Merge branch 'topic/error' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/regulator into regulator-fixed

65 files changed, 5016 insertions, 3114 deletions

diff --git a/fs/btrfs/acl.c b/fs/btrfs/acl.c
index 67a607709d4f..247b8dfaf6e5 100644
--- a/fs/btrfs/acl.c
+++ b/fs/btrfs/acl.c
@@ -55,8 +55,7 @@ struct posix_acl *btrfs_get_acl(struct inode *inode, int type)
 	}
 	if (size > 0) {
 		acl = posix_acl_from_xattr(&init_user_ns, value, size);
-	} else if (size == -ENOENT || size == -ENODATA || size == 0) {
-		/* FIXME, who returns -ENOENT?  I think nobody */
+	} else if (size == -ERANGE || size == -ENODATA || size == 0) {
 		acl = NULL;
 	} else {
 		acl = ERR_PTR(-EIO);
@@ -80,11 +79,9 @@ static int __btrfs_set_acl(struct btrfs_trans_handle *trans,
 	case ACL_TYPE_ACCESS:
 		name = XATTR_NAME_POSIX_ACL_ACCESS;
 		if (acl) {
-			ret = posix_acl_equiv_mode(acl, &inode->i_mode);
-			if (ret < 0)
+			ret = posix_acl_update_mode(inode, &inode->i_mode, &acl);
+			if (ret)
 				return ret;
-			if (ret == 0)
-				acl = NULL;
 		}
 		ret = 0;
 		break;
diff --git a/fs/btrfs/async-thread.c b/fs/btrfs/async-thread.c
index 5fb60ea7eee2..e0f071f6b5a7 100644
--- a/fs/btrfs/async-thread.c
+++ b/fs/btrfs/async-thread.c
@@ -34,6 +34,10 @@
 
 struct __btrfs_workqueue {
 	struct workqueue_struct *normal_wq;
+
+	/* File system this workqueue services */
+	struct btrfs_fs_info *fs_info;
+
 	/* List head pointing to ordered work list */
 	struct list_head ordered_list;
 
@@ -70,6 +74,18 @@ void btrfs_##name(struct work_struct *arg)				\
 	normal_work_helper(work);					\
 }
 
+struct btrfs_fs_info *
+btrfs_workqueue_owner(struct __btrfs_workqueue *wq)
+{
+	return wq->fs_info;
+}
+
+struct btrfs_fs_info *
+btrfs_work_owner(struct btrfs_work *work)
+{
+	return work->wq->fs_info;
+}
+
 BTRFS_WORK_HELPER(worker_helper);
 BTRFS_WORK_HELPER(delalloc_helper);
 BTRFS_WORK_HELPER(flush_delalloc_helper);
@@ -94,14 +110,15 @@ BTRFS_WORK_HELPER(scrubnc_helper);
 BTRFS_WORK_HELPER(scrubparity_helper);
 
 static struct __btrfs_workqueue *
-__btrfs_alloc_workqueue(const char *name, unsigned int flags, int limit_active,
-			 int thresh)
+__btrfs_alloc_workqueue(struct btrfs_fs_info *fs_info, const char *name,
+			unsigned int flags, int limit_active, int thresh)
 {
 	struct __btrfs_workqueue *ret = kzalloc(sizeof(*ret), GFP_KERNEL);
 
 	if (!ret)
 		return NULL;
 
+	ret->fs_info = fs_info;
 	ret->limit_active = limit_active;
 	atomic_set(&ret->pending, 0);
 	if (thresh == 0)
@@ -143,7 +160,8 @@ __btrfs_alloc_workqueue(const char *name, unsigned int flags, int limit_active,
 static inline void
 __btrfs_destroy_workqueue(struct __btrfs_workqueue *wq);
 
-struct btrfs_workqueue *btrfs_alloc_workqueue(const char *name,
+struct btrfs_workqueue *btrfs_alloc_workqueue(struct btrfs_fs_info *fs_info,
+					      const char *name,
 					      unsigned int flags,
 					      int limit_active,
 					      int thresh)
@@ -153,7 +171,8 @@ struct btrfs_workqueue *btrfs_alloc_workqueue(const char *name,
 	if (!ret)
 		return NULL;
 
-	ret->normal = __btrfs_alloc_workqueue(name, flags & ~WQ_HIGHPRI,
+	ret->normal = __btrfs_alloc_workqueue(fs_info, name,
+					      flags & ~WQ_HIGHPRI,
 					      limit_active, thresh);
 	if (!ret->normal) {
 		kfree(ret);
@@ -161,8 +180,8 @@ struct btrfs_workqueue *btrfs_alloc_workqueue(const char *name,
 	}
 
 	if (flags & WQ_HIGHPRI) {
-		ret->high = __btrfs_alloc_workqueue(name, flags, limit_active,
-						    thresh);
+		ret->high = __btrfs_alloc_workqueue(fs_info, name, flags,
+						    limit_active, thresh);
 		if (!ret->high) {
 			__btrfs_destroy_workqueue(ret->normal);
 			kfree(ret);
diff --git a/fs/btrfs/async-thread.h b/fs/btrfs/async-thread.h
index ad4d0647d1a6..8e52484cd461 100644
--- a/fs/btrfs/async-thread.h
+++ b/fs/btrfs/async-thread.h
@@ -21,6 +21,7 @@
 #define __BTRFS_ASYNC_THREAD_
 #include <linux/workqueue.h>
 
+struct btrfs_fs_info;
 struct btrfs_workqueue;
 /* Internal use only */
 struct __btrfs_workqueue;
@@ -67,7 +68,8 @@ BTRFS_WORK_HELPER_PROTO(scrubnc_helper);
 BTRFS_WORK_HELPER_PROTO(scrubparity_helper);
 
 
-struct btrfs_workqueue *btrfs_alloc_workqueue(const char *name,
+struct btrfs_workqueue *btrfs_alloc_workqueue(struct btrfs_fs_info *fs_info,
+					      const char *name,
 					      unsigned int flags,
 					      int limit_active,
 					      int thresh);
@@ -80,4 +82,6 @@ void btrfs_queue_work(struct btrfs_workqueue *wq,
 void btrfs_destroy_workqueue(struct btrfs_workqueue *wq);
 void btrfs_workqueue_set_max(struct btrfs_workqueue *wq, int max);
 void btrfs_set_work_high_priority(struct btrfs_work *work);
+struct btrfs_fs_info *btrfs_work_owner(struct btrfs_work *work);
+struct btrfs_fs_info *btrfs_workqueue_owner(struct __btrfs_workqueue *wq);
 #endif
diff --git a/fs/btrfs/backref.c b/fs/btrfs/backref.c
index 8bb3509099e8..85dc7ab8f89e 100644
--- a/fs/btrfs/backref.c
+++ b/fs/btrfs/backref.c
@@ -17,6 +17,7 @@
  */
 
 #include <linux/vmalloc.h>
+#include <linux/rbtree.h>
 #include "ctree.h"
 #include "disk-io.h"
 #include "backref.h"
@@ -34,6 +35,265 @@ struct extent_inode_elem {
 	struct extent_inode_elem *next;
 };
 
+/*
+ * ref_root is used as the root of the ref tree that hold a collection
+ * of unique references.
+ */
+struct ref_root {
+	struct rb_root rb_root;
+
+	/*
+	 * The unique_refs represents the number of ref_nodes with a positive
+	 * count stored in the tree. Even if a ref_node (the count is greater
+	 * than one) is added, the unique_refs will only increase by one.
+	 */
+	unsigned int unique_refs;
+};
+
+/* ref_node is used to store a unique reference to the ref tree. */
+struct ref_node {
+	struct rb_node rb_node;
+
+	/* For NORMAL_REF, otherwise all these fields should be set to 0 */
+	u64 root_id;
+	u64 object_id;
+	u64 offset;
+
+	/* For SHARED_REF, otherwise parent field should be set to 0 */
+	u64 parent;
+
+	/* Ref to the ref_mod of btrfs_delayed_ref_node */
+	int ref_mod;
+};
+
+/* Dynamically allocate and initialize a ref_root */
+static struct ref_root *ref_root_alloc(void)
+{
+	struct ref_root *ref_tree;
+
+	ref_tree = kmalloc(sizeof(*ref_tree), GFP_NOFS);
+	if (!ref_tree)
+		return NULL;
+
+	ref_tree->rb_root = RB_ROOT;
+	ref_tree->unique_refs = 0;
+
+	return ref_tree;
+}
+
+/* Free all nodes in the ref tree, and reinit ref_root */
+static void ref_root_fini(struct ref_root *ref_tree)
+{
+	struct ref_node *node;
+	struct rb_node *next;
+
+	while ((next = rb_first(&ref_tree->rb_root)) != NULL) {
+		node = rb_entry(next, struct ref_node, rb_node);
+		rb_erase(next, &ref_tree->rb_root);
+		kfree(node);
+	}
+
+	ref_tree->rb_root = RB_ROOT;
+	ref_tree->unique_refs = 0;
+}
+
+static void ref_root_free(struct ref_root *ref_tree)
+{
+	if (!ref_tree)
+		return;
+
+	ref_root_fini(ref_tree);
+	kfree(ref_tree);
+}
+
+/*
+ * Compare ref_node with (root_id, object_id, offset, parent)
+ *
+ * The function compares two ref_node a and b. It returns an integer less
+ * than, equal to, or greater than zero , respectively, to be less than, to
+ * equal, or be greater than b.
+ */
+static int ref_node_cmp(struct ref_node *a, struct ref_node *b)
+{
+	if (a->root_id < b->root_id)
+		return -1;
+	else if (a->root_id > b->root_id)
+		return 1;
+
+	if (a->object_id < b->object_id)
+		return -1;
+	else if (a->object_id > b->object_id)
+		return 1;
+
+	if (a->offset < b->offset)
+		return -1;
+	else if (a->offset > b->offset)
+		return 1;
+
+	if (a->parent < b->parent)
+		return -1;
+	else if (a->parent > b->parent)
+		return 1;
+
+	return 0;
+}
+
+/*
+ * Search ref_node with (root_id, object_id, offset, parent) in the tree
+ *
+ * if found, the pointer of the ref_node will be returned;
+ * if not found, NULL will be returned and pos will point to the rb_node for
+ * insert, pos_parent will point to pos'parent for insert;
+*/
+static struct ref_node *__ref_tree_search(struct ref_root *ref_tree,
+					  struct rb_node ***pos,
+					  struct rb_node **pos_parent,
+					  u64 root_id, u64 object_id,
+					  u64 offset, u64 parent)
+{
+	struct ref_node *cur = NULL;
+	struct ref_node entry;
+	int ret;
+
+	entry.root_id = root_id;
+	entry.object_id = object_id;
+	entry.offset = offset;
+	entry.parent = parent;
+
+	*pos = &ref_tree->rb_root.rb_node;
+
+	while (**pos) {
+		*pos_parent = **pos;
+		cur = rb_entry(*pos_parent, struct ref_node, rb_node);
+
+		ret = ref_node_cmp(cur, &entry);
+		if (ret > 0)
+			*pos = &(**pos)->rb_left;
+		else if (ret < 0)
+			*pos = &(**pos)->rb_right;
+		else
+			return cur;
+	}
+
+	return NULL;
+}
+
+/*
+ * Insert a ref_node to the ref tree
+ * @pos used for specifiy the position to insert
+ * @pos_parent for specifiy pos's parent
+ *
+ * success, return 0;
+ * ref_node already exists, return -EEXIST;
+*/
+static int ref_tree_insert(struct ref_root *ref_tree, struct rb_node **pos,
+			   struct rb_node *pos_parent, struct ref_node *ins)
+{
+	struct rb_node **p = NULL;
+	struct rb_node *parent = NULL;
+	struct ref_node *cur = NULL;
+
+	if (!pos) {
+		cur = __ref_tree_search(ref_tree, &p, &parent, ins->root_id,
+					ins->object_id, ins->offset,
+					ins->parent);
+		if (cur)
+			return -EEXIST;
+	} else {
+		p = pos;
+		parent = pos_parent;
+	}
+
+	rb_link_node(&ins->rb_node, parent, p);
+	rb_insert_color(&ins->rb_node, &ref_tree->rb_root);
+
+	return 0;
+}
+
+/* Erase and free ref_node, caller should update ref_root->unique_refs */
+static void ref_tree_remove(struct ref_root *ref_tree, struct ref_node *node)
+{
+	rb_erase(&node->rb_node, &ref_tree->rb_root);
+	kfree(node);
+}
+
+/*
+ * Update ref_root->unique_refs
+ *
+ * Call __ref_tree_search
+ *	1. if ref_node doesn't exist, ref_tree_insert this node, and update
+ *	ref_root->unique_refs:
+ *		if ref_node->ref_mod > 0, ref_root->unique_refs++;
+ *		if ref_node->ref_mod < 0, do noting;
+ *
+ *	2. if ref_node is found, then get origin ref_node->ref_mod, and update
+ *	ref_node->ref_mod.
+ *		if ref_node->ref_mod is equal to 0,then call ref_tree_remove
+ *
+ *		according to origin_mod and new_mod, update ref_root->items
+ *		+----------------+--------------+-------------+
+ *		|		 |new_count <= 0|new_count > 0|
+ *		+----------------+--------------+-------------+
+ *		|origin_count < 0|       0      |      1      |
+ *		+----------------+--------------+-------------+
+ *		|origin_count > 0|      -1      |      0      |
+ *		+----------------+--------------+-------------+
+ *
+ * In case of allocation failure, -ENOMEM is returned and the ref_tree stays
+ * unaltered.
+ * Success, return 0
+ */
+static int ref_tree_add(struct ref_root *ref_tree, u64 root_id, u64 object_id,
+			u64 offset, u64 parent, int count)
+{
+	struct ref_node *node = NULL;
+	struct rb_node **pos = NULL;
+	struct rb_node *pos_parent = NULL;
+	int origin_count;
+	int ret;
+
+	if (!count)
+		return 0;
+
+	node = __ref_tree_search(ref_tree, &pos, &pos_parent, root_id,
+				 object_id, offset, parent);
+	if (node == NULL) {
+		node = kmalloc(sizeof(*node), GFP_NOFS);
+		if (!node)
+			return -ENOMEM;
+
+		node->root_id = root_id;
+		node->object_id = object_id;
+		node->offset = offset;
+		node->parent = parent;
+		node->ref_mod = count;
+
+		ret = ref_tree_insert(ref_tree, pos, pos_parent, node);
+		ASSERT(!ret);
+		if (ret) {
+			kfree(node);
+			return ret;
+		}
+
+		ref_tree->unique_refs += node->ref_mod > 0 ? 1 : 0;
+
+		return 0;
+	}
+
+	origin_count = node->ref_mod;
+	node->ref_mod += count;
+
+	if (node->ref_mod > 0)
+		ref_tree->unique_refs += origin_count > 0 ? 0 : 1;
+	else if (node->ref_mod <= 0)
+		ref_tree->unique_refs += origin_count > 0 ? -1 : 0;
+
+	if (!node->ref_mod)
+		ref_tree_remove(ref_tree, node);
+
+	return 0;
+}
+
 static int check_extent_in_eb(struct btrfs_key *key, struct extent_buffer *eb,
 				struct btrfs_file_extent_item *fi,
 				u64 extent_item_pos,
@@ -139,7 +399,7 @@ int __init btrfs_prelim_ref_init(void)
 	btrfs_prelim_ref_cache = kmem_cache_create("btrfs_prelim_ref",
 					sizeof(struct __prelim_ref),
 					0,
-					SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD,
+					SLAB_MEM_SPREAD,
 					NULL);
 	if (!btrfs_prelim_ref_cache)
 		return -ENOMEM;
@@ -361,7 +621,7 @@ static int __resolve_indirect_ref(struct btrfs_fs_info *fs_info,
 		goto out;
 	}
 
