data, u32 total_size, int nr);
-
 
 struct btrfs_path *btrfs_alloc_path(void)
 {
@@ -3559,11 +3554,10 @@ int btrfs_insert_some_items(struct btrfs_trans_handle *trans,
  * to save stack depth by doing the bulk of the work in a function
  * that doesn't call btrfs_search_slot
  */
-static noinline_for_stack int
-setup_items_for_insert(struct btrfs_trans_handle *trans,
-		      struct btrfs_root *root, struct btrfs_path *path,
-		      struct btrfs_key *cpu_key, u32 *data_size,
-		      u32 total_data, u32 total_size, int nr)
+int setup_items_for_insert(struct btrfs_trans_handle *trans,
+			   struct btrfs_root *root, struct btrfs_path *path,
+			   struct btrfs_key *cpu_key, u32 *data_size,
+			   u32 total_data, u32 total_size, int nr)
 {
 	struct btrfs_item *item;
 	int i;

fs/btrfs/ctree.h

@@ -870,6 +870,7 @@ struct btrfs_block_group_cache {
 struct reloc_control;
 struct btrfs_device;
 struct btrfs_fs_devices;
+struct btrfs_delayed_root;
 struct btrfs_fs_info {
 	u8 fsid[BTRFS_FSID_SIZE];
 	u8 chunk_tree_uuid[BTRFS_UUID_SIZE];
@@ -896,7 +897,10 @@ struct btrfs_fs_info {
 	/* logical->physical extent mapping */
 	struct btrfs_mapping_tree mapping_tree;
 
-	/* block reservation for extent, checksum and root tree */
+	/*
+	 * block reservation for extent, checksum, root tree and
+	 * delayed dir index item
+	 */
 	struct btrfs_block_rsv global_block_rsv;
 	/* block reservation for delay allocation */
 	struct btrfs_block_rsv delalloc_block_rsv;
@@ -1023,6 +1027,7 @@ struct btrfs_fs_info {
 	 * for the sys_munmap function call path
 	 */
 	struct btrfs_workers fixup_workers;
+	struct btrfs_workers delayed_workers;
 	struct task_struct *transaction_kthread;
 	struct task_struct *cleaner_kthread;
 	int thread_pool_size;
@@ -1080,6 +1085,8 @@ struct btrfs_fs_info {
 
 	/* filesystem state */
 	u64 fs_state;
+
+	struct btrfs_delayed_root *delayed_root;
 };
 
 /*
@@ -1172,6 +1179,11 @@ struct btrfs_root {
 	/* red-black tree that keeps track of in-memory inodes */
 	struct rb_root inode_tree;
 
+	/*
+	 * radix tree that keeps track of delayed nodes of every inode,
+	 * protected by inode_lock
+	 */
+	struct radix_tree_root delayed_nodes_tree;
 	/*
 	 * right now this just gets used so that a root has its own devid
 	 * for stat.  It may be used for more later
@@ -2110,6 +2122,13 @@ static inline bool btrfs_mixed_space_info(struct btrfs_space_info *space_info)
 }
 
 /* extent-tree.c */
+static inline u64 btrfs_calc_trans_metadata_size(struct btrfs_root *root,
+						 int num_items)
+{
+	return (root->leafsize + root->nodesize * (BTRFS_MAX_LEVEL - 1)) *
+		3 * num_items;
+}
+
 void btrfs_put_block_group(struct btrfs_block_group_cache *cache);
 int btrfs_run_delayed_refs(struct btrfs_trans_handle *trans,
 			   struct btrfs_root *root, unsigned long count);
@@ -2305,6 +2324,8 @@ void btrfs_release_path(struct btrfs_root *root, struct btrfs_path *p);
 struct btrfs_path *btrfs_alloc_path(void);
 void btrfs_free_path(struct btrfs_path *p);
 void btrfs_set_path_blocking(struct btrfs_path *p);
+void btrfs_clear_path_blocking(struct btrfs_path *p,
+			       struct extent_buffer *held);
 void btrfs_unlock_up_safe(struct btrfs_path *p, int level);
 
 int btrfs_del_items(struct btrfs_trans_handle *trans, struct btrfs_root *root,
@@ -2316,6 +2337,10 @@ static inline int btrfs_del_item(struct btrfs_trans_handle *trans,
 	return btrfs_del_items(trans, root, path, path->slots[0], 1);
 }
 
+int setup_items_for_insert(struct btrfs_trans_handle *trans,
+			   struct btrfs_root *root, struct btrfs_path *path,
+			   struct btrfs_key *cpu_key, u32 *data_size,
+			   u32 total_data, u32 total_size, int nr);
 int btrfs_insert_item(struct btrfs_trans_handle *trans, struct btrfs_root
 		      *root, struct btrfs_key *key, void *data, u32 data_size);
 int btrfs_insert_some_items(struct btrfs_trans_handle *trans,
@@ -2379,7 +2404,7 @@ void btrfs_check_and_init_root_item(struct btrfs_root_item *item);
 /* dir-item.c */
 int btrfs_insert_dir_item(struct btrfs_trans_handle *trans,
 			  struct btrfs_root *root, const char *name,
-			  int name_len, u64 dir,
+			  int name_len, struct inode *dir,
 			  struct btrfs_key *location, u8 type, u64 index);
 struct btrfs_dir_item *btrfs_lookup_dir_item(struct btrfs_trans_handle *trans,
 					     struct btrfs_root *root,