-	if (btrfs_test_is_dummy_root(root)) {
+	if (btrfs_is_testing(fs_info)) {
 		srcu_read_unlock(&fs_info->subvol_srcu, index);
 		ret = -ENOENT;
 		goto out;
@@ -390,8 +650,8 @@ static int __resolve_indirect_ref(struct btrfs_fs_info *fs_info,
 	/* root node has been locked, we can release @subvol_srcu safely here */
 	srcu_read_unlock(&fs_info->subvol_srcu, index);
 
-	pr_debug("search slot in root %llu (level %d, ref count %d) returned "
-		 "%d for key (%llu %u %llu)\n",
+	btrfs_debug(fs_info,
+		"search slot in root %llu (level %d, ref count %d) returned %d for key (%llu %u %llu)",
 		 ref->root_id, level, ref->count, ret,
 		 ref->key_for_search.objectid, ref->key_for_search.type,
 		 ref->key_for_search.offset);
@@ -589,6 +849,7 @@ static void __merge_refs(struct list_head *head, int mode)
 
 			list_del(&ref2->list);
 			kmem_cache_free(btrfs_prelim_ref_cache, ref2);
+			cond_resched();
 		}
 
 	}
@@ -699,6 +960,7 @@ static int __add_delayed_refs(struct btrfs_delayed_ref_head *head, u64 seq,
 static int __add_inline_refs(struct btrfs_fs_info *fs_info,
 			     struct btrfs_path *path, u64 bytenr,
 			     int *info_level, struct list_head *prefs,
+			     struct ref_root *ref_tree,
 			     u64 *total_refs, u64 inum)
 {
 	int ret = 0;
@@ -766,6 +1028,13 @@ static int __add_inline_refs(struct btrfs_fs_info *fs_info,
 			count = btrfs_shared_data_ref_count(leaf, sdref);
 			ret = __add_prelim_ref(prefs, 0, NULL, 0, offset,
 					       bytenr, count, GFP_NOFS);
+			if (ref_tree) {
+				if (!ret)
+					ret = ref_tree_add(ref_tree, 0, 0, 0,
+							   bytenr, count);
+				if (!ret && ref_tree->unique_refs > 1)
+					ret = BACKREF_FOUND_SHARED;
+			}
 			break;
 		}
 		case BTRFS_TREE_BLOCK_REF_KEY:
@@ -793,6 +1062,15 @@ static int __add_inline_refs(struct btrfs_fs_info *fs_info,
 			root = btrfs_extent_data_ref_root(leaf, dref);
 			ret = __add_prelim_ref(prefs, root, &key, 0, 0,
 					       bytenr, count, GFP_NOFS);
+			if (ref_tree) {
+				if (!ret)
+					ret = ref_tree_add(ref_tree, root,
+							   key.objectid,
+							   key.offset, 0,
+							   count);
+				if (!ret && ref_tree->unique_refs > 1)
+					ret = BACKREF_FOUND_SHARED;
+			}
 			break;
 		}
 		default:
@@ -811,7 +1089,8 @@ static int __add_inline_refs(struct btrfs_fs_info *fs_info,
  */
 static int __add_keyed_refs(struct btrfs_fs_info *fs_info,
 			    struct btrfs_path *path, u64 bytenr,
-			    int info_level, struct list_head *prefs, u64 inum)
+			    int info_level, struct list_head *prefs,
+			    struct ref_root *ref_tree, u64 inum)
 {
 	struct btrfs_root *extent_root = fs_info->extent_root;
 	int ret;
@@ -854,6 +1133,13 @@ static int __add_keyed_refs(struct btrfs_fs_info *fs_info,
 			count = btrfs_shared_data_ref_count(leaf, sdref);
 			ret = __add_prelim_ref(prefs, 0, NULL, 0, key.offset,
 						bytenr, count, GFP_NOFS);
+			if (ref_tree) {
+				if (!ret)
+					ret = ref_tree_add(ref_tree, 0, 0, 0,
+							   bytenr, count);
+				if (!ret && ref_tree->unique_refs > 1)
+					ret = BACKREF_FOUND_SHARED;
+			}
 			break;
 		}
 		case BTRFS_TREE_BLOCK_REF_KEY:
@@ -882,6 +1168,15 @@ static int __add_keyed_refs(struct btrfs_fs_info *fs_info,
 			root = btrfs_extent_data_ref_root(leaf, dref);
 			ret = __add_prelim_ref(prefs, root, &key, 0, 0,
 					       bytenr, count, GFP_NOFS);
+			if (ref_tree) {
+				if (!ret)
+					ret = ref_tree_add(ref_tree, root,
+							   key.objectid,
+							   key.offset, 0,
+							   count);
+				if (!ret && ref_tree->unique_refs > 1)
+					ret = BACKREF_FOUND_SHARED;
+			}
 			break;
 		}
 		default:
@@ -908,13 +1203,16 @@ static int __add_keyed_refs(struct btrfs_fs_info *fs_info,
  * commit root.
  * The special case is for qgroup to search roots in commit_transaction().
  *
+ * If check_shared is set to 1, any extent has more than one ref item, will
+ * be returned BACKREF_FOUND_SHARED immediately.
+ *
  * FIXME some caching might speed things up
  */
 static int find_parent_nodes(struct btrfs_trans_handle *trans,
 			     struct btrfs_fs_info *fs_info, u64 bytenr,
 			     u64 time_seq, struct ulist *refs,
 			     struct ulist *roots, const u64 *extent_item_pos,
-			     u64 root_objectid, u64 inum)
+			     u64 root_objectid, u64 inum, int check_shared)
 {
 	struct btrfs_key key;
 	struct btrfs_path *path;
@@ -926,6 +1224,7 @@ static int find_parent_nodes(struct btrfs_trans_handle *trans,
 	struct list_head prefs;
 	struct __prelim_ref *ref;
 	struct extent_inode_elem *eie = NULL;
+	struct ref_root *ref_tree = NULL;
 	u64 total_refs = 0;
 
 	INIT_LIST_HEAD(&prefs);
@@ -957,6 +1256,18 @@ static int find_parent_nodes(struct btrfs_trans_handle *trans,
 again:
 	head = NULL;
 
+	if (check_shared) {
+		if (!ref_tree) {
+			ref_tree = ref_root_alloc();
+			if (!ref_tree) {
+				ret = -ENOMEM;
+				goto out;
+			}
+		} else {
+			ref_root_fini(ref_tree);
+		}
+	}
+
 	ret = btrfs_search_slot(trans, fs_info->extent_root, &key, path, 0, 0);
 	if (ret < 0)
 		goto out;
@@ -1001,6 +1312,36 @@ again:
 		} else {
 			spin_unlock(&delayed_refs->lock);
 		}
+
+		if (check_shared && !list_empty(&prefs_delayed)) {
+			/*
+			 * Add all delay_ref to the ref_tree and check if there
+			 * are multiple ref items added.
+			 */
+			list_for_each_entry(ref, &prefs_delayed, list) {
+				if (ref->key_for_search.type) {
+					ret = ref_tree_add(ref_tree,
+						ref->root_id,
+						ref->key_for_search.objectid,
+						ref->key_for_search.offset,
+						0, ref->count);
+					if (ret)
+						goto out;
+				} else {
+					ret = ref_tree_add(ref_tree, 0, 0, 0,
+						     ref->parent, ref->count);
+					if (ret)
+						goto out;
+				}
+
+			}
+
+			if (ref_tree->unique_refs > 1) {
+				ret = BACKREF_FOUND_SHARED;
+				goto out;
+			}
+
+		}
 	}
 
 	if (path->slots[0]) {
@@ -1016,11 +1357,13 @@ again:
 		     key.type == BTRFS_METADATA_ITEM_KEY)) {
 			ret = __add_inline_refs(fs_info, path, bytenr,
 						&info_level, &prefs,
-						&total_refs, inum);
+						ref_tree, &total_refs,
+						inum);
 			if (ret)
 				goto out;
 			ret = __add_keyed_refs(fs_info, path, bytenr,
-					       info_level, &prefs, inum);
+					       info_level, &prefs,
+					       ref_tree, inum);
 			if (ret)
 				goto out;
 		}
@@ -1105,6 +1448,7 @@ again:
 
 out:
 	btrfs_free_path(path);
+	ref_root_free(ref_tree);
 	while (!list_empty(&prefs)) {
 		ref = list_first_entry(&prefs, struct __prelim_ref, list);
 		list_del(&ref->list);
@@ -1158,8 +1502,8 @@ static int btrfs_find_all_leafs(struct btrfs_trans_handle *trans,
 	if (!*leafs)
 		return -ENOMEM;
 
-	ret = find_parent_nodes(trans, fs_info, bytenr,
-				time_seq, *leafs, NULL, extent_item_pos, 0, 0);
+	ret = find_parent_nodes(trans, fs_info, bytenr, time_seq,
+				*leafs, NULL, extent_item_pos, 0, 0, 0);
 	if (ret < 0 && ret != -ENOENT) {
 		free_leaf_list(*leafs);
 		return ret;
@@ -1201,8 +1545,8 @@ static int __btrfs_find_all_roots(struct btrfs_trans_handle *trans,
 
 	ULIST_ITER_INIT(&uiter);
 	while (1) {
-		ret = find_parent_nodes(trans, fs_info, bytenr,
-					time_seq, tmp, *roots, NULL, 0, 0);
+		ret = find_parent_nodes(trans, fs_info, bytenr, time_seq,
+					tmp, *roots, NULL, 0, 0, 0);
 		if (ret < 0 && ret != -ENOENT) {
 			ulist_free(tmp);
 			ulist_free(*roots);
@@ -1272,7 +1616,7 @@ int btrfs_check_shared(struct btrfs_trans_handle *trans,
 	ULIST_ITER_INIT(&uiter);
 	while (1) {
 		ret = find_parent_nodes(trans, fs_info, bytenr, elem.seq, tmp,
-					roots, NULL, root_objectid, inum);
+					roots, NULL, root_objectid, inum, 1);
 		if (ret == BACKREF_FOUND_SHARED) {
 			/* this is the only condition under which we return 1 */
 			ret = 1;
@@ -1491,7 +1835,8 @@ int extent_from_logical(struct btrfs_fs_info *fs_info, u64 logical,
 
 	if (found_key->objectid > logical ||
 	    found_key->objectid + size <= logical) {
-		pr_debug("logical %llu is not within any extent\n", logical);
+		btrfs_debug(fs_info,
+			"logical %llu is not within any extent", logical);
 		return -ENOENT;
 	}
 
@@ -1502,8 +1847,8 @@ int extent_from_logical(struct btrfs_fs_info *fs_info, u64 logical,
 	ei = btrfs_item_ptr(eb, path->slots[0], struct btrfs_extent_item);
 	flags = btrfs_extent_flags(eb, ei);
 
-	pr_debug("logical %llu is at position %llu within the extent (%llu "
-		 "EXTENT_ITEM %llu) flags %#llx size %u\n",
+	btrfs_debug(fs_info,
+		"logical %llu is at position %llu within the extent (%llu EXTENT_ITEM %llu) flags %#llx size %u",
 		 logical, logical - found_key->objectid, found_key->objectid,
 		 found_key->offset, flags, item_size);
 
@@ -1624,21 +1969,24 @@ int tree_backref_for_extent(unsigned long *ptr, struct extent_buffer *eb,
 	return 0;
 }
 
-static int iterate_leaf_refs(struct extent_inode_elem *inode_list,
-				u64 root, u64 extent_item_objectid,
-				iterate_extent_inodes_t *iterate, void *ctx)
+static int iterate_leaf_refs(struct btrfs_fs_info *fs_info,
+			     struct extent_inode_elem *inode_list,
+			     u64 root, u64 extent_item_objectid,
+			     iterate_extent_inodes_t *iterate, void *ctx)
 {
 	struct extent_inode_elem *eie;
 	int ret = 0;
 
 	for (eie = inode_list; eie; eie = eie->next) {
-		pr_debug("ref for %llu resolved, key (%llu EXTEND_DATA %llu), "
-			 "root %llu\n", extent_item_objectid,
-			 eie->inum, eie->offset, root);
+		btrfs_debug(fs_info,
+			    "ref for %llu resolved, key (%llu EXTEND_DATA %llu), root %llu",
+			    extent_item_objectid, eie->inum,
+			    eie->offset, root);
 		ret = iterate(eie->inum, eie->offset, root, ctx);
 		if (ret) {
-			pr_debug("stopping iteration for %llu due to ret=%d\n",
-				 extent_item_objectid, ret);
+			btrfs_debug(fs_info,
+				    "stopping iteration for %llu due to ret=%d",
+				    extent_item_objectid, ret);
 			break;
 		}
 	}
@@ -1666,7 +2014,7 @@ int iterate_extent_inodes(struct btrfs_fs_info *fs_info,
 	struct ulist_iterator ref_uiter;
 	struct ulist_iterator root_uiter;
 
-	pr_debug("resolving all inodes for extent %llu\n",
+	btrfs_debug(fs_info, "resolving all inodes for extent %llu",
 			extent_item_objectid);
 
 	if (!search_commit_root) {
@@ -1692,10 +2040,12 @@ int iterate_extent_inodes(struct btrfs_fs_info *fs_info,
 			break;
 		ULIST_ITER_INIT(&root_uiter);
 		while (!ret && (root_node = ulist_next(roots, &root_uiter))) {
-			pr_debug("root %llu references leaf %llu, data list "
-				 "%#llx\n", root_node->val, ref_node->val,
-				 ref_node->aux);
-			ret = iterate_leaf_refs((struct extent_inode_elem *)
+			btrfs_debug(fs_info,
+				    "root %llu references leaf %llu, data list %#llx",
+				    root_node->val, ref_node->val,
+				    ref_node->aux);
+			ret = iterate_leaf_refs(fs_info,
+						(struct extent_inode_elem *)
 						(uintptr_t)ref_node->aux,
 						root_node->val,
 						extent_item_objectid,
@@ -1791,9 +2141,9 @@ static int iterate_inode_refs(u64 inum, struct btrfs_root *fs_root,
 		for (cur = 0; cur < btrfs_item_size(eb, item); cur += len) {
 			name_len = btrfs_inode_ref_name_len(eb, iref);
 			/* path must be released before calling iterate()! */
-			pr_debug("following ref at offset %u for inode %llu in "
-				 "tree %llu\n", cur, found_key.objectid,
-				 fs_root->objectid);
+			btrfs_debug(fs_root->fs_info,
+				"following ref at offset %u for inode %llu in tree %llu",
+				cur, found_key.objectid, fs_root->objectid);
 			ret = iterate(parent, name_len,
 				      (unsigned long)(iref + 1), eb, ctx);
 			if (ret)
diff --git a/fs/btrfs/btrfs_inode.h b/fs/btrfs/btrfs_inode.h
index 4919aedb5fc1..1a8fa46ff87e 100644
--- a/fs/btrfs/btrfs_inode.h
+++ b/fs/btrfs/btrfs_inode.h
@@ -44,17 +44,6 @@
 #define BTRFS_INODE_IN_DELALLOC_LIST		9
 #define BTRFS_INODE_READDIO_NEED_LOCK		10
 #define BTRFS_INODE_HAS_PROPS		        11
-/*
- * The following 3 bits are meant only for the btree inode.
- * When any of them is set, it means an error happened while writing an
- * extent buffer belonging to:
- * 1) a non-log btree
- * 2) a log btree and first log sub-transaction
- * 3) a log btree and second log sub-transaction
- */
-#define BTRFS_INODE_BTREE_ERR		        12
-#define BTRFS_INODE_BTREE_LOG1_ERR		13
-#define BTRFS_INODE_BTREE_LOG2_ERR		14
 
 /* in memory btrfs inode */
 struct btrfs_inode {
diff --git a/fs/btrfs/check-integrity.c b/fs/btrfs/check-integrity.c
index b677a6ea6001..8e99251650b3 100644
--- a/fs/btrfs/check-integrity.c
+++ b/fs/btrfs/check-integrity.c
@@ -656,7 +656,7 @@ static int btrfsic_process_superblock(struct btrfsic_state *state,
 	BUG_ON(NULL == state);
 	selected_super = kzalloc(sizeof(*selected_super), GFP_NOFS);
 	if (NULL == selected_super) {
-		printk(KERN_INFO "btrfsic: error, kmalloc failed!\n");
+		pr_info("btrfsic: error, kmalloc failed!\n");
 		return -ENOMEM;
 	}
 
@@ -681,7 +681,7 @@ static int btrfsic_process_superblock(struct btrfsic_state *state,
 	}
 
 	if (NULL == state->latest_superblock) {
-		printk(KERN_INFO "btrfsic: no superblock found!\n");
+		pr_info("btrfsic: no superblock found!\n");
 		kfree(selected_super);
 		return -1;
 	}
@@ -698,13 +698,13 @@ static int btrfsic_process_superblock(struct btrfsic_state *state,
 			next_bytenr = btrfs_super_root(selected_super);
 			if (state->print_mask &
 			    BTRFSIC_PRINT_MASK_ROOT_CHUNK_LOG_TREE_LOCATION)
-				printk(KERN_INFO "root@%llu\n", next_bytenr);
+				pr_info("root@%llu\n", next_bytenr);
 			break;
 		case 1:
 			next_bytenr = btrfs_super_chunk_root(selected_super);
 			if (state->print_mask &
 			    BTRFSIC_PRINT_MASK_ROOT_CHUNK_LOG_TREE_LOCATION)
-				printk(KERN_INFO "chunk@%llu\n", next_bytenr);
+				pr_info("chunk@%llu\n", next_bytenr);
 			break;
 		case 2:
 			next_bytenr = btrfs_super_log_root(selected_super);
@@ -712,7 +712,7 @@ static int btrfsic_process_superblock(struct btrfsic_state *state,
 				continue;
 			if (state->print_mask &
 			    BTRFSIC_PRINT_MASK_ROOT_CHUNK_LOG_TREE_LOCATION)
-				printk(KERN_INFO "log@%llu\n", next_bytenr);
+				pr_info("log@%llu\n", next_bytenr);
 			break;
 		}
 
@@ -720,7 +720,7 @@ static int btrfsic_process_superblock(struct btrfsic_state *state,
 		    btrfs_num_copies(state->root->fs_info,
 				     next_bytenr, state->metablock_size);
 		if (state->print_mask & BTRFSIC_PRINT_MASK_NUM_COPIES)
-			printk(KERN_INFO "num_copies(log_bytenr=%llu) = %d\n",
+			pr_info("num_copies(log_bytenr=%llu) = %d\n",
 			       next_bytenr, num_copies);
 
 		for (mirror_num = 1; mirror_num <= num_copies; mirror_num++) {
@@ -733,9 +733,7 @@ static int btrfsic_process_superblock(struct btrfsic_state *state,
 						&tmp_next_block_ctx,
 						mirror_num);
 			if (ret) {
-				printk(KERN_INFO "btrfsic:"
-				       " btrfsic_map_block(root @%llu,"
-				       " mirror %d) failed!\n",
+				pr_info("btrfsic: btrfsic_map_block(root @%llu, mirror %d) failed!\n",
 				       next_bytenr, mirror_num);
 				kfree(selected_super);
 				return -1;
@@ -758,8 +756,7 @@ static int btrfsic_process_superblock(struct btrfsic_state *state,
 
 			ret = btrfsic_read_block(state, &tmp_next_block_ctx);
 			if (ret < (int)PAGE_SIZE) {
-				printk(KERN_INFO
-				       "btrfsic: read @logical %llu failed!\n",
+				pr_info("btrfsic: read @logical %llu failed!\n",
 				       tmp_next_block_ctx.start);
 				btrfsic_release_block_ctx(&tmp_next_block_ctx);
 				kfree(selected_super);
@@ -820,7 +817,7 @@ static int btrfsic_process_superblock_dev_mirror(
 	if (NULL == superblock_tmp) {
 		superblock_tmp = btrfsic_block_alloc();
 		if (NULL == superblock_tmp) {
-			printk(KERN_INFO "btrfsic: error, kmalloc failed!\n");
+			pr_info("btrfsic: error, kmalloc failed!\n");
 			brelse(bh);
 			return -1;
 		}
@@ -894,7 +891,7 @@ static int btrfsic_process_superblock_dev_mirror(
 		    btrfs_num_copies(state->root->fs_info,
 				     next_bytenr, state->metablock_size);
 		if (state->print_mask & BTRFSIC_PRINT_MASK_NUM_COPIES)
-			printk(KERN_INFO "num_copies(log_bytenr=%llu) = %d\n",
+			pr_info("num_copies(log_bytenr=%llu) = %d\n",
 			       next_bytenr, num_copies);
 		for (mirror_num = 1; mirror_num <= num_copies; mirror_num++) {
 			struct btrfsic_block *next_block;
@@ -905,8 +902,7 @@ static int btrfsic_process_superblock_dev_mirror(
 					      state->metablock_size,
 					      &tmp_next_block_ctx,
 					      mirror_num)) {
-				printk(KERN_INFO "btrfsic: btrfsic_map_block("
-				       "bytenr @%llu, mirror %d) failed!\n",
+				pr_info("btrfsic: btrfsic_map_block(bytenr @%llu, mirror %d) failed!\n",
 				       next_bytenr, mirror_num);
 				brelse(bh);
 				return -1;
@@ -948,7 +944,7 @@ static struct btrfsic_stack_frame *btrfsic_stack_frame_alloc(void)
 
 	sf = kzalloc(sizeof(*sf), GFP_NOFS);
 	if (NULL == sf)
-		printk(KERN_INFO "btrfsic: alloc memory failed!\n");
+		pr_info("btrfsic: alloc memory failed!\n");
 	else
 		sf->magic = BTRFSIC_BLOCK_STACK_FRAME_MAGIC_NUMBER;
 	return sf;
@@ -994,9 +990,7 @@ continue_with_new_stack_frame:
 			sf->nr = btrfs_stack_header_nritems(&leafhdr->header);
 
 			if (state->print_mask & BTRFSIC_PRINT_MASK_VERBOSE)
-				printk(KERN_INFO
-				       "leaf %llu items %d generation %llu"
-				       " owner %llu\n",
+				pr_info("leaf %llu items %d generation %llu owner %llu\n",
 				       sf->block_ctx->start, sf->nr,
 				       btrfs_stack_header_generation(
 					       &leafhdr->header),
@@ -1023,8 +1017,7 @@ continue_with_current_leaf_stack_frame:
 			if (disk_item_offset + sizeof(struct btrfs_item) >
 			    sf->block_ctx->len) {
 leaf_item_out_of_bounce_error:
-				printk(KERN_INFO
-				       "btrfsic: leaf item out of bounce at logical %llu, dev %s\n",
+				pr_info("btrfsic: leaf item out of bounce at logical %llu, dev %s\n",
 				       sf->block_ctx->start,
 				       sf->block_ctx->dev->name);
 				goto one_stack_frame_backwards;
@@ -1120,8 +1113,7 @@ leaf_item_out_of_bounce_error:
 			sf->nr = btrfs_stack_header_nritems(&nodehdr->header);
 
 			if (state->print_mask & BTRFSIC_PRINT_MASK_VERBOSE)
-				printk(KERN_INFO "node %llu level %d items %d"
-				       " generation %llu owner %llu\n",
+				pr_info("node %llu level %d items %d generation %llu owner %llu\n",
 				       sf->block_ctx->start,
 				       nodehdr->header.level, sf->nr,
 				       btrfs_stack_header_generation(
@@ -1145,8 +1137,7 @@ continue_with_current_node_stack_frame:
 					  (uintptr_t)nodehdr;
 			if (key_ptr_offset + sizeof(struct btrfs_key_ptr) >
 			    sf->block_ctx->len) {
-				printk(KERN_INFO
-				       "btrfsic: node item out of bounce at logical %llu, dev %s\n",
+				pr_info("btrfsic: node item out of bounce at logical %llu, dev %s\n",
 				       sf->block_ctx->start,
 				       sf->block_ctx->dev->name);
 				goto one_stack_frame_backwards;
@@ -1275,7 +1266,7 @@ static int btrfsic_create_link_to_next_block(
 		    btrfs_num_copies(state->root->fs_info,
 				     next_bytenr, state->metablock_size);
 		if (state->print_mask & BTRFSIC_PRINT_MASK_NUM_COPIES)
-			printk(KERN_INFO "num_copies(log_bytenr=%llu) = %d\n",
+			pr_info("num_copies(log_bytenr=%llu) = %d\n",
 			       next_bytenr, *num_copiesp);
 		*mirror_nump = 1;
 	}
@@ -1284,15 +1275,13 @@ static int btrfsic_create_link_to_next_block(
 		return 0;
 
 	if (state->print_mask & BTRFSIC_PRINT_MASK_VERBOSE)
-		printk(KERN_INFO
-		       "btrfsic_create_link_to_next_block(mirror_num=%d)\n",
+		pr_info("btrfsic_create_link_to_next_block(mirror_num=%d)\n",
 		       *mirror_nump);
 	ret = btrfsic_map_block(state, next_bytenr,
 				state->metablock_size,
 				next_block_ctx, *mirror_nump);
 	if (ret) {
-		printk(KERN_INFO
-		       "btrfsic: btrfsic_map_block(@%llu, mirror=%d) failed!\n",
+		pr_info("btrfsic: btrfsic_map_block(@%llu, mirror=%d) failed!\n",
 		       next_bytenr, *mirror_nump);
 		btrfsic_release_block_ctx(next_block_ctx);
 		*next_blockp = NULL;
@@ -1318,16 +1307,14 @@ static int btrfsic_create_link_to_next_block(
 			if (next_block->logical_bytenr != next_bytenr &&
 			    !(!next_block->is_metadata &&
 			      0 == next_block->logical_bytenr))
-				printk(KERN_INFO
-				       "Referenced block @%llu (%s/%llu/%d) found in hash table, %c, bytenr mismatch (!= stored %llu).\n",
+				pr_info("Referenced block @%llu (%s/%llu/%d) found in hash table, %c, bytenr mismatch (!= stored %llu).\n",
 				       next_bytenr, next_block_ctx->dev->name,
 				       next_block_ctx->dev_bytenr, *mirror_nump,
 				       btrfsic_get_block_type(state,
 							      next_block),
 				       next_block->logical_bytenr);
 			else
-				printk(KERN_INFO
-				       "Referenced block @%llu (%s/%llu/%d) found in hash table, %c.\n",
+				pr_info("Referenced block @%llu (%s/%llu/%d) found in hash table, %c.\n",
 				       next_bytenr, next_block_ctx->dev->name,
 				       next_block_ctx->dev_bytenr, *mirror_nump,
 				       btrfsic_get_block_type(state,
@@ -1348,7 +1335,7 @@ static int btrfsic_create_link_to_next_block(
 	if (NULL == l) {
 		l = btrfsic_block_link_alloc();
 		if (NULL == l) {
-			printk(KERN_INFO "btrfsic: error, kmalloc failed!\n");
+			pr_info("btrfsic: error, kmalloc failed!\n");
 			btrfsic_release_block_ctx(next_block_ctx);
 			*next_blockp = NULL;
 			return -1;
@@ -1381,8 +1368,7 @@ static int btrfsic_create_link_to_next_block(
 	if (limit_nesting > 0 && did_alloc_block_link) {
 		ret = btrfsic_read_block(state, next_block_ctx);
 		if (ret < (int)next_block_ctx->len) {
-			printk(KERN_INFO
-			       "btrfsic: read block @logical %llu failed!\n",
+			pr_info("btrfsic: read block @logical %llu failed!\n",
 			       next_bytenr);
 			btrfsic_release_block_ctx(next_block_ctx);
 			*next_blockp = NULL;
@@ -1417,8 +1403,7 @@ static int btrfsic_handle_extent_data(
 	if (file_extent_item_offset +
 	    offsetof(struct btrfs_file_extent_item, disk_num_bytes) >
 	    block_ctx->len) {
-		printk(KERN_INFO
-		       "btrfsic: file item out of bounce at logical %llu, dev %s\n",
+		pr_info("btrfsic: file item out of bounce at logical %llu, dev %s\n",
 		       block_ctx->start, block_ctx->dev->name);
 		return -1;
 	}
@@ -1429,7 +1414,7 @@ static int btrfsic_handle_extent_data(
 	if (BTRFS_FILE_EXTENT_REG != file_extent_item.type ||
 	    btrfs_stack_file_extent_disk_bytenr(&file_extent_item) == 0) {
 		if (state->print_mask & BTRFSIC_PRINT_MASK_VERY_VERBOSE)
-			printk(KERN_INFO "extent_data: type %u, disk_bytenr = %llu\n",
+			pr_info("extent_data: type %u, disk_bytenr = %llu\n",
 			       file_extent_item.type,
 			       btrfs_stack_file_extent_disk_bytenr(
 			       &file_extent_item));
@@ -1438,8 +1423,7 @@ static int btrfsic_handle_extent_data(
 