fs/btrfs/delayed-inode.h

+/*
+ * Copyright (C) 2011 Fujitsu.  All rights reserved.
+ * Written by Miao Xie <miaox@cn.fujitsu.com>
+ *
+ * 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.
+ */
+
+#ifndef __DELAYED_TREE_OPERATION_H
+#define __DELAYED_TREE_OPERATION_H
+
+#include <linux/rbtree.h>
+#include <linux/spinlock.h>
+#include <linux/mutex.h>
+#include <linux/list.h>
+#include <linux/wait.h>
+#include <asm/atomic.h>
+
+#include "ctree.h"
+
+/* types of the delayed item */
+#define BTRFS_DELAYED_INSERTION_ITEM	1
+#define BTRFS_DELAYED_DELETION_ITEM	2
+
+struct btrfs_delayed_root {
+	spinlock_t lock;
+	struct list_head node_list;
+	/*
+	 * Used for delayed nodes which is waiting to be dealt with by the
+	 * worker. If the delayed node is inserted into the work queue, we
+	 * drop it from this list.
+	 */
+	struct list_head prepare_list;
+	atomic_t items;		/* for delayed items */
+	int nodes;		/* for delayed nodes */
+	wait_queue_head_t wait;
+};
+
+struct btrfs_delayed_node {
+	u64 inode_id;
+	u64 bytes_reserved;
+	struct btrfs_root *root;
+	/* Used to add the node into the delayed root's node list. */
+	struct list_head n_list;
+	/*
+	 * Used to add the node into the prepare list, the nodes in this list
+	 * is waiting to be dealt with by the async worker.
+	 */
+	struct list_head p_list;
+	struct rb_root ins_root;
+	struct rb_root del_root;
+	struct mutex mutex;
+	struct btrfs_inode_item inode_item;
+	atomic_t refs;
+	u64 index_cnt;
+	bool in_list;
+	bool inode_dirty;
+	int count;
+};
+
+struct btrfs_delayed_item {
+	struct rb_node rb_node;
+	struct btrfs_key key;
+	struct list_head tree_list;	/* used for batch insert/delete items */
+	struct list_head readdir_list;	/* used for readdir items */
+	u64 bytes_reserved;
+	struct btrfs_block_rsv *block_rsv;
+	struct btrfs_delayed_node *delayed_node;
+	atomic_t refs;
+	int ins_or_del;
+	u32 data_len;
+	char data[0];
+};
+
+static inline void btrfs_init_delayed_root(
+				struct btrfs_delayed_root *delayed_root)
+{
+	atomic_set(&delayed_root->items, 0);
+	delayed_root->nodes = 0;
+	spin_lock_init(&delayed_root->lock);
+	init_waitqueue_head(&delayed_root->wait);
+	INIT_LIST_HEAD(&delayed_root->node_list);
+	INIT_LIST_HEAD(&delayed_root->prepare_list);
+}
+
+int btrfs_insert_delayed_dir_index(struct btrfs_trans_handle *trans,
+				   struct btrfs_root *root, const char *name,
+				   int name_len, struct inode *dir,
+				   struct btrfs_disk_key *disk_key, u8 type,
+				   u64 index);
+
+int btrfs_delete_delayed_dir_index(struct btrfs_trans_handle *trans,
+				   struct btrfs_root *root, struct inode *dir,
+				   u64 index);
+
+int btrfs_inode_delayed_dir_index_count(struct inode *inode);
+
+int btrfs_run_delayed_items(struct btrfs_trans_handle *trans,
+			    struct btrfs_root *root);
+
+void btrfs_balance_delayed_items(struct btrfs_root *root);
+
+int btrfs_commit_inode_delayed_items(struct btrfs_trans_handle *trans,
+				     struct inode *inode);
+/* Used for evicting the inode. */
+void btrfs_remove_delayed_node(struct inode *inode);
+void btrfs_kill_delayed_inode_items(struct inode *inode);
+
+
+int btrfs_delayed_update_inode(struct btrfs_trans_handle *trans,
+			       struct btrfs_root *root, struct inode *inode);
+
+/* Used for drop dead root */
+void btrfs_kill_all_delayed_nodes(struct btrfs_root *root);
+
+/* Used for readdir() */
+void btrfs_get_delayed_items(struct inode *inode, struct list_head *ins_list,
+			     struct list_head *del_list);
+void btrfs_put_delayed_items(struct list_head *ins_list,
+			     struct list_head *del_list);
+int btrfs_should_delete_dir_index(struct list_head *del_list,
+				  u64 index);
+int btrfs_readdir_delayed_dir_index(struct file *filp, void *dirent,
+				    filldir_t filldir,
+				    struct list_head *ins_list);
+
+/* for init */
+int __init btrfs_delayed_inode_init(void);
+void btrfs_delayed_inode_exit(void);
+#endif

fs/btrfs/dir-item.c

@@ -124,8 +124,9 @@ int btrfs_insert_xattr_item(struct btrfs_trans_handle *trans,
  * to use for the second index (if one is created).
  */
 int btrfs_insert_dir_item(struct btrfs_trans_handle *trans, struct btrfs_root
-			  *root, const char *name, int name_len, u64 dir,
-			  struct btrfs_key *location, u8 type, u64 index)
+			  *root, const char *name, int name_len,
+			  struct inode *dir, struct btrfs_key *location,
+			  u8 type, u64 index)
 {
 	int ret = 0;
 	int ret2 = 0;
@@ -137,13 +138,17 @@ int btrfs_insert_dir_item(struct btrfs_trans_handle *trans, struct btrfs_root
 	struct btrfs_disk_key disk_key;
 	u32 data_size;
 