 	if (file_extent_item_offset + sizeof(struct btrfs_file_extent_item) >
 	    block_ctx->len) {
-		printk(KERN_INFO
-		       "btrfsic: file item out of bounce at logical %llu, dev %s\n",
+		pr_info("btrfsic: file item out of bounce at logical %llu, dev %s\n",
 		       block_ctx->start, block_ctx->dev->name);
 		return -1;
 	}
@@ -1457,8 +1441,7 @@ static int btrfsic_handle_extent_data(
 	generation = btrfs_stack_file_extent_generation(&file_extent_item);
 
 	if (state->print_mask & BTRFSIC_PRINT_MASK_VERY_VERBOSE)
-		printk(KERN_INFO "extent_data: type %u, disk_bytenr = %llu,"
-		       " offset = %llu, num_bytes = %llu\n",
+		pr_info("extent_data: type %u, disk_bytenr = %llu, offset = %llu, num_bytes = %llu\n",
 		       file_extent_item.type,
 		       btrfs_stack_file_extent_disk_bytenr(&file_extent_item),
 		       btrfs_stack_file_extent_offset(&file_extent_item),
@@ -1477,7 +1460,7 @@ static int btrfsic_handle_extent_data(
 		    btrfs_num_copies(state->root->fs_info,
 				     next_bytenr, state->datablock_size);
 		if (state->print_mask & BTRFSIC_PRINT_MASK_NUM_COPIES)
-			printk(KERN_INFO "num_copies(log_bytenr=%llu) = %d\n",
+			pr_info("num_copies(log_bytenr=%llu) = %d\n",
 			       next_bytenr, num_copies);
 		for (mirror_num = 1; mirror_num <= num_copies; mirror_num++) {
 			struct btrfsic_block_data_ctx next_block_ctx;
@@ -1485,19 +1468,16 @@ static int btrfsic_handle_extent_data(
 			int block_was_created;
 
 			if (state->print_mask & BTRFSIC_PRINT_MASK_VERBOSE)
-				printk(KERN_INFO "btrfsic_handle_extent_data("
-				       "mirror_num=%d)\n", mirror_num);
+				pr_info("btrfsic_handle_extent_data(mirror_num=%d)\n",
+					mirror_num);
 			if (state->print_mask & BTRFSIC_PRINT_MASK_VERY_VERBOSE)
-				printk(KERN_INFO
-				       "\tdisk_bytenr = %llu, num_bytes %u\n",
+				pr_info("\tdisk_bytenr = %llu, num_bytes %u\n",
 				       next_bytenr, chunk_len);
 			ret = btrfsic_map_block(state, next_bytenr,
 						chunk_len, &next_block_ctx,
 						mirror_num);
 			if (ret) {
-				printk(KERN_INFO
-				       "btrfsic: btrfsic_map_block(@%llu,"
-				       " mirror=%d) failed!\n",
+				pr_info("btrfsic: btrfsic_map_block(@%llu, mirror=%d) failed!\n",
 				       next_bytenr, mirror_num);
 				return -1;
 			}
@@ -1512,8 +1492,7 @@ static int btrfsic_handle_extent_data(
 					mirror_num,
 					&block_was_created);
 			if (NULL == next_block) {
-				printk(KERN_INFO
-				       "btrfsic: error, kmalloc failed!\n");
+				pr_info("btrfsic: error, kmalloc failed!\n");
 				btrfsic_release_block_ctx(&next_block_ctx);
 				return -1;
 			}
@@ -1523,12 +1502,7 @@ static int btrfsic_handle_extent_data(
 				    next_block->logical_bytenr != next_bytenr &&
 				    !(!next_block->is_metadata &&
 				      0 == next_block->logical_bytenr)) {
-					printk(KERN_INFO
-					       "Referenced block"
-					       " @%llu (%s/%llu/%d)"
-					       " found in hash table, D,"
-					       " bytenr mismatch"
-					       " (!= stored %llu).\n",
+					pr_info("Referenced block @%llu (%s/%llu/%d) found in hash table, D, bytenr mismatch (!= stored %llu).\n",
 					       next_bytenr,
 					       next_block_ctx.dev->name,
 					       next_block_ctx.dev_bytenr,
@@ -1592,7 +1566,7 @@ static int btrfsic_map_block(struct btrfsic_state *state, u64 bytenr, u32 len,
 	kfree(multi);
 	if (NULL == block_ctx_out->dev) {
 		ret = -ENXIO;
-		printk(KERN_INFO "btrfsic: error, cannot lookup dev (#1)!\n");
+		pr_info("btrfsic: error, cannot lookup dev (#1)!\n");
 	}
 
 	return ret;
@@ -1638,8 +1612,7 @@ static int btrfsic_read_block(struct btrfsic_state *state,
 	BUG_ON(block_ctx->pagev);
 	BUG_ON(block_ctx->mem_to_free);
 	if (block_ctx->dev_bytenr & ((u64)PAGE_SIZE - 1)) {
-		printk(KERN_INFO
-		       "btrfsic: read_block() with unaligned bytenr %llu\n",
+		pr_info("btrfsic: read_block() with unaligned bytenr %llu\n",
 		       block_ctx->dev_bytenr);
 		return -1;
 	}
@@ -1666,13 +1639,13 @@ static int btrfsic_read_block(struct btrfsic_state *state,
 
 		bio = btrfs_io_bio_alloc(GFP_NOFS, num_pages - i);
 		if (!bio) {
-			printk(KERN_INFO
-			       "btrfsic: bio_alloc() for %u pages failed!\n",
+			pr_info("btrfsic: bio_alloc() for %u pages failed!\n",
 			       num_pages - i);
 			return -1;
 		}
 		bio->bi_bdev = block_ctx->dev->bdev;
 		bio->bi_iter.bi_sector = dev_bytenr >> 9;
+		bio_set_op_attrs(bio, REQ_OP_READ, 0);
 
 		for (j = i; j < num_pages; j++) {
 			ret = bio_add_page(bio, block_ctx->pagev[j],
@@ -1681,13 +1654,11 @@ static int btrfsic_read_block(struct btrfsic_state *state,
 				break;
 		}
 		if (j == i) {
-			printk(KERN_INFO
-			       "btrfsic: error, failed to add a single page!\n");
+			pr_info("btrfsic: error, failed to add a single page!\n");
 			return -1;
 		}
-		if (submit_bio_wait(READ, bio)) {
-			printk(KERN_INFO
-			       "btrfsic: read error at logical %llu dev %s!\n",
+		if (submit_bio_wait(bio)) {
+			pr_info("btrfsic: read error at logical %llu dev %s!\n",
 			       block_ctx->start, block_ctx->dev->name);
 			bio_put(bio);
 			return -1;
@@ -1699,7 +1670,7 @@ static int btrfsic_read_block(struct btrfsic_state *state,
 	for (i = 0; i < num_pages; i++) {
 		block_ctx->datav[i] = kmap(block_ctx->pagev[i]);
 		if (!block_ctx->datav[i]) {
-			printk(KERN_INFO "btrfsic: kmap() failed (dev %s)!\n",
+			pr_info("btrfsic: kmap() failed (dev %s)!\n",
 			       block_ctx->dev->name);
 			return -1;
 		}
@@ -1714,19 +1685,17 @@ static void btrfsic_dump_database(struct btrfsic_state *state)
 
 	BUG_ON(NULL == state);
 
-	printk(KERN_INFO "all_blocks_list:\n");
+	pr_info("all_blocks_list:\n");
 	list_for_each_entry(b_all, &state->all_blocks_list, all_blocks_node) {
 		const struct btrfsic_block_link *l;
 
-		printk(KERN_INFO "%c-block @%llu (%s/%llu/%d)\n",
+		pr_info("%c-block @%llu (%s/%llu/%d)\n",
 		       btrfsic_get_block_type(state, b_all),
 		       b_all->logical_bytenr, b_all->dev_state->name,
 		       b_all->dev_bytenr, b_all->mirror_num);
 
 		list_for_each_entry(l, &b_all->ref_to_list, node_ref_to) {
-			printk(KERN_INFO " %c @%llu (%s/%llu/%d)"
-			       " refers %u* to"
-			       " %c @%llu (%s/%llu/%d)\n",
+			pr_info(" %c @%llu (%s/%llu/%d) refers %u* to %c @%llu (%s/%llu/%d)\n",
 			       btrfsic_get_block_type(state, b_all),
 			       b_all->logical_bytenr, b_all->dev_state->name,
 			       b_all->dev_bytenr, b_all->mirror_num,
@@ -1739,9 +1708,7 @@ static void btrfsic_dump_database(struct btrfsic_state *state)
 		}
 
 		list_for_each_entry(l, &b_all->ref_from_list, node_ref_from) {
-			printk(KERN_INFO " %c @%llu (%s/%llu/%d)"
-			       " is ref %u* from"
-			       " %c @%llu (%s/%llu/%d)\n",
+			pr_info(" %c @%llu (%s/%llu/%d) is ref %u* from %c @%llu (%s/%llu/%d)\n",
 			       btrfsic_get_block_type(state, b_all),
 			       b_all->logical_bytenr, b_all->dev_state->name,
 			       b_all->dev_bytenr, b_all->mirror_num,
@@ -1753,7 +1720,7 @@ static void btrfsic_dump_database(struct btrfsic_state *state)
 			       l->block_ref_from->mirror_num);
 		}
 
-		printk(KERN_INFO "\n");
+		pr_info("\n");
 	}
 }
 
@@ -1828,8 +1795,7 @@ again:
 						    mapped_datav[0]);
 			if (num_pages * PAGE_SIZE <
 			    BTRFS_SUPER_INFO_SIZE) {
-				printk(KERN_INFO
-				       "btrfsic: cannot work with too short bios!\n");
+				pr_info("btrfsic: cannot work with too short bios!\n");
 				return;
 			}
 			is_metadata = 1;
@@ -1837,8 +1803,7 @@ again:
 			processed_len = BTRFS_SUPER_INFO_SIZE;
 			if (state->print_mask &
 			    BTRFSIC_PRINT_MASK_TREE_BEFORE_SB_WRITE) {
-				printk(KERN_INFO
-				       "[before new superblock is written]:\n");
+				pr_info("[before new superblock is written]:\n");
 				btrfsic_dump_tree_sub(state, block, 0);
 			}
 		}
@@ -1846,8 +1811,7 @@ again:
 			if (!block->is_superblock) {
 				if (num_pages * PAGE_SIZE <
 				    state->metablock_size) {
-					printk(KERN_INFO
-					       "btrfsic: cannot work with too short bios!\n");
+					pr_info("btrfsic: cannot work with too short bios!\n");
 					return;
 				}
 				processed_len = state->metablock_size;
@@ -1862,8 +1826,7 @@ again:
 				if (block->logical_bytenr != bytenr &&
 				    !(!block->is_metadata &&
 				      block->logical_bytenr == 0))
-					printk(KERN_INFO
-					       "Written block @%llu (%s/%llu/%d) found in hash table, %c, bytenr mismatch (!= stored %llu).\n",
+					pr_info("Written block @%llu (%s/%llu/%d) found in hash table, %c, bytenr mismatch (!= stored %llu).\n",
 					       bytenr, dev_state->name,
 					       dev_bytenr,
 					       block->mirror_num,
@@ -1871,8 +1834,7 @@ again:
 								      block),
 					       block->logical_bytenr);
 				else
-					printk(KERN_INFO
-					       "Written block @%llu (%s/%llu/%d) found in hash table, %c.\n",
+					pr_info("Written block @%llu (%s/%llu/%d) found in hash table, %c.\n",
 					       bytenr, dev_state->name,
 					       dev_bytenr, block->mirror_num,
 					       btrfsic_get_block_type(state,
@@ -1882,33 +1844,24 @@ again:
 		} else {
 			if (num_pages * PAGE_SIZE <
 			    state->datablock_size) {
-				printk(KERN_INFO
-				       "btrfsic: cannot work with too short bios!\n");
+				pr_info("btrfsic: cannot work with too short bios!\n");
 				return;
 			}
 			processed_len = state->datablock_size;
 			bytenr = block->logical_bytenr;
 			if (state->print_mask & BTRFSIC_PRINT_MASK_VERBOSE)
-				printk(KERN_INFO
-				       "Written block @%llu (%s/%llu/%d)"
-				       " found in hash table, %c.\n",
+				pr_info("Written block @%llu (%s/%llu/%d) found in hash table, %c.\n",
 				       bytenr, dev_state->name, dev_bytenr,
 				       block->mirror_num,
 				       btrfsic_get_block_type(state, block));
 		}
 
 		if (state->print_mask & BTRFSIC_PRINT_MASK_VERBOSE)
-			printk(KERN_INFO
-			       "ref_to_list: %cE, ref_from_list: %cE\n",
+			pr_info("ref_to_list: %cE, ref_from_list: %cE\n",
 			       list_empty(&block->ref_to_list) ? ' ' : '!',
 			       list_empty(&block->ref_from_list) ? ' ' : '!');
 		if (btrfsic_is_block_ref_by_superblock(state, block, 0)) {
-			printk(KERN_INFO "btrfs: attempt to overwrite %c-block"
-			       " @%llu (%s/%llu/%d), old(gen=%llu,"
-			       " objectid=%llu, type=%d, offset=%llu),"
-			       " new(gen=%llu),"
-			       " which is referenced by most recent superblock"
-			       " (superblockgen=%llu)!\n",
+			pr_info("btrfs: attempt to overwrite %c-block @%llu (%s/%llu/%d), old(gen=%llu, objectid=%llu, type=%d, offset=%llu), new(gen=%llu), which is referenced by most recent superblock (superblockgen=%llu)!\n",
 			       btrfsic_get_block_type(state, block), bytenr,
 			       dev_state->name, dev_bytenr, block->mirror_num,
 			       block->generation,
@@ -1922,9 +1875,7 @@ again:
 		}
 
 		if (!block->is_iodone && !block->never_written) {
-			printk(KERN_INFO "btrfs: attempt to overwrite %c-block"
-			       " @%llu (%s/%llu/%d), oldgen=%llu, newgen=%llu,"
-			       " which is not yet iodone!\n",
+			pr_info("btrfs: attempt to overwrite %c-block @%llu (%s/%llu/%d), oldgen=%llu, newgen=%llu, which is not yet iodone!\n",
 			       btrfsic_get_block_type(state, block), bytenr,
 			       dev_state->name, dev_bytenr, block->mirror_num,
 			       block->generation,
@@ -2022,8 +1973,7 @@ again:
 						mapped_datav[0]);
 				if (state->print_mask &
 				    BTRFSIC_PRINT_MASK_TREE_AFTER_SB_WRITE) {
-					printk(KERN_INFO
-					"[after new superblock is written]:\n");
+					pr_info("[after new superblock is written]:\n");
 					btrfsic_dump_tree_sub(state, block, 0);
 				}
 			} else {
@@ -2035,9 +1985,7 @@ again:
 						0, 0);
 			}
 			if (ret)
-				printk(KERN_INFO
-				       "btrfsic: btrfsic_process_metablock"
-				       "(root @%llu) failed!\n",
+				pr_info("btrfsic: btrfsic_process_metablock(root @%llu) failed!\n",
 				       dev_bytenr);
 		} else {
 			block->is_metadata = 0;
@@ -2064,8 +2012,7 @@ again:
 		if (!is_metadata) {
 			processed_len = state->datablock_size;
 			if (state->print_mask & BTRFSIC_PRINT_MASK_VERBOSE)
-				printk(KERN_INFO "Written block (%s/%llu/?)"
-				       " !found in hash table, D.\n",
+				pr_info("Written block (%s/%llu/?) !found in hash table, D.\n",
 				       dev_state->name, dev_bytenr);
 			if (!state->include_extent_data) {
 				/* ignore that written D block */
@@ -2083,9 +2030,7 @@ again:
 			btrfsic_cmp_log_and_dev_bytenr(state, bytenr, dev_state,
 						       dev_bytenr);
 			if (state->print_mask & BTRFSIC_PRINT_MASK_VERBOSE)
-				printk(KERN_INFO
-				       "Written block @%llu (%s/%llu/?)"
-				       " !found in hash table, M.\n",
+				pr_info("Written block @%llu (%s/%llu/?) !found in hash table, M.\n",
 				       bytenr, dev_state->name, dev_bytenr);
 		}
 