-	key.objectid = dir;
+	key.objectid = dir->i_ino;
 	btrfs_set_key_type(&key, BTRFS_DIR_ITEM_KEY);
 	key.offset = btrfs_name_hash(name, name_len);
 
 	path = btrfs_alloc_path();
+	if (!path)
+		return -ENOMEM;
 	path->leave_spinning = 1;
 
+	btrfs_cpu_key_to_disk(&disk_key, location);
+
 	data_size = sizeof(*dir_item) + name_len;
 	dir_item = insert_with_overflow(trans, root, path, &key, data_size,
 					name, name_len);
@@ -155,7 +160,6 @@ int btrfs_insert_dir_item(struct btrfs_trans_handle *trans, struct btrfs_root
 	}
 
 	leaf = path->nodes[0];
-	btrfs_cpu_key_to_disk(&disk_key, location);
 	btrfs_set_dir_item_key(leaf, dir_item, &disk_key);
 	btrfs_set_dir_type(leaf, dir_item, type);
 	btrfs_set_dir_data_len(leaf, dir_item, 0);
@@ -174,27 +178,9 @@ int btrfs_insert_dir_item(struct btrfs_trans_handle *trans, struct btrfs_root
 	}
 	btrfs_release_path(root, path);
 
-	btrfs_set_key_type(&key, BTRFS_DIR_INDEX_KEY);
-	key.offset = index;
-	dir_item = insert_with_overflow(trans, root, path, &key, data_size,
-					name, name_len);
-	if (IS_ERR(dir_item)) {
-		ret2 = PTR_ERR(dir_item);
-		goto out_free;
-	}
-	leaf = path->nodes[0];
-	btrfs_cpu_key_to_disk(&disk_key, location);
-	btrfs_set_dir_item_key(leaf, dir_item, &disk_key);
-	btrfs_set_dir_type(leaf, dir_item, type);
-	btrfs_set_dir_data_len(leaf, dir_item, 0);
-	btrfs_set_dir_name_len(leaf, dir_item, name_len);
-	btrfs_set_dir_transid(leaf, dir_item, trans->transid);
-	name_ptr = (unsigned long)(dir_item + 1);
-	write_extent_buffer(leaf, name, name_ptr, name_len);
-	btrfs_mark_buffer_dirty(leaf);
-
+	ret2 = btrfs_insert_delayed_dir_index(trans, root, name, name_len, dir,
+					      &disk_key, type, index);
 out_free:
-
 	btrfs_free_path(path);
 	if (ret)
 		return ret;

fs/btrfs/disk-io.c

@@ -1059,6 +1059,7 @@ static int __setup_root(u32 nodesize, u32 leafsize, u32 sectorsize,
 	root->name = NULL;
 	root->in_sysfs = 0;
 	root->inode_tree = RB_ROOT;
+	INIT_RADIX_TREE(&root->delayed_nodes_tree, GFP_ATOMIC);
 	root->block_rsv = NULL;
 	root->orphan_block_rsv = NULL;
 
@@ -1707,6 +1708,13 @@ struct btrfs_root *open_ctree(struct super_block *sb,
 
 	INIT_LIST_HEAD(&fs_info->ordered_extents);
 	spin_lock_init(&fs_info->ordered_extent_lock);
+	fs_info->delayed_root = kmalloc(sizeof(struct btrfs_delayed_root),
+					GFP_NOFS);
+	if (!fs_info->delayed_root) {
+		err = -ENOMEM;
+		goto fail_iput;
+	}
+	btrfs_init_delayed_root(fs_info->delayed_root);
 
 	sb->s_blocksize = 4096;
 	sb->s_blocksize_bits = blksize_bits(4096);
@@ -1774,7 +1782,7 @@ struct btrfs_root *open_ctree(struct super_block *sb,
 	bh = btrfs_read_dev_super(fs_devices->latest_bdev);
 	if (!bh) {
 		err = -EINVAL;
-		goto fail_iput;
+		goto fail_alloc;
 	}
 
 	memcpy(&fs_info->super_copy, bh->b_data, sizeof(fs_info->super_copy));
@@ -1786,7 +1794,7 @@ struct btrfs_root *open_ctree(struct super_block *sb,
 
 	disk_super = &fs_info->super_copy;
 	if (!btrfs_super_root(disk_super))
-		goto fail_iput;
+		goto fail_alloc;
 
 	/* check FS state, whether FS is broken. */
 	fs_info->fs_state |= btrfs_super_flags(disk_super);
@@ -1802,7 +1810,7 @@ struct btrfs_root *open_ctree(struct super_block *sb,
 	ret = btrfs_parse_options(tree_root, options);
 	if (ret) {
 		err = ret;
-		goto fail_iput;
+		goto fail_alloc;
 	}
 
 	features = btrfs_super_incompat_flags(disk_super) &
@@ -1812,7 +1820,7 @@ struct btrfs_root *open_ctree(struct super_block *sb,
 		       "unsupported optional features (%Lx).\n",
 		       (unsigned long long)features);
 		err = -EINVAL;
-		goto fail_iput;
+		goto fail_alloc;
 	}
 
 	features = btrfs_super_incompat_flags(disk_super);
@@ -1828,7 +1836,7 @@ struct btrfs_root *open_ctree(struct super_block *sb,
 		       "unsupported option features (%Lx).\n",
 		       (unsigned long long)features);
 		err = -EINVAL;
-		goto fail_iput;
+		goto fail_alloc;
 	}
 
 	btrfs_init_workers(&fs_info->generic_worker,
@@ -1875,6 +1883,9 @@ struct btrfs_root *open_ctree(struct super_block *sb,
 			   &fs_info->generic_worker);
 	btrfs_init_workers(&fs_info->endio_freespace_worker, "freespace-write",
 			   1, &fs_info->generic_worker);
+	btrfs_init_workers(&fs_info->delayed_workers, "delayed-meta",
+			   fs_info->thread_pool_size,
+			   &fs_info->generic_worker);
 
 	/*
 	 * endios are largely parallel and should have a very
@@ -1896,6 +1907,7 @@ struct btrfs_root *open_ctree(struct super_block *sb,
 	btrfs_start_workers(&fs_info->endio_meta_write_workers, 1);
 	btrfs_start_workers(&fs_info->endio_write_workers, 1);
 	btrfs_start_workers(&fs_info->endio_freespace_worker, 1);
+	btrfs_start_workers(&fs_info->delayed_workers, 1);
 
 	fs_info->bdi.ra_pages *= btrfs_super_num_devices(disk_super);
 	fs_info->bdi.ra_pages = max(fs_info->bdi.ra_pages,
@@ -2152,6 +2164,9 @@ struct btrfs_root *open_ctree(struct super_block *sb,
 	btrfs_stop_workers(&fs_info->endio_write_workers);
 	btrfs_stop_workers(&fs_info->endio_freespace_worker);
 	btrfs_stop_workers(&fs_info->submit_workers);
+	btrfs_stop_workers(&fs_info->delayed_workers);
+fail_alloc:
+	kfree(fs_info->delayed_root);
 fail_iput:
 	invalidate_inode_pages2(fs_info->btree_inode->i_mapping);
 	iput(fs_info->btree_inode);
@@ -2597,6 +2612,7 @@ int close_ctree(struct btrfs_root *root)
 	del_fs_roots(fs_info);
 
 	iput(fs_info->btree_inode);
+	kfree(fs_info->delayed_root);
 
 	btrfs_stop_workers(&fs_info->generic_worker);
 	btrfs_stop_workers(&fs_info->fixup_workers);
@@ -2608,6 +2624,7 @@ int close_ctree(struct btrfs_root *root)
 	btrfs_stop_workers(&fs_info->endio_write_workers);
 	btrfs_stop_workers(&fs_info->endio_freespace_worker);
 	btrfs_stop_workers(&fs_info->submit_workers);
+	btrfs_stop_workers(&fs_info->delayed_workers);
 
 	btrfs_close_devices(fs_info->fs_devices);
 	btrfs_mapping_tree_free(&fs_info->mapping_tree);
@@ -2681,6 +2698,29 @@ void btrfs_btree_balance_dirty(struct btrfs_root *root, unsigned long nr)
 	u64 num_dirty;
 	unsigned long thresh = 32 * 1024 * 1024;
 
+	if (current->flags & PF_MEMALLOC)
+		return;
+
+	btrfs_balance_delayed_items(root);
+
+	num_dirty = root->fs_info->dirty_metadata_bytes;
+
+	if (num_dirty > thresh) {
+		balance_dirty_pages_ratelimited_nr(
+				   root->fs_info->btree_inode->i_mapping, 1);
+	}
+	return;
+}
+
+void __btrfs_btree_balance_dirty(struct btrfs_root *root, unsigned long nr)
+{
+	/*
+	 * looks as though older kernels can get into trouble with
+	 * this code, they end up stuck in balance_dirty_pages forever
+	 */
+	u64 num_dirty;
+	unsigned long thresh = 32 * 1024 * 1024;
+
 	if (current->flags & PF_MEMALLOC)
 		return;
 

fs/btrfs/disk-io.h

@@ -71,6 +71,7 @@ int btrfs_insert_dev_radix(struct btrfs_root *root,
 			   u64 block_start,
 			   u64 num_blocks);
 void btrfs_btree_balance_dirty(struct btrfs_root *root, unsigned long nr);
+void __btrfs_btree_balance_dirty(struct btrfs_root *root, unsigned long nr);
 int btrfs_free_fs_root(struct btrfs_fs_info *fs_info, struct btrfs_root *root);
 void btrfs_mark_buffer_dirty(struct extent_buffer *buf);
 void btrfs_mark_buffer_dirty_nonblocking(struct extent_buffer *buf);

fs/btrfs/extent-tree.c

@@ -3975,12 +3975,6 @@ static void release_global_block_rsv(struct btrfs_fs_info *fs_info)
 	WARN_ON(fs_info->chunk_block_rsv.reserved > 0);
 }
 