@@ -2099,7 +2044,7 @@ again:
 
 		block = btrfsic_block_alloc();
 		if (NULL == block) {
-			printk(KERN_INFO "btrfsic: error, kmalloc failed!\n");
+			pr_info("btrfsic: error, kmalloc failed!\n");
 			btrfsic_release_block_ctx(&block_ctx);
 			goto continue_loop;
 		}
@@ -2149,8 +2094,7 @@ again:
 			block->next_in_same_bio = NULL;
 		}
 		if (state->print_mask & BTRFSIC_PRINT_MASK_VERBOSE)
-			printk(KERN_INFO
-			       "New written %c-block @%llu (%s/%llu/%d)\n",
+			pr_info("New written %c-block @%llu (%s/%llu/%d)\n",
 			       is_metadata ? 'M' : 'D',
 			       block->logical_bytenr, block->dev_state->name,
 			       block->dev_bytenr, block->mirror_num);
@@ -2161,9 +2105,7 @@ again:
 			ret = btrfsic_process_metablock(state, block,
 							&block_ctx, 0, 0);
 			if (ret)
-				printk(KERN_INFO
-				       "btrfsic: process_metablock(root @%llu)"
-				       " failed!\n",
+				pr_info("btrfsic: process_metablock(root @%llu) failed!\n",
 				       dev_bytenr);
 		}
 		btrfsic_release_block_ctx(&block_ctx);
@@ -2198,20 +2140,18 @@ static void btrfsic_bio_end_io(struct bio *bp)
 
 		if ((dev_state->state->print_mask &
 		     BTRFSIC_PRINT_MASK_END_IO_BIO_BH))
-			printk(KERN_INFO
-			       "bio_end_io(err=%d) for %c @%llu (%s/%llu/%d)\n",
+			pr_info("bio_end_io(err=%d) for %c @%llu (%s/%llu/%d)\n",
 			       bp->bi_error,
 			       btrfsic_get_block_type(dev_state->state, block),
 			       block->logical_bytenr, dev_state->name,
 			       block->dev_bytenr, block->mirror_num);
 		next_block = block->next_in_same_bio;
 		block->iodone_w_error = iodone_w_error;
-		if (block->submit_bio_bh_rw & REQ_FLUSH) {
+		if (block->submit_bio_bh_rw & REQ_PREFLUSH) {
 			dev_state->last_flush_gen++;
 			if ((dev_state->state->print_mask &
 			     BTRFSIC_PRINT_MASK_END_IO_BIO_BH))
-				printk(KERN_INFO
-				       "bio_end_io() new %s flush_gen=%llu\n",
+				pr_info("bio_end_io() new %s flush_gen=%llu\n",
 				       dev_state->name,
 				       dev_state->last_flush_gen);
 		}
@@ -2234,20 +2174,18 @@ static void btrfsic_bh_end_io(struct buffer_head *bh, int uptodate)
 	BUG_ON(NULL == block);
 	dev_state = block->dev_state;
 	if ((dev_state->state->print_mask & BTRFSIC_PRINT_MASK_END_IO_BIO_BH))
-		printk(KERN_INFO
-		       "bh_end_io(error=%d) for %c @%llu (%s/%llu/%d)\n",
+		pr_info("bh_end_io(error=%d) for %c @%llu (%s/%llu/%d)\n",
 		       iodone_w_error,
 		       btrfsic_get_block_type(dev_state->state, block),
 		       block->logical_bytenr, block->dev_state->name,
 		       block->dev_bytenr, block->mirror_num);
 
 	block->iodone_w_error = iodone_w_error;
-	if (block->submit_bio_bh_rw & REQ_FLUSH) {
+	if (block->submit_bio_bh_rw & REQ_PREFLUSH) {
 		dev_state->last_flush_gen++;
 		if ((dev_state->state->print_mask &
 		     BTRFSIC_PRINT_MASK_END_IO_BIO_BH))
-			printk(KERN_INFO
-			       "bh_end_io() new %s flush_gen=%llu\n",
+			pr_info("bh_end_io() new %s flush_gen=%llu\n",
 			       dev_state->name, dev_state->last_flush_gen);
 	}
 	if (block->submit_bio_bh_rw & REQ_FUA)
@@ -2270,9 +2208,7 @@ static int btrfsic_process_written_superblock(
 	if (!(superblock->generation > state->max_superblock_generation ||
 	      0 == state->max_superblock_generation)) {
 		if (state->print_mask & BTRFSIC_PRINT_MASK_SUPERBLOCK_WRITE)
-			printk(KERN_INFO
-			       "btrfsic: superblock @%llu (%s/%llu/%d)"
-			       " with old gen %llu <= %llu\n",
+			pr_info("btrfsic: superblock @%llu (%s/%llu/%d) with old gen %llu <= %llu\n",
 			       superblock->logical_bytenr,
 			       superblock->dev_state->name,
 			       superblock->dev_bytenr, superblock->mirror_num,
@@ -2280,9 +2216,7 @@ static int btrfsic_process_written_superblock(
 			       state->max_superblock_generation);
 	} else {
 		if (state->print_mask & BTRFSIC_PRINT_MASK_SUPERBLOCK_WRITE)
-			printk(KERN_INFO
-			       "btrfsic: got new superblock @%llu (%s/%llu/%d)"
-			       " with new gen %llu > %llu\n",
+			pr_info("btrfsic: got new superblock @%llu (%s/%llu/%d) with new gen %llu > %llu\n",
 			       superblock->logical_bytenr,
 			       superblock->dev_state->name,
 			       superblock->dev_bytenr, superblock->mirror_num,
@@ -2317,7 +2251,7 @@ static int btrfsic_process_written_superblock(
 			next_bytenr = btrfs_super_root(super_hdr);
 			if (state->print_mask &
 			    BTRFSIC_PRINT_MASK_ROOT_CHUNK_LOG_TREE_LOCATION)
-				printk(KERN_INFO "root@%llu\n", next_bytenr);
+				pr_info("root@%llu\n", next_bytenr);
 			break;
 		case 1:
 			btrfs_set_disk_key_objectid(&tmp_disk_key,
@@ -2326,7 +2260,7 @@ static int btrfsic_process_written_superblock(
 			next_bytenr = btrfs_super_chunk_root(super_hdr);
 			if (state->print_mask &
 			    BTRFSIC_PRINT_MASK_ROOT_CHUNK_LOG_TREE_LOCATION)
-				printk(KERN_INFO "chunk@%llu\n", next_bytenr);
+				pr_info("chunk@%llu\n", next_bytenr);
 			break;
 		case 2:
 			btrfs_set_disk_key_objectid(&tmp_disk_key,
@@ -2337,7 +2271,7 @@ static int btrfsic_process_written_superblock(
 				continue;
 			if (state->print_mask &
 			    BTRFSIC_PRINT_MASK_ROOT_CHUNK_LOG_TREE_LOCATION)
-				printk(KERN_INFO "log@%llu\n", next_bytenr);
+				pr_info("log@%llu\n", next_bytenr);
 			break;
 		}
 
@@ -2345,23 +2279,19 @@ static int btrfsic_process_written_superblock(
 		    btrfs_num_copies(state->root->fs_info,
 				     next_bytenr, BTRFS_SUPER_INFO_SIZE);
 		if (state->print_mask & BTRFSIC_PRINT_MASK_NUM_COPIES)
-			printk(KERN_INFO "num_copies(log_bytenr=%llu) = %d\n",
+			pr_info("num_copies(log_bytenr=%llu) = %d\n",
 			       next_bytenr, num_copies);
 		for (mirror_num = 1; mirror_num <= num_copies; mirror_num++) {
 			int was_created;
 
 			if (state->print_mask & BTRFSIC_PRINT_MASK_VERBOSE)
-				printk(KERN_INFO
-				       "btrfsic_process_written_superblock("
-				       "mirror_num=%d)\n", mirror_num);
+				pr_info("btrfsic_process_written_superblock(mirror_num=%d)\n", mirror_num);
 			ret = btrfsic_map_block(state, next_bytenr,
 						BTRFS_SUPER_INFO_SIZE,
 						&tmp_next_block_ctx,
 						mirror_num);
 			if (ret) {
-				printk(KERN_INFO
-				       "btrfsic: btrfsic_map_block(@%llu,"
-				       " mirror=%d) failed!\n",
+				pr_info("btrfsic: btrfsic_map_block(@%llu, mirror=%d) failed!\n",
 				       next_bytenr, mirror_num);
 				return -1;
 			}
@@ -2374,8 +2304,7 @@ static int btrfsic_process_written_superblock(
 					mirror_num,
 					&was_created);
 			if (NULL == next_block) {
-				printk(KERN_INFO
-				       "btrfsic: error, kmalloc failed!\n");
+				pr_info("btrfsic: error, kmalloc failed!\n");
 				btrfsic_release_block_ctx(&tmp_next_block_ctx);
 				return -1;
 			}
@@ -2424,8 +2353,7 @@ static int btrfsic_check_all_ref_blocks(struct btrfsic_state *state,
 		 * by the most recent super block.
 		 */
 		if (state->print_mask & BTRFSIC_PRINT_MASK_VERBOSE)
-			printk(KERN_INFO
-			       "btrfsic: abort cyclic linkage (case 1).\n");
+			pr_info("btrfsic: abort cyclic linkage (case 1).\n");
 
 		return ret;
 	}
@@ -2436,9 +2364,7 @@ static int btrfsic_check_all_ref_blocks(struct btrfsic_state *state,
 	 */
 	list_for_each_entry(l, &block->ref_to_list, node_ref_to) {
 		if (state->print_mask & BTRFSIC_PRINT_MASK_VERBOSE)
-			printk(KERN_INFO
-			       "rl=%d, %c @%llu (%s/%llu/%d)"
-			       " %u* refers to %c @%llu (%s/%llu/%d)\n",
+			pr_info("rl=%d, %c @%llu (%s/%llu/%d) %u* refers to %c @%llu (%s/%llu/%d)\n",
 			       recursion_level,
 			       btrfsic_get_block_type(state, block),
 			       block->logical_bytenr, block->dev_state->name,
@@ -2450,9 +2376,7 @@ static int btrfsic_check_all_ref_blocks(struct btrfsic_state *state,
 			       l->block_ref_to->dev_bytenr,
 			       l->block_ref_to->mirror_num);
 		if (l->block_ref_to->never_written) {
-			printk(KERN_INFO "btrfs: attempt to write superblock"
-			       " which references block %c @%llu (%s/%llu/%d)"
-			       " which is never written!\n",
+			pr_info("btrfs: attempt to write superblock which references block %c @%llu (%s/%llu/%d) which is never written!\n",
 			       btrfsic_get_block_type(state, l->block_ref_to),
 			       l->block_ref_to->logical_bytenr,
 			       l->block_ref_to->dev_state->name,
@@ -2460,9 +2384,7 @@ static int btrfsic_check_all_ref_blocks(struct btrfsic_state *state,
 			       l->block_ref_to->mirror_num);
 			ret = -1;
 		} else if (!l->block_ref_to->is_iodone) {
-			printk(KERN_INFO "btrfs: attempt to write superblock"
-			       " which references block %c @%llu (%s/%llu/%d)"
-			       " which is not yet iodone!\n",
+			pr_info("btrfs: attempt to write superblock which references block %c @%llu (%s/%llu/%d) which is not yet iodone!\n",
 			       btrfsic_get_block_type(state, l->block_ref_to),
 			       l->block_ref_to->logical_bytenr,
 			       l->block_ref_to->dev_state->name,
@@ -2470,9 +2392,7 @@ static int btrfsic_check_all_ref_blocks(struct btrfsic_state *state,
 			       l->block_ref_to->mirror_num);
 			ret = -1;
 		} else if (l->block_ref_to->iodone_w_error) {
-			printk(KERN_INFO "btrfs: attempt to write superblock"
-			       " which references block %c @%llu (%s/%llu/%d)"
-			       " which has write error!\n",
+			pr_info("btrfs: attempt to write superblock which references block %c @%llu (%s/%llu/%d) which has write error!\n",
 			       btrfsic_get_block_type(state, l->block_ref_to),
 			       l->block_ref_to->logical_bytenr,
 			       l->block_ref_to->dev_state->name,
@@ -2485,10 +2405,7 @@ static int btrfsic_check_all_ref_blocks(struct btrfsic_state *state,
 			   l->parent_generation &&
 			   BTRFSIC_GENERATION_UNKNOWN !=
 			   l->block_ref_to->generation) {
-			printk(KERN_INFO "btrfs: attempt to write superblock"
-			       " which references block %c @%llu (%s/%llu/%d)"
-			       " with generation %llu !="
-			       " parent generation %llu!\n",
+			pr_info("btrfs: attempt to write superblock which references block %c @%llu (%s/%llu/%d) with generation %llu != parent generation %llu!\n",
 			       btrfsic_get_block_type(state, l->block_ref_to),
 			       l->block_ref_to->logical_bytenr,
 			       l->block_ref_to->dev_state->name,
@@ -2499,11 +2416,7 @@ static int btrfsic_check_all_ref_blocks(struct btrfsic_state *state,
 			ret = -1;
 		} else if (l->block_ref_to->flush_gen >
 			   l->block_ref_to->dev_state->last_flush_gen) {
-			printk(KERN_INFO "btrfs: attempt to write superblock"
-			       " which references block %c @%llu (%s/%llu/%d)"
-			       " which is not flushed out of disk's write cache"
-			       " (block flush_gen=%llu,"
-			       " dev->flush_gen=%llu)!\n",
+			pr_info("btrfs: attempt to write superblock which references block %c @%llu (%s/%llu/%d) which is not flushed out of disk's write cache (block flush_gen=%llu, dev->flush_gen=%llu)!\n",
 			       btrfsic_get_block_type(state, l->block_ref_to),
 			       l->block_ref_to->logical_bytenr,
 			       l->block_ref_to->dev_state->name,
@@ -2532,8 +2445,7 @@ static int btrfsic_is_block_ref_by_superblock(
 	if (recursion_level >= 3 + BTRFS_MAX_LEVEL) {
 		/* refer to comment at "abort cyclic linkage (case 1)" */
 		if (state->print_mask & BTRFSIC_PRINT_MASK_VERBOSE)
-			printk(KERN_INFO
-			       "btrfsic: abort cyclic linkage (case 2).\n");
+			pr_info("btrfsic: abort cyclic linkage (case 2).\n");
 