-static u64 calc_trans_metadata_size(struct btrfs_root *root, int num_items)
-{
-	return (root->leafsize + root->nodesize * (BTRFS_MAX_LEVEL - 1)) *
-		3 * num_items;
-}
-
 int btrfs_trans_reserve_metadata(struct btrfs_trans_handle *trans,
 				 struct btrfs_root *root,
 				 int num_items)
@@ -3991,7 +3985,7 @@ int btrfs_trans_reserve_metadata(struct btrfs_trans_handle *trans,
 	if (num_items == 0 || root->fs_info->chunk_root == root)
 		return 0;
 
-	num_bytes = calc_trans_metadata_size(root, num_items);
+	num_bytes = btrfs_calc_trans_metadata_size(root, num_items);
 	ret = btrfs_block_rsv_add(trans, root, &root->fs_info->trans_block_rsv,
 				  num_bytes);
 	if (!ret) {
@@ -4030,14 +4024,14 @@ int btrfs_orphan_reserve_metadata(struct btrfs_trans_handle *trans,
 	 * If all of the metadata space is used, we can commit
 	 * transaction and use space it freed.
 	 */
-	u64 num_bytes = calc_trans_metadata_size(root, 4);
+	u64 num_bytes = btrfs_calc_trans_metadata_size(root, 4);
 	return block_rsv_migrate_bytes(src_rsv, dst_rsv, num_bytes);
 }
 
 void btrfs_orphan_release_metadata(struct inode *inode)
 {
 	struct btrfs_root *root = BTRFS_I(inode)->root;
-	u64 num_bytes = calc_trans_metadata_size(root, 4);
+	u64 num_bytes = btrfs_calc_trans_metadata_size(root, 4);
 	btrfs_block_rsv_release(root, root->orphan_block_rsv, num_bytes);
 }
 
@@ -4051,7 +4045,7 @@ int btrfs_snap_reserve_metadata(struct btrfs_trans_handle *trans,
 	 * two for root back/forward refs, two for directory entries
 	 * and one for root of the snapshot.
 	 */
-	u64 num_bytes = calc_trans_metadata_size(root, 5);
+	u64 num_bytes = btrfs_calc_trans_metadata_size(root, 5);
 	dst_rsv->space_info = src_rsv->space_info;
 	return block_rsv_migrate_bytes(src_rsv, dst_rsv, num_bytes);
 }
@@ -4080,7 +4074,7 @@ int btrfs_delalloc_reserve_metadata(struct inode *inode, u64 num_bytes)
 
 	if (nr_extents > reserved_extents) {
 		nr_extents -= reserved_extents;
-		to_reserve = calc_trans_metadata_size(root, nr_extents);
+		to_reserve = btrfs_calc_trans_metadata_size(root, nr_extents);
 	} else {
 		nr_extents = 0;
 		to_reserve = 0;
@@ -4134,7 +4128,7 @@ void btrfs_delalloc_release_metadata(struct inode *inode, u64 num_bytes)
 
 	to_free = calc_csum_metadata_size(inode, num_bytes);
 	if (nr_extents > 0)
-		to_free += calc_trans_metadata_size(root, nr_extents);
+		to_free += btrfs_calc_trans_metadata_size(root, nr_extents);
 
 	btrfs_block_rsv_release(root, &root->fs_info->delalloc_block_rsv,
 				to_free);

fs/btrfs/inode.c

@@ -2659,11 +2659,26 @@ noinline int btrfs_update_inode(struct btrfs_trans_handle *trans,
 	struct extent_buffer *leaf;
 	int ret;
 
+	/*
+	 * If root is tree root, it means this inode is used to
+	 * store free space information. And these inodes are updated
+	 * when committing the transaction, so they needn't delaye to
+	 * be updated, or deadlock will occured.
+	 */
+	if (!is_free_space_inode(root, inode)) {
+		ret = btrfs_delayed_update_inode(trans, root, inode);
+		if (!ret)
+			btrfs_set_inode_last_trans(trans, inode);
+		return ret;
+	}
+
 	path = btrfs_alloc_path();
-	BUG_ON(!path);
+	if (!path)
+		return -ENOMEM;
+
 	path->leave_spinning = 1;
-	ret = btrfs_lookup_inode(trans, root, path,
-				 &BTRFS_I(inode)->location, 1);
+	ret = btrfs_lookup_inode(trans, root, path, &BTRFS_I(inode)->location,
+				 1);
 	if (ret) {
 		if (ret > 0)
 			ret = -ENOENT;
@@ -2673,7 +2688,7 @@ noinline int btrfs_update_inode(struct btrfs_trans_handle *trans,
 	btrfs_unlock_up_safe(path, 1);
 	leaf = path->nodes[0];
 	inode_item = btrfs_item_ptr(leaf, path->slots[0],
-				  struct btrfs_inode_item);
+				    struct btrfs_inode_item);
 
 	fill_inode_item(trans, leaf, inode_item, inode);
 	btrfs_mark_buffer_dirty(leaf);
@@ -2684,7 +2699,6 @@ noinline int btrfs_update_inode(struct btrfs_trans_handle *trans,
 	return ret;
 }
 