 		return 0;
 	}
@@ -2544,9 +2456,7 @@ static int btrfsic_is_block_ref_by_superblock(
 	 */
 	list_for_each_entry(l, &block->ref_from_list, node_ref_from) {
 		if (state->print_mask & BTRFSIC_PRINT_MASK_VERBOSE)
-			printk(KERN_INFO
-			       "rl=%d, %c @%llu (%s/%llu/%d)"
-			       " is ref %u* from %c @%llu (%s/%llu/%d)\n",
+			pr_info("rl=%d, %c @%llu (%s/%llu/%d) is ref %u* from %c @%llu (%s/%llu/%d)\n",
 			       recursion_level,
 			       btrfsic_get_block_type(state, block),
 			       block->logical_bytenr, block->dev_state->name,
@@ -2576,9 +2486,7 @@ static int btrfsic_is_block_ref_by_superblock(
 static void btrfsic_print_add_link(const struct btrfsic_state *state,
 				   const struct btrfsic_block_link *l)
 {
-	printk(KERN_INFO
-	       "Add %u* link from %c @%llu (%s/%llu/%d)"
-	       " to %c @%llu (%s/%llu/%d).\n",
+	pr_info("Add %u* link from %c @%llu (%s/%llu/%d) to %c @%llu (%s/%llu/%d).\n",
 	       l->ref_cnt,
 	       btrfsic_get_block_type(state, l->block_ref_from),
 	       l->block_ref_from->logical_bytenr,
@@ -2593,9 +2501,7 @@ static void btrfsic_print_add_link(const struct btrfsic_state *state,
 static void btrfsic_print_rem_link(const struct btrfsic_state *state,
 				   const struct btrfsic_block_link *l)
 {
-	printk(KERN_INFO
-	       "Rem %u* link from %c @%llu (%s/%llu/%d)"
-	       " to %c @%llu (%s/%llu/%d).\n",
+	pr_info("Rem %u* link from %c @%llu (%s/%llu/%d) to %c @%llu (%s/%llu/%d).\n",
 	       l->ref_cnt,
 	       btrfsic_get_block_type(state, l->block_ref_from),
 	       l->block_ref_from->logical_bytenr,
@@ -2645,7 +2551,7 @@ static void btrfsic_dump_tree_sub(const struct btrfsic_state *state,
 	 * This algorithm is recursive because the amount of used stack space
 	 * is very small and the max recursion depth is limited.
 	 */
-	indent_add = sprintf(buf, "%c-%llu(%s/%llu/%d)",
+	indent_add = sprintf(buf, "%c-%llu(%s/%llu/%u)",
 			     btrfsic_get_block_type(state, block),
 			     block->logical_bytenr, block->dev_state->name,
 			     block->dev_bytenr, block->mirror_num);
@@ -2707,8 +2613,7 @@ static struct btrfsic_block_link *btrfsic_block_link_lookup_or_add(
 	if (NULL == l) {
 		l = btrfsic_block_link_alloc();
 		if (NULL == l) {
-			printk(KERN_INFO
-			       "btrfsic: error, kmalloc" " failed!\n");
+			pr_info("btrfsic: error, kmalloc failed!\n");
 			return NULL;
 		}
 
@@ -2755,13 +2660,12 @@ static struct btrfsic_block *btrfsic_block_lookup_or_add(
 
 		block = btrfsic_block_alloc();
 		if (NULL == block) {
-			printk(KERN_INFO "btrfsic: error, kmalloc failed!\n");
+			pr_info("btrfsic: error, kmalloc failed!\n");
 			return NULL;
 		}
 		dev_state = btrfsic_dev_state_lookup(block_ctx->dev->bdev);
 		if (NULL == dev_state) {
-			printk(KERN_INFO
-			       "btrfsic: error, lookup dev_state failed!\n");
+			pr_info("btrfsic: error, lookup dev_state failed!\n");
 			btrfsic_block_free(block);
 			return NULL;
 		}
@@ -2773,8 +2677,7 @@ static struct btrfsic_block *btrfsic_block_lookup_or_add(
 		block->never_written = never_written;
 		block->mirror_num = mirror_num;
 		if (state->print_mask & BTRFSIC_PRINT_MASK_VERBOSE)
-			printk(KERN_INFO
-			       "New %s%c-block @%llu (%s/%llu/%d)\n",
+			pr_info("New %s%c-block @%llu (%s/%llu/%d)\n",
 			       additional_string,
 			       btrfsic_get_block_type(state, block),
 			       block->logical_bytenr, dev_state->name,
@@ -2809,9 +2712,7 @@ static void btrfsic_cmp_log_and_dev_bytenr(struct btrfsic_state *state,
 		ret = btrfsic_map_block(state, bytenr, state->metablock_size,
 					&block_ctx, mirror_num);
 		if (ret) {
-			printk(KERN_INFO "btrfsic:"
-			       " btrfsic_map_block(logical @%llu,"
-			       " mirror %d) failed!\n",
+			pr_info("btrfsic: btrfsic_map_block(logical @%llu, mirror %d) failed!\n",
 			       bytenr, mirror_num);
 			continue;
 		}
@@ -2826,9 +2727,7 @@ static void btrfsic_cmp_log_and_dev_bytenr(struct btrfsic_state *state,
 	}
 
 	if (WARN_ON(!match)) {
-		printk(KERN_INFO "btrfs: attempt to write M-block which contains logical bytenr that doesn't map to dev+physical bytenr of submit_bio,"
-		       " buffer->log_bytenr=%llu, submit_bio(bdev=%s,"
-		       " phys_bytenr=%llu)!\n",
+		pr_info("btrfs: attempt to write M-block which contains logical bytenr that doesn't map to dev+physical bytenr of submit_bio, buffer->log_bytenr=%llu, submit_bio(bdev=%s, phys_bytenr=%llu)!\n",
 		       bytenr, dev_state->name, dev_bytenr);
 		for (mirror_num = 1; mirror_num <= num_copies; mirror_num++) {
 			ret = btrfsic_map_block(state, bytenr,
@@ -2837,8 +2736,7 @@ static void btrfsic_cmp_log_and_dev_bytenr(struct btrfsic_state *state,
 			if (ret)
 				continue;
 
-			printk(KERN_INFO "Read logical bytenr @%llu maps to"
-			       " (%s/%llu/%d)\n",
+			pr_info("Read logical bytenr @%llu maps to (%s/%llu/%d)\n",
 			       bytenr, block_ctx.dev->name,
 			       block_ctx.dev_bytenr, mirror_num);
 		}
@@ -2848,19 +2746,16 @@ static void btrfsic_cmp_log_and_dev_bytenr(struct btrfsic_state *state,
 static struct btrfsic_dev_state *btrfsic_dev_state_lookup(
 		struct block_device *bdev)
 {
-	struct btrfsic_dev_state *ds;
-
-	ds = btrfsic_dev_state_hashtable_lookup(bdev,
-						&btrfsic_dev_state_hashtable);
-	return ds;
+	return btrfsic_dev_state_hashtable_lookup(bdev,
+						  &btrfsic_dev_state_hashtable);
 }
 
-int btrfsic_submit_bh(int rw, struct buffer_head *bh)
+int btrfsic_submit_bh(int op, int op_flags, struct buffer_head *bh)
 {
 	struct btrfsic_dev_state *dev_state;
 
 	if (!btrfsic_is_initialized)
-		return submit_bh(rw, bh);
+		return submit_bh(op, op_flags, bh);
 
 	mutex_lock(&btrfsic_mutex);
 	/* since btrfsic_submit_bh() might also be called before
@@ -2869,34 +2764,28 @@ int btrfsic_submit_bh(int rw, struct buffer_head *bh)
 
 	/* Only called to write the superblock (incl. FLUSH/FUA) */
 	if (NULL != dev_state &&
-	    (rw & WRITE) && bh->b_size > 0) {
+	    (op == REQ_OP_WRITE) && bh->b_size > 0) {
 		u64 dev_bytenr;
 
 		dev_bytenr = 4096 * bh->b_blocknr;
 		if (dev_state->state->print_mask &
 		    BTRFSIC_PRINT_MASK_SUBMIT_BIO_BH)
-			printk(KERN_INFO
-			       "submit_bh(rw=0x%x, blocknr=%llu (bytenr %llu),"
-			       " size=%zu, data=%p, bdev=%p)\n",
-			       rw, (unsigned long long)bh->b_blocknr,
+			pr_info("submit_bh(op=0x%x,0x%x, blocknr=%llu (bytenr %llu), size=%zu, data=%p, bdev=%p)\n",
+			       op, op_flags, (unsigned long long)bh->b_blocknr,
 			       dev_bytenr, bh->b_size, bh->b_data, bh->b_bdev);
 		btrfsic_process_written_block(dev_state, dev_bytenr,
 					      &bh->b_data, 1, NULL,
-					      NULL, bh, rw);
-	} else if (NULL != dev_state && (rw & REQ_FLUSH)) {
+					      NULL, bh, op_flags);
+	} else if (NULL != dev_state && (op_flags & REQ_PREFLUSH)) {
 		if (dev_state->state->print_mask &
 		    BTRFSIC_PRINT_MASK_SUBMIT_BIO_BH)
-			printk(KERN_INFO
-			       "submit_bh(rw=0x%x FLUSH, bdev=%p)\n",
-			       rw, bh->b_bdev);
+			pr_info("submit_bh(op=0x%x,0x%x FLUSH, bdev=%p)\n",
+			       op, op_flags, bh->b_bdev);
 		if (!dev_state->dummy_block_for_bio_bh_flush.is_iodone) {
 			if ((dev_state->state->print_mask &
 			     (BTRFSIC_PRINT_MASK_SUBMIT_BIO_BH |
 			      BTRFSIC_PRINT_MASK_VERBOSE)))
-				printk(KERN_INFO
-				       "btrfsic_submit_bh(%s) with FLUSH"
-				       " but dummy block already in use"
-				       " (ignored)!\n",
+				pr_info("btrfsic_submit_bh(%s) with FLUSH but dummy block already in use (ignored)!\n",
 				       dev_state->name);
 		} else {
 			struct btrfsic_block *const block =
@@ -2906,7 +2795,7 @@ int btrfsic_submit_bh(int rw, struct buffer_head *bh)
 			block->never_written = 0;
 			block->iodone_w_error = 0;
 			block->flush_gen = dev_state->last_flush_gen + 1;
-			block->submit_bio_bh_rw = rw;
+			block->submit_bio_bh_rw = op_flags;
 			block->orig_bio_bh_private = bh->b_private;
 			block->orig_bio_bh_end_io.bh = bh->b_end_io;
 			block->next_in_same_bio = NULL;
@@ -2915,10 +2804,10 @@ int btrfsic_submit_bh(int rw, struct buffer_head *bh)
 		}
 	}
 	mutex_unlock(&btrfsic_mutex);
-	return submit_bh(rw, bh);
+	return submit_bh(op, op_flags, bh);
 }
 
-static void __btrfsic_submit_bio(int rw, struct bio *bio)
+static void __btrfsic_submit_bio(struct bio *bio)
 {
 	struct btrfsic_dev_state *dev_state;
 
@@ -2930,7 +2819,7 @@ static void __btrfsic_submit_bio(int rw, struct bio *bio)
 	 * btrfsic_mount(), this might return NULL */
 	dev_state = btrfsic_dev_state_lookup(bio->bi_bdev);
 	if (NULL != dev_state &&
-	    (rw & WRITE) && NULL != bio->bi_io_vec) {
+	    (bio_op(bio) == REQ_OP_WRITE) && NULL != bio->bi_io_vec) {
 		unsigned int i;
 		u64 dev_bytenr;
 		u64 cur_bytenr;
@@ -2941,10 +2830,8 @@ static void __btrfsic_submit_bio(int rw, struct bio *bio)
 		bio_is_patched = 0;
 		if (dev_state->state->print_mask &
 		    BTRFSIC_PRINT_MASK_SUBMIT_BIO_BH)
-			printk(KERN_INFO
-			       "submit_bio(rw=0x%x, bi_vcnt=%u,"
-			       " bi_sector=%llu (bytenr %llu), bi_bdev=%p)\n",
-			       rw, bio->bi_vcnt,
+			pr_info("submit_bio(rw=%d,0x%x, bi_vcnt=%u, bi_sector=%llu (bytenr %llu), bi_bdev=%p)\n",
+			       bio_op(bio), bio->bi_opf, bio->bi_vcnt,
 			       (unsigned long long)bio->bi_iter.bi_sector,
 			       dev_bytenr, bio->bi_bdev);
 
@@ -2966,8 +2853,7 @@ static void __btrfsic_submit_bio(int rw, struct bio *bio)
 			}
 			if (dev_state->state->print_mask &
 			    BTRFSIC_PRINT_MASK_SUBMIT_BIO_BH_VERBOSE)
-				printk(KERN_INFO
-				       "#%u: bytenr=%llu, len=%u, offset=%u\n",
+				pr_info("#%u: bytenr=%llu, len=%u, offset=%u\n",
 				       i, cur_bytenr, bio->bi_io_vec[i].bv_len,
 				       bio->bi_io_vec[i].bv_offset);
 			cur_bytenr += bio->bi_io_vec[i].bv_len;
@@ -2975,26 +2861,22 @@ static void __btrfsic_submit_bio(int rw, struct bio *bio)
 		btrfsic_process_written_block(dev_state, dev_bytenr,
 					      mapped_datav, bio->bi_vcnt,
 					      bio, &bio_is_patched,
-					      NULL, rw);
+					      NULL, bio->bi_opf);
 		while (i > 0) {
 			i--;
 			kunmap(bio->bi_io_vec[i].bv_page);
 		}
 		kfree(mapped_datav);
-	} else if (NULL != dev_state && (rw & REQ_FLUSH)) {
+	} else if (NULL != dev_state && (bio->bi_opf & REQ_PREFLUSH)) {
 		if (dev_state->state->print_mask &
 		    BTRFSIC_PRINT_MASK_SUBMIT_BIO_BH)
-			printk(KERN_INFO
-			       "submit_bio(rw=0x%x FLUSH, bdev=%p)\n",
-			       rw, bio->bi_bdev);
+			pr_info("submit_bio(rw=%d,0x%x FLUSH, bdev=%p)\n",
+			       bio_op(bio), bio->bi_opf, bio->bi_bdev);
 		if (!dev_state->dummy_block_for_bio_bh_flush.is_iodone) {
 			if ((dev_state->state->print_mask &
 			     (BTRFSIC_PRINT_MASK_SUBMIT_BIO_BH |
 			      BTRFSIC_PRINT_MASK_VERBOSE)))
-				printk(KERN_INFO
-				       "btrfsic_submit_bio(%s) with FLUSH"
-				       " but dummy block already in use"
-				       " (ignored)!\n",
+				pr_info("btrfsic_submit_bio(%s) with FLUSH but dummy block already in use (ignored)!\n",
 				       dev_state->name);
 		} else {
 			struct btrfsic_block *const block =
@@ -3004,7 +2886,7 @@ static void __btrfsic_submit_bio(int rw, struct bio *bio)
 			block->never_written = 0;
 			block->iodone_w_error = 0;
 			block->flush_gen = dev_state->last_flush_gen + 1;
-			block->submit_bio_bh_rw = rw;
+			block->submit_bio_bh_rw = bio->bi_opf;
 			block->orig_bio_bh_private = bio->bi_private;
 			block->orig_bio_bh_end_io.bio = bio->bi_end_io;
 			block->next_in_same_bio = NULL;
@@ -3016,16 +2898,16 @@ leave:
 	mutex_unlock(&btrfsic_mutex);
 }
 