-
 /*
  * unlink helper that gets used here in inode.c and in the tree logging
  * recovery code.  It remove a link in a directory with a given name, and
@@ -2737,18 +2751,9 @@ static int __btrfs_unlink_inode(struct btrfs_trans_handle *trans,
 		goto err;
 	}
 
-	di = btrfs_lookup_dir_index_item(trans, root, path, dir_ino,
-					 index, name, name_len, -1);
-	if (IS_ERR(di)) {
-		ret = PTR_ERR(di);
-		goto err;
-	}
-	if (!di) {
-		ret = -ENOENT;
+	ret = btrfs_delete_delayed_dir_index(trans, root, dir, index);
+	if (ret)
 		goto err;
-	}
-	ret = btrfs_delete_one_dir_name(trans, root, path, di);
-	btrfs_release_path(root, path);
 
 	ret = btrfs_del_inode_ref_in_log(trans, root, name, name_len,
 					 inode, dir_ino);
@@ -2939,6 +2944,14 @@ static struct btrfs_trans_handle *__unlink_start_trans(struct inode *dir,
 	index = btrfs_inode_ref_index(path->nodes[0], ref);
 	btrfs_release_path(root, path);
 
+	/*
+	 * This is a commit root search, if we can lookup inode item and other
+	 * relative items in the commit root, it means the transaction of
+	 * dir/file creation has been committed, and the dir index item that we
+	 * delay to insert has also been inserted into the commit root. So
+	 * we needn't worry about the delayed insertion of the dir index item
+	 * here.
+	 */
 	di = btrfs_lookup_dir_index_item(trans, root, path, dir_ino, index,
 				dentry->d_name.name, dentry->d_name.len, 0);
 	if (IS_ERR(di)) {
@@ -3045,24 +3058,16 @@ int btrfs_unlink_subvol(struct btrfs_trans_handle *trans,
 		btrfs_release_path(root, path);
 		index = key.offset;
 	}
+	btrfs_release_path(root, path);
 
-	di = btrfs_lookup_dir_index_item(trans, root, path, dir_ino,
-					 index, name, name_len, -1);
-	BUG_ON(!di || IS_ERR(di));
-
-	leaf = path->nodes[0];
-	btrfs_dir_item_key_to_cpu(leaf, di, &key);
-	WARN_ON(key.type != BTRFS_ROOT_ITEM_KEY || key.objectid != objectid);
-	ret = btrfs_delete_one_dir_name(trans, root, path, di);
+	ret = btrfs_delete_delayed_dir_index(trans, root, dir, index);
 	BUG_ON(ret);
-	btrfs_release_path(root, path);
 
 	btrfs_i_size_write(dir, dir->i_size - name_len * 2);
 	dir->i_mtime = dir->i_ctime = CURRENT_TIME;
 	ret = btrfs_update_inode(trans, root, dir);
 	BUG_ON(ret);
 
-	btrfs_free_path(path);
 	return 0;
 }
 
@@ -3323,6 +3328,15 @@ int btrfs_truncate_inode_items(struct btrfs_trans_handle *trans,
 	if (root->ref_cows || root == root->fs_info->tree_root)
 		btrfs_drop_extent_cache(inode, new_size & (~mask), (u64)-1, 0);
 
+	/*
+	 * This function is also used to drop the items in the log tree before
+	 * we relog the inode, so if root != BTRFS_I(inode)->root, it means
+	 * it is used to drop the loged items. So we shouldn't kill the delayed
+	 * items.
+	 */
+	if (min_type == 0 && root == BTRFS_I(inode)->root)
+		btrfs_kill_delayed_inode_items(inode);
+
 	path = btrfs_alloc_path();
 	BUG_ON(!path);
 	path->reada = -1;
@@ -4232,7 +4246,7 @@ static struct dentry *btrfs_lookup(struct inode *dir, struct dentry *dentry,
 	return d_splice_alias(inode, dentry);
 }
 
-static unsigned char btrfs_filetype_table[] = {
+unsigned char btrfs_filetype_table[] = {
 	DT_UNKNOWN, DT_REG, DT_DIR, DT_CHR, DT_BLK, DT_FIFO, DT_SOCK, DT_LNK
 };
 
@@ -4246,6 +4260,8 @@ static int btrfs_real_readdir(struct file *filp, void *dirent,
 	struct btrfs_key key;
 	struct btrfs_key found_key;
 	struct btrfs_path *path;
+	struct list_head ins_list;
+	struct list_head del_list;
 	int ret;
 	struct extent_buffer *leaf;
 	int slot;
@@ -4258,6 +4274,7 @@ static int btrfs_real_readdir(struct file *filp, void *dirent,
 	char tmp_name[32];
 	char *name_ptr;
 	int name_len;
+	int is_curr = 0;	/* filp->f_pos points to the current index? */
 
 	/* FIXME, use a real flag for deciding about the key type */
 	if (root->fs_info->tree_root == root)
@@ -4280,8 +4297,16 @@ static int btrfs_real_readdir(struct file *filp, void *dirent,
 		filp->f_pos = 2;
 	}
 	path = btrfs_alloc_path();
+	if (!path)
+		return -ENOMEM;
 	path->reada = 2;
 
+	if (key_type == BTRFS_DIR_INDEX_KEY) {
+		INIT_LIST_HEAD(&ins_list);
+		INIT_LIST_HEAD(&del_list);
+		btrfs_get_delayed_items(inode, &ins_list, &del_list);
+	}
+
 	btrfs_set_key_type(&key, key_type);
 	key.offset = filp->f_pos;
 	key.objectid = btrfs_ino(inode);
@@ -4311,8 +4336,13 @@ static int btrfs_real_readdir(struct file *filp, void *dirent,
 			break;
 		if (found_key.offset < filp->f_pos)
 			goto next;
+		if (key_type == BTRFS_DIR_INDEX_KEY &&
+		    btrfs_should_delete_dir_index(&del_list,
+						  found_key.offset))
+			goto next;
 
 		filp->f_pos = found_key.offset;
+		is_curr = 1;
 
 		di = btrfs_item_ptr(leaf, slot, struct btrfs_dir_item);
 		di_cur = 0;
@@ -4367,6 +4397,15 @@ static int btrfs_real_readdir(struct file *filp, void *dirent,
 		path->slots[0]++;
 	}
 
+	if (key_type == BTRFS_DIR_INDEX_KEY) {
+		if (is_curr)
+			filp->f_pos++;
+		ret = btrfs_readdir_delayed_dir_index(filp, dirent, filldir,
+						      &ins_list);
+		if (ret)
+			goto nopos;
+	}
+
 	/* Reached end of directory/root. Bump pos past the last item. */
 	if (key_type == BTRFS_DIR_INDEX_KEY)
 		/*
@@ -4379,6 +4418,8 @@ static int btrfs_real_readdir(struct file *filp, void *dirent,
 nopos:
 	ret = 0;
 err:
+	if (key_type == BTRFS_DIR_INDEX_KEY)
+		btrfs_put_delayed_items(&ins_list, &del_list);
 	btrfs_free_path(path);
 	return ret;
 }
@@ -4459,6 +4500,8 @@ void btrfs_dirty_inode(struct inode *inode)
 		}
 	}
 	btrfs_end_transaction(trans, root);
+	if (BTRFS_I(inode)->delayed_node)
+		btrfs_balance_delayed_items(root);
 }
 
 /*
@@ -4527,9 +4570,12 @@ int btrfs_set_inode_index(struct inode *dir, u64 *index)
 	int ret = 0;
 
 	if (BTRFS_I(dir)->index_cnt == (u64)-1) {
-		ret = btrfs_set_inode_index_count(dir);
-		if (ret)
-			return ret;
+		ret = btrfs_inode_delayed_dir_index_count(dir);
+		if (ret) {
+			ret = btrfs_set_inode_index_count(dir);
+			if (ret)
+				return ret;
+		}
 	}
 
 	*index = BTRFS_I(dir)->index_cnt;
@@ -4701,7 +4747,7 @@ int btrfs_add_link(struct btrfs_trans_handle *trans,
 
 	if (ret == 0) {
 		ret = btrfs_insert_dir_item(trans, root, name, name_len,
-					    parent_ino, &key,
+					    parent_inode, &key,
 					    btrfs_inode_type(inode), index);
 		BUG_ON(ret);
 
@@ -6816,6 +6862,8 @@ struct inode *btrfs_alloc_inode(struct super_block *sb)
 	ei->dummy_inode = 0;
 	ei->force_compress = BTRFS_COMPRESS_NONE;
 
+	ei->delayed_node = NULL;
+
 	inode = &ei->vfs_inode;
 	extent_map_tree_init(&ei->extent_tree, GFP_NOFS);
 	extent_io_tree_init(&ei->io_tree, &inode->i_data, GFP_NOFS);
@@ -6906,6 +6954,7 @@ void btrfs_destroy_inode(struct inode *inode)
 	inode_tree_del(inode);
 	btrfs_drop_extent_cache(inode, 0, (u64)-1, 0);
 free:
+	btrfs_remove_delayed_node(inode);
 	call_rcu(&inode->i_rcu, btrfs_i_callback);
 }
 

fs/btrfs/ioctl.c

@@ -422,7 +422,7 @@ static noinline int create_subvol(struct btrfs_root *root,
 	BUG_ON(ret);
 
 	ret = btrfs_insert_dir_item(trans, root,
-				    name, namelen, btrfs_ino(dir), &key,
+				    name, namelen, dir, &key,
 				    BTRFS_FT_DIR, index);
 	if (ret)
 		goto fail;

fs/btrfs/super.c

@@ -40,6 +40,7 @@
 #include <linux/magic.h>
 #include <linux/slab.h>
 #include "compat.h"
+#include "delayed-inode.h"
 #include "ctree.h"
 #include "disk-io.h"
 #include "transaction.h"
@@ -1206,10 +1207,14 @@ static int __init init_btrfs_fs(void)
 	if (err)
 		goto free_extent_io;
 
-	err = btrfs_interface_init();
+	err = btrfs_delayed_inode_init();
 	if (err)
 		goto free_extent_map;
 