-void btrfsic_submit_bio(int rw, struct bio *bio)
+void btrfsic_submit_bio(struct bio *bio)
 {
-	__btrfsic_submit_bio(rw, bio);
-	submit_bio(rw, bio);
+	__btrfsic_submit_bio(bio);
+	submit_bio(bio);
 }
 
-int btrfsic_submit_bio_wait(int rw, struct bio *bio)
+int btrfsic_submit_bio_wait(struct bio *bio)
 {
-	__btrfsic_submit_bio(rw, bio);
-	return submit_bio_wait(rw, bio);
+	__btrfsic_submit_bio(bio);
+	return submit_bio_wait(bio);
 }
 
 int btrfsic_mount(struct btrfs_root *root,
@@ -3038,14 +2920,12 @@ int btrfsic_mount(struct btrfs_root *root,
 	struct btrfs_device *device;
 
 	if (root->nodesize & ((u64)PAGE_SIZE - 1)) {
-		printk(KERN_INFO
-		       "btrfsic: cannot handle nodesize %d not being a multiple of PAGE_SIZE %ld!\n",
+		pr_info("btrfsic: cannot handle nodesize %d not being a multiple of PAGE_SIZE %ld!\n",
 		       root->nodesize, PAGE_SIZE);
 		return -1;
 	}
 	if (root->sectorsize & ((u64)PAGE_SIZE - 1)) {
-		printk(KERN_INFO
-		       "btrfsic: cannot handle sectorsize %d not being a multiple of PAGE_SIZE %ld!\n",
+		pr_info("btrfsic: cannot handle sectorsize %d not being a multiple of PAGE_SIZE %ld!\n",
 		       root->sectorsize, PAGE_SIZE);
 		return -1;
 	}
@@ -3053,7 +2933,7 @@ int btrfsic_mount(struct btrfs_root *root,
 	if (!state) {
 		state = vzalloc(sizeof(*state));
 		if (!state) {
-			printk(KERN_INFO "btrfs check-integrity: vzalloc() failed!\n");
+			pr_info("btrfs check-integrity: vzalloc() failed!\n");
 			return -1;
 		}
 	}
@@ -3085,8 +2965,7 @@ int btrfsic_mount(struct btrfs_root *root,
 
 		ds = btrfsic_dev_state_alloc();
 		if (NULL == ds) {
-			printk(KERN_INFO
-			       "btrfs check-integrity: kmalloc() failed!\n");
+			pr_info("btrfs check-integrity: kmalloc() failed!\n");
 			mutex_unlock(&btrfsic_mutex);
 			return -1;
 		}
@@ -3147,9 +3026,7 @@ void btrfsic_unmount(struct btrfs_root *root,
 	}
 
 	if (NULL == state) {
-		printk(KERN_INFO
-		       "btrfsic: error, cannot find state information"
-		       " on umount!\n");
+		pr_info("btrfsic: error, cannot find state information on umount!\n");
 		mutex_unlock(&btrfsic_mutex);
 		return;
 	}
@@ -3176,9 +3053,7 @@ void btrfsic_unmount(struct btrfs_root *root,
 		if (b_all->is_iodone || b_all->never_written)
 			btrfsic_block_free(b_all);
 		else
-			printk(KERN_INFO "btrfs: attempt to free %c-block"
-			       " @%llu (%s/%llu/%d) on umount which is"
-			       " not yet iodone!\n",
+			pr_info("btrfs: attempt to free %c-block @%llu (%s/%llu/%d) on umount which is not yet iodone!\n",
 			       btrfsic_get_block_type(state, b_all),
 			       b_all->logical_bytenr, b_all->dev_state->name,
 			       b_all->dev_bytenr, b_all->mirror_num);
diff --git a/fs/btrfs/check-integrity.h b/fs/btrfs/check-integrity.h
index 13b8566c97ab..f78dff1c7e86 100644
--- a/fs/btrfs/check-integrity.h
+++ b/fs/btrfs/check-integrity.h
@@ -20,9 +20,9 @@
 #define __BTRFS_CHECK_INTEGRITY__
 
 #ifdef CONFIG_BTRFS_FS_CHECK_INTEGRITY
-int btrfsic_submit_bh(int rw, struct buffer_head *bh);
-void btrfsic_submit_bio(int rw, struct bio *bio);
-int btrfsic_submit_bio_wait(int rw, struct bio *bio);
+int btrfsic_submit_bh(int op, int op_flags, struct buffer_head *bh);
+void btrfsic_submit_bio(struct bio *bio);
+int btrfsic_submit_bio_wait(struct bio *bio);
 #else
 #define btrfsic_submit_bh submit_bh
 #define btrfsic_submit_bio submit_bio
diff --git a/fs/btrfs/compression.c b/fs/btrfs/compression.c
index 658c39b70fba..ccc70d96958d 100644
--- a/fs/btrfs/compression.c
+++ b/fs/btrfs/compression.c
@@ -363,6 +363,7 @@ int btrfs_submit_compressed_write(struct inode *inode, u64 start,
 		kfree(cb);
 		return -ENOMEM;
 	}
+	bio_set_op_attrs(bio, REQ_OP_WRITE, 0);
 	bio->bi_private = cb;
 	bio->bi_end_io = end_compressed_bio_write;
 	atomic_inc(&cb->pending_bios);
@@ -373,7 +374,7 @@ int btrfs_submit_compressed_write(struct inode *inode, u64 start,
 		page = compressed_pages[pg_index];
 		page->mapping = inode->i_mapping;
 		if (bio->bi_iter.bi_size)
-			ret = io_tree->ops->merge_bio_hook(WRITE, page, 0,
+			ret = io_tree->ops->merge_bio_hook(page, 0,
 							   PAGE_SIZE,
 							   bio, 0);
 		else
@@ -401,13 +402,17 @@ int btrfs_submit_compressed_write(struct inode *inode, u64 start,
 				BUG_ON(ret); /* -ENOMEM */
 			}
 
-			ret = btrfs_map_bio(root, WRITE, bio, 0, 1);
-			BUG_ON(ret); /* -ENOMEM */
+			ret = btrfs_map_bio(root, bio, 0, 1);
+			if (ret) {
+				bio->bi_error = ret;
+				bio_endio(bio);
+			}
 
 			bio_put(bio);
 
 			bio = compressed_bio_alloc(bdev, first_byte, GFP_NOFS);
 			BUG_ON(!bio);
+			bio_set_op_attrs(bio, REQ_OP_WRITE, 0);
 			bio->bi_private = cb;
 			bio->bi_end_io = end_compressed_bio_write;
 			bio_add_page(bio, page, PAGE_SIZE, 0);
@@ -431,8 +436,11 @@ int btrfs_submit_compressed_write(struct inode *inode, u64 start,
 		BUG_ON(ret); /* -ENOMEM */
 	}
 
-	ret = btrfs_map_bio(root, WRITE, bio, 0, 1);
-	BUG_ON(ret); /* -ENOMEM */
+	ret = btrfs_map_bio(root, bio, 0, 1);
+	if (ret) {
+		bio->bi_error = ret;
+		bio_endio(bio);
+	}
 
 	bio_put(bio);
 	return 0;
@@ -646,6 +654,7 @@ int btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
 	comp_bio = compressed_bio_alloc(bdev, cur_disk_byte, GFP_NOFS);
 	if (!comp_bio)
 		goto fail2;
+	bio_set_op_attrs (comp_bio, REQ_OP_READ, 0);
 	comp_bio->bi_private = cb;
 	comp_bio->bi_end_io = end_compressed_bio_read;
 	atomic_inc(&cb->pending_bios);
@@ -656,7 +665,7 @@ int btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
 		page->index = em_start >> PAGE_SHIFT;
 
 		if (comp_bio->bi_iter.bi_size)
-			ret = tree->ops->merge_bio_hook(READ, page, 0,
+			ret = tree->ops->merge_bio_hook(page, 0,
 							PAGE_SIZE,
 							comp_bio, 0);
 		else
@@ -687,8 +696,7 @@ int btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
 			sums += DIV_ROUND_UP(comp_bio->bi_iter.bi_size,
 					     root->sectorsize);
 
-			ret = btrfs_map_bio(root, READ, comp_bio,
-					    mirror_num, 0);
+			ret = btrfs_map_bio(root, comp_bio, mirror_num, 0);
 			if (ret) {
 				bio->bi_error = ret;
 				bio_endio(comp_bio);
@@ -699,6 +707,7 @@ int btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
 			comp_bio = compressed_bio_alloc(bdev, cur_disk_byte,
 							GFP_NOFS);
 			BUG_ON(!comp_bio);
+			bio_set_op_attrs(comp_bio, REQ_OP_READ, 0);
 			comp_bio->bi_private = cb;
 			comp_bio->bi_end_io = end_compressed_bio_read;
 
@@ -717,7 +726,7 @@ int btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
 		BUG_ON(ret); /* -ENOMEM */
 	}
 
-	ret = btrfs_map_bio(root, READ, comp_bio, mirror_num, 0);
+	ret = btrfs_map_bio(root, comp_bio, mirror_num, 0);
 	if (ret) {
 		bio->bi_error = ret;
 		bio_endio(comp_bio);
@@ -774,8 +783,7 @@ void __init btrfs_init_compress(void)
 		 */
 		workspace = btrfs_compress_op[i]->alloc_workspace();
 		if (IS_ERR(workspace)) {
-			printk(KERN_WARNING
-	"BTRFS: cannot preallocate compression workspace, will try later");
+			pr_warn("BTRFS: cannot preallocate compression workspace, will try later\n");
 		} else {
 			atomic_set(&btrfs_comp_ws[i].total_ws, 1);
 			btrfs_comp_ws[i].free_ws = 1;
@@ -845,8 +853,7 @@ again:
 					/* no burst */ 1);
 
 			if (__ratelimit(&_rs)) {
-				printk(KERN_WARNING
-			    "no compression workspaces, low memory, retrying");
+				pr_warn("BTRFS: no compression workspaces, low memory, retrying\n");
 			}
 		}
 		goto again;
diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c
index 427c36b430a6..f6ba165d3f81 100644
--- a/fs/btrfs/ctree.c
+++ b/fs/btrfs/ctree.c
@@ -45,9 +45,7 @@ static int tree_mod_log_free_eb(struct btrfs_fs_info *fs_info,
 
 struct btrfs_path *btrfs_alloc_path(void)
 {
-	struct btrfs_path *path;
-	path = kmem_cache_zalloc(btrfs_path_cachep, GFP_NOFS);
-	return path;
+	return kmem_cache_zalloc(btrfs_path_cachep, GFP_NOFS);
 }
 
 /*
@@ -1102,7 +1100,7 @@ static noinline int __btrfs_cow_block(struct btrfs_trans_handle *trans,
 	int level, ret;
 	int last_ref = 0;
 	int unlock_orig = 0;
-	u64 parent_start;
+	u64 parent_start = 0;
 
 	if (*cow_ret == buf)
 		unlock_orig = 1;
@@ -1121,13 +1119,8 @@ static noinline int __btrfs_cow_block(struct btrfs_trans_handle *trans,
 	else
 		btrfs_node_key(buf, &disk_key, 0);
 
-	if (root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID) {
-		if (parent)
-			parent_start = parent->start;
-		else
-			parent_start = 0;
-	} else
-		parent_start = 0;
+	if ((root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID) && parent)
+		parent_start = parent->start;
 
 	cow = btrfs_alloc_tree_block(trans, root, parent_start,
 			root->root_key.objectid, &disk_key, level,
@@ -1153,14 +1146,14 @@ static noinline int __btrfs_cow_block(struct btrfs_trans_handle *trans,
 
 	ret = update_ref_for_cow(trans, root, buf, cow, &last_ref);
 	if (ret) {
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 		return ret;
 	}
 
 	if (test_bit(BTRFS_ROOT_REF_COWS, &root->state)) {
 		ret = btrfs_reloc_cow_block(trans, root, buf, cow);
 		if (ret) {
-			btrfs_abort_transaction(trans, root, ret);
+			btrfs_abort_transaction(trans, ret);
 			return ret;
 		}
 	}
@@ -1170,8 +1163,6 @@ static noinline int __btrfs_cow_block(struct btrfs_trans_handle *trans,
 		if (root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID ||
 		    btrfs_header_backref_rev(buf) < BTRFS_MIXED_BACKREF_REV)
 			parent_start = buf->start;
-		else
-			parent_start = 0;
 
 		extent_buffer_get(cow);
 		tree_mod_log_set_root_pointer(root, cow, 1);
@@ -1182,11 +1173,6 @@ static noinline int __btrfs_cow_block(struct btrfs_trans_handle *trans,
 		free_extent_buffer(buf);
 		add_root_to_dirty_list(root);
 	} else {
-		if (root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID)
-			parent_start = parent->start;
-		else
-			parent_start = 0;
-
 		WARN_ON(trans->transid != btrfs_header_generation(parent));
 		tree_mod_log_insert_key(root->fs_info, parent, parent_slot,
 					MOD_LOG_KEY_REPLACE, GFP_NOFS);
@@ -1198,7 +1184,7 @@ static noinline int __btrfs_cow_block(struct btrfs_trans_handle *trans,
 		if (last_ref) {
 			ret = tree_mod_log_free_eb(root->fs_info, buf);
 			if (ret) {
-				btrfs_abort_transaction(trans, root, ret);
+				btrfs_abort_transaction(trans, ret);
 				return ret;
 			}
 		}
@@ -1373,7 +1359,8 @@ tree_mod_log_rewind(struct btrfs_fs_info *fs_info, struct btrfs_path *path,
 
 	if (tm->op == MOD_LOG_KEY_REMOVE_WHILE_FREEING) {
 		BUG_ON(tm->slot != 0);
-		eb_rewin = alloc_dummy_extent_buffer(fs_info, eb->start);
+		eb_rewin = alloc_dummy_extent_buffer(fs_info, eb->start,
+						eb->len);
 		if (!eb_rewin) {
 			btrfs_tree_read_unlock_blocking(eb);
 			free_extent_buffer(eb);
@@ -1454,7 +1441,8 @@ get_old_root(struct btrfs_root *root, u64 time_seq)
 	} else if (old_root) {
 		btrfs_tree_read_unlock(eb_root);
 		free_extent_buffer(eb_root);
-		eb = alloc_dummy_extent_buffer(root->fs_info, logical);
+		eb = alloc_dummy_extent_buffer(root->fs_info, logical,
+					root->nodesize);
 	} else {
 		btrfs_set_lock_blocking_rw(eb_root, BTRFS_READ_LOCK);
 		eb = btrfs_clone_extent_buffer(eb_root);
@@ -1503,7 +1491,7 @@ static inline int should_cow_block(struct btrfs_trans_handle *trans,
 				   struct btrfs_root *root,
 				   struct extent_buffer *buf)
 {
-	if (btrfs_test_is_dummy_root(root))
+	if (btrfs_is_testing(root->fs_info))
 		return 0;
 