+	err = btrfs_interface_init();
+	if (err)
+		goto free_delayed_inode;
+
 	err = register_filesystem(&btrfs_fs_type);
 	if (err)
 		goto unregister_ioctl;
@@ -1219,6 +1224,8 @@ static int __init init_btrfs_fs(void)
 
 unregister_ioctl:
 	btrfs_interface_exit();
+free_delayed_inode:
+	btrfs_delayed_inode_exit();
 free_extent_map:
 	extent_map_exit();
 free_extent_io:
@@ -1235,6 +1242,7 @@ static int __init init_btrfs_fs(void)
 static void __exit exit_btrfs_fs(void)
 {
 	btrfs_destroy_cachep();
+	btrfs_delayed_inode_exit();
 	extent_map_exit();
 	extent_io_exit();
 	btrfs_interface_exit();

fs/btrfs/transaction.c

@@ -488,19 +488,40 @@ static int __btrfs_end_transaction(struct btrfs_trans_handle *trans,
 int btrfs_end_transaction(struct btrfs_trans_handle *trans,
 			  struct btrfs_root *root)
 {
-	return __btrfs_end_transaction(trans, root, 0, 1);
+	int ret;
+
+	ret = __btrfs_end_transaction(trans, root, 0, 1);
+	if (ret)
+		return ret;
+	return 0;
 }
 
 int btrfs_end_transaction_throttle(struct btrfs_trans_handle *trans,
 				   struct btrfs_root *root)
 {
-	return __btrfs_end_transaction(trans, root, 1, 1);
+	int ret;
+
+	ret = __btrfs_end_transaction(trans, root, 1, 1);
+	if (ret)
+		return ret;
+	return 0;
 }
 
 int btrfs_end_transaction_nolock(struct btrfs_trans_handle *trans,
 				 struct btrfs_root *root)
 {
-	return __btrfs_end_transaction(trans, root, 0, 0);
+	int ret;
+
+	ret = __btrfs_end_transaction(trans, root, 0, 0);
+	if (ret)
+		return ret;
+	return 0;
+}
+
+int btrfs_end_transaction_dmeta(struct btrfs_trans_handle *trans,
+				struct btrfs_root *root)
+{
+	return __btrfs_end_transaction(trans, root, 1, 1);
 }
 
 /*
@@ -974,7 +995,7 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
 	BUG_ON(ret);
 	ret = btrfs_insert_dir_item(trans, parent_root,
 				dentry->d_name.name, dentry->d_name.len,
-				btrfs_ino(parent_inode), &key,
+				parent_inode, &key,
 				BTRFS_FT_DIR, index);
 	BUG_ON(ret);
 
@@ -1044,6 +1065,14 @@ static noinline int create_pending_snapshots(struct btrfs_trans_handle *trans,
 	int ret;
 
 	list_for_each_entry(pending, head, list) {
+		/*
+		 * We must deal with the delayed items before creating
+		 * snapshots, or we will create a snapthot with inconsistent
+		 * information.
+		*/
+		ret = btrfs_run_delayed_items(trans, fs_info->fs_root);
+		BUG_ON(ret);
+
 		ret = create_pending_snapshot(trans, fs_info, pending);
 		BUG_ON(ret);
 	}
@@ -1297,6 +1326,9 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans,
 			BUG_ON(ret);
 		}
 
+		ret = btrfs_run_delayed_items(trans, root);
+		BUG_ON(ret);
+
 		/*
 		 * rename don't use btrfs_join_transaction, so, once we
 		 * set the transaction to blocked above, we aren't going
@@ -1323,6 +1355,9 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans,
 	ret = create_pending_snapshots(trans, root->fs_info);
 	BUG_ON(ret);
 
+	ret = btrfs_run_delayed_items(trans, root);
+	BUG_ON(ret);
+
 	ret = btrfs_run_delayed_refs(trans, root, (unsigned long)-1);
 	BUG_ON(ret);
 
@@ -1439,6 +1474,8 @@ int btrfs_clean_old_snapshots(struct btrfs_root *root)
 		root = list_entry(list.next, struct btrfs_root, root_list);
 		list_del(&root->root_list);
 
+		btrfs_kill_all_delayed_nodes(root);
+
 		if (btrfs_header_backref_rev(root->node) <
 		    BTRFS_MIXED_BACKREF_REV)
 			btrfs_drop_snapshot(root, NULL, 0);

fs/btrfs/transaction.h

@@ -115,6 +115,8 @@ int btrfs_commit_transaction_async(struct btrfs_trans_handle *trans,
 				   int wait_for_unblock);
 int btrfs_end_transaction_throttle(struct btrfs_trans_handle *trans,
 				   struct btrfs_root *root);
+int btrfs_end_transaction_dmeta(struct btrfs_trans_handle *trans,
+				struct btrfs_root *root);
 int btrfs_should_end_transaction(struct btrfs_trans_handle *trans,
 				 struct btrfs_root *root);
 void btrfs_throttle(struct btrfs_root *root);

fs/btrfs/tree-log.c

@@ -2774,6 +2774,13 @@ static int btrfs_log_inode(struct btrfs_trans_handle *trans,
 		max_key.type = (u8)-1;
 	max_key.offset = (u64)-1;
 
+	ret = btrfs_commit_inode_delayed_items(trans, inode);
+	if (ret) {
+		btrfs_free_path(path);
+		btrfs_free_path(dst_path);
+		return ret;
+	}
+
 	mutex_lock(&BTRFS_I(inode)->log_mutex);
 
 	/*