 	/* ensure we can see the force_cow */
@@ -1552,6 +1540,7 @@ noinline int btrfs_cow_block(struct btrfs_trans_handle *trans,
 		       trans->transid, root->fs_info->generation);
 
 	if (!should_cow_block(trans, root, buf)) {
+		trans->dirty = true;
 		*cow_ret = buf;
 		return 0;
 	}
@@ -1726,20 +1715,6 @@ int btrfs_realloc_node(struct btrfs_trans_handle *trans,
 	return err;
 }
 
-/*
- * The leaf data grows from end-to-front in the node.
- * this returns the address of the start of the last item,
- * which is the stop of the leaf data stack
- */
-static inline unsigned int leaf_data_end(struct btrfs_root *root,
-					 struct extent_buffer *leaf)
-{
-	u32 nr = btrfs_header_nritems(leaf);
-	if (nr == 0)
-		return BTRFS_LEAF_DATA_SIZE(root);
-	return btrfs_item_offset_nr(leaf, nr - 1);
-}
-
 
 /*
  * search for key in the extent_buffer.  The items start at offset p,
@@ -1768,6 +1743,14 @@ static noinline int generic_bin_search(struct extent_buffer *eb,
 	unsigned long map_len = 0;
 	int err;
 
+	if (low > high) {
+		btrfs_err(eb->fs_info,
+		 "%s: low (%d) > high (%d) eb %llu owner %llu level %d",
+			  __func__, low, high, eb->start,
+			  btrfs_header_owner(eb), btrfs_header_level(eb));
+		return -EINVAL;
+	}
+
 	while (low < high) {
 		mid = (low + high) / 2;
 		offset = p + mid * item_size;
@@ -1783,10 +1766,12 @@ static noinline int generic_bin_search(struct extent_buffer *eb,
 			if (!err) {
 				tmp = (struct btrfs_disk_key *)(kaddr + offset -
 							map_start);
-			} else {
+			} else if (err == 1) {
 				read_extent_buffer(eb, &unaligned,
 						   offset, sizeof(unaligned));
 				tmp = &unaligned;
+			} else {
+				return err;
 			}
 
 		} else {
@@ -1853,7 +1838,6 @@ static void root_sub_used(struct btrfs_root *root, u32 size)
 
 /* given a node and slot number, this reads the blocks it points to.  The
  * extent buffer is returned with a reference taken (but unlocked).
- * NULL is returned on error.
  */
 static noinline struct extent_buffer *read_node_slot(struct btrfs_root *root,
 				   struct extent_buffer *parent, int slot)
@@ -1861,19 +1845,16 @@ static noinline struct extent_buffer *read_node_slot(struct btrfs_root *root,
 	int level = btrfs_header_level(parent);
 	struct extent_buffer *eb;
 
-	if (slot < 0)
-		return NULL;
-	if (slot >= btrfs_header_nritems(parent))
-		return NULL;
+	if (slot < 0 || slot >= btrfs_header_nritems(parent))
+		return ERR_PTR(-ENOENT);
 
 	BUG_ON(level == 0);
 
 	eb = read_tree_block(root, btrfs_node_blockptr(parent, slot),
 			     btrfs_node_ptr_generation(parent, slot));
-	if (IS_ERR(eb) || !extent_buffer_uptodate(eb)) {
-		if (!IS_ERR(eb))
-			free_extent_buffer(eb);
-		eb = NULL;
+	if (!IS_ERR(eb) && !extent_buffer_uptodate(eb)) {
+		free_extent_buffer(eb);
+		eb = ERR_PTR(-EIO);
 	}
 
 	return eb;
@@ -1926,8 +1907,8 @@ static noinline int balance_level(struct btrfs_trans_handle *trans,
 
 		/* promote the child to a root */
 		child = read_node_slot(root, mid, 0);
-		if (!child) {
-			ret = -EROFS;
+		if (IS_ERR(child)) {
+			ret = PTR_ERR(child);
 			btrfs_handle_fs_error(root->fs_info, ret, NULL);
 			goto enospc;
 		}
@@ -1965,6 +1946,9 @@ static noinline int balance_level(struct btrfs_trans_handle *trans,
 		return 0;
 
 	left = read_node_slot(root, parent, pslot - 1);
+	if (IS_ERR(left))
+		left = NULL;
+
 	if (left) {
 		btrfs_tree_lock(left);
 		btrfs_set_lock_blocking(left);
@@ -1975,7 +1959,11 @@ static noinline int balance_level(struct btrfs_trans_handle *trans,
 			goto enospc;
 		}
 	}
+
 	right = read_node_slot(root, parent, pslot + 1);
+	if (IS_ERR(right))
+		right = NULL;
+
 	if (right) {
 		btrfs_tree_lock(right);
 		btrfs_set_lock_blocking(right);
@@ -2130,6 +2118,8 @@ static noinline int push_nodes_for_insert(struct btrfs_trans_handle *trans,
 		return 1;
 
 	left = read_node_slot(root, parent, pslot - 1);
+	if (IS_ERR(left))
+		left = NULL;
 
 	/* first, try to make some room in the middle buffer */
 	if (left) {
@@ -2180,6 +2170,8 @@ static noinline int push_nodes_for_insert(struct btrfs_trans_handle *trans,
 		free_extent_buffer(left);
 	}
 	right = read_node_slot(root, parent, pslot + 1);
+	if (IS_ERR(right))
+		right = NULL;
 
 	/*
 	 * then try to empty the right most buffer into the middle
@@ -2248,7 +2240,6 @@ static void reada_for_search(struct btrfs_root *root,
 	u64 search;
 	u64 target;
 	u64 nread = 0;
-	u64 gen;
 	struct extent_buffer *eb;
 	u32 nr;
 	u32 blocksize;
@@ -2293,7 +2284,6 @@ static void reada_for_search(struct btrfs_root *root,
 		search = btrfs_node_blockptr(node, nr);
 		if ((search <= target && target - search <= 65536) ||
 		    (search > target && search - target <= 65536)) {
-			gen = btrfs_node_ptr_generation(node, nr);
 			readahead_tree_block(root, search);
 			nread += blocksize;
 		}
@@ -2510,6 +2500,8 @@ read_block_for_search(struct btrfs_trans_handle *trans,
 		if (!btrfs_buffer_uptodate(tmp, 0, 0))
 			ret = -EIO;
 		free_extent_buffer(tmp);
+	} else {
+		ret = PTR_ERR(tmp);
 	}
 	return ret;
 }
@@ -2773,8 +2765,10 @@ again:
 			 * then we don't want to set the path blocking,
 			 * so we test it here
 			 */
-			if (!should_cow_block(trans, root, b))
+			if (!should_cow_block(trans, root, b)) {
+				trans->dirty = true;
 				goto cow_done;
+			}
 
 			/*
 			 * must have write locks on this node and the
@@ -2823,6 +2817,8 @@ cow_done:
 		}
 
 		ret = key_search(b, key, level, &prev_cmp, &slot);
+		if (ret < 0)
+			goto done;
 
 		if (level != 0) {
 			int dec = 0;
@@ -3229,7 +3225,7 @@ static int push_node_left(struct btrfs_trans_handle *trans,
 	ret = tree_mod_log_eb_copy(root->fs_info, dst, src, dst_nritems, 0,
 				   push_items);
 	if (ret) {
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 		return ret;
 	}
 	copy_extent_buffer(dst, src,
@@ -3304,7 +3300,7 @@ static int balance_node_right(struct btrfs_trans_handle *trans,
 	ret = tree_mod_log_eb_copy(root->fs_info, dst, src, 0,
 				   src_nritems - push_items, push_items);
 	if (ret) {
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 		return ret;
 	}
 	copy_extent_buffer(dst, src,
@@ -3508,7 +3504,7 @@ static noinline int split_node(struct btrfs_trans_handle *trans,
 	ret = tree_mod_log_eb_copy(root->fs_info, split, c, 0,
 				   mid, c_nritems - mid);
 	if (ret) {
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 		return ret;
 	}
 	copy_extent_buffer(split, c,
@@ -3762,7 +3758,11 @@ static int push_leaf_right(struct btrfs_trans_handle *trans, struct btrfs_root
 	btrfs_assert_tree_locked(path->nodes[1]);
 
 	right = read_node_slot(root, upper, slot + 1);
-	if (right == NULL)
+	/*
+	 * slot + 1 is not valid or we fail to read the right node,
+	 * no big deal, just return.
+	 */
+	if (IS_ERR(right))
 		return 1;
 
 	btrfs_tree_lock(right);
@@ -3992,7 +3992,11 @@ static int push_leaf_left(struct btrfs_trans_handle *trans, struct btrfs_root
 	btrfs_assert_tree_locked(path->nodes[1]);
 
 	left = read_node_slot(root, path->nodes[1], slot - 1);
-	if (left == NULL)
+	/*
+	 * slot - 1 is not valid or we fail to read the left node,
+	 * no big deal, just return.
+	 */
+	if (IS_ERR(left))
 		return 1;
 
 	btrfs_tree_lock(left);
@@ -4307,7 +4311,11 @@ again:
 			if (path->slots[1] == 0)
 				fixup_low_keys(fs_info, path, &disk_key, 1);
 		}
-		btrfs_mark_buffer_dirty(right);
+		/*
+		 * We create a new leaf 'right' for the required ins_len and
+		 * we'll do btrfs_mark_buffer_dirty() on this leaf after copying
+		 * the content of ins_len to 'right'.
+		 */
 		return ret;
 	}
 
@@ -4738,8 +4746,9 @@ void setup_items_for_insert(struct btrfs_root *root, struct btrfs_path *path,
 
 	if (btrfs_leaf_free_space(root, leaf) < total_size) {
 		btrfs_print_leaf(root, leaf);
-		btrfs_crit(root->fs_info, "not enough freespace need %u have %d",
-		       total_size, btrfs_leaf_free_space(root, leaf));
+		btrfs_crit(root->fs_info,
+			   "not enough freespace need %u have %d",
+			   total_size, btrfs_leaf_free_space(root, leaf));
 		BUG();
 	}
 
@@ -4748,8 +4757,9 @@ void setup_items_for_insert(struct btrfs_root *root, struct btrfs_path *path,
 
 		if (old_data < data_end) {
 			btrfs_print_leaf(root, leaf);
-			btrfs_crit(root->fs_info, "slot %d old_data %d data_end %d",
-			       slot, old_data, data_end);
+			btrfs_crit(root->fs_info,
+				   "slot %d old_data %d data_end %d",
+				   slot, old_data, data_end);
 			BUG_ON(1);
 		}
 		/*
@@ -4759,7 +4769,7 @@ void setup_items_for_insert(struct btrfs_root *root, struct btrfs_path *path,
 		for (i = slot; i < nritems; i++) {
 			u32 ioff;
 
-			item = btrfs_item_nr( i);
+			item = btrfs_item_nr(i);
 			ioff = btrfs_token_item_offset(leaf, item, &token);
 			btrfs_set_token_item_offset(leaf, item,
 						    ioff - total_data, &token);
@@ -5199,7 +5209,10 @@ find_next_key:
 		}
 		btrfs_set_path_blocking(path);
 		cur = read_node_slot(root, cur, slot);
-		BUG_ON(!cur); /* -ENOMEM */
+		if (IS_ERR(cur)) {
+			ret = PTR_ERR(cur);
+			goto out;
+		}
 
 		btrfs_tree_read_lock(cur);
 
@@ -5218,15 +5231,21 @@ out:
 	return ret;
 }
 
-static void tree_move_down(struct btrfs_root *root,
+static int tree_move_down(struct btrfs_root *root,
 			   struct btrfs_path *path,
 			   int *level, int root_level)
 {
+	struct extent_buffer *eb;
+
 	BUG_ON(*level == 0);
-	path->nodes[*level - 1] = read_node_slot(root, path->nodes[*level],
-					path->slots[*level]);
+	eb = read_node_slot(root, path->nodes[*level], path->slots[*level]);
+	if (IS_ERR(eb))
+		return PTR_ERR(eb);
+
+	path->nodes[*level - 1] = eb;
 	path->slots[*level - 1] = 0;
 	(*level)--;
+	return 0;
 }
 
 static int tree_move_next_or_upnext(struct btrfs_root *root,
@@ -5271,8 +5290,7 @@ static int tree_advance(struct btrfs_root *root,
 	if (*level == 0 || !allow_down) {
 		ret = tree_move_next_or_upnext(root, path, level, root_level);
 	} else {
-		tree_move_down(root, path, level, root_level);
-		ret = 0;
+		ret = tree_move_down(root, path, level, root_level);
 	}
 	if (ret >= 0) {
 		if (*level == 0)
@@ -5446,8 +5464,10 @@ int btrfs_compare_trees(struct btrfs_root *left_root,
 					left_root_level,
 					advance_left != ADVANCE_ONLY_NEXT,
 					&left_key);
-			if (ret < 0)
+			if (ret == -1)
 				left_end_reached = ADVANCE;
+			else if (ret < 0)
+				goto out;
 			advance_left = 0;
 		}
 		if (advance_right && !right_end_reached) {
@@ -5455,8 +5475,10 @@ int btrfs_compare_trees(struct btrfs_root *left_root,
 					right_root_level,
 					advance_right != ADVANCE_ONLY_NEXT,
 					&right_key);
-			if (ret < 0)
+			if (ret == -1)
 				right_end_reached = ADVANCE;
+			else if (ret < 0)
+				goto out;
 			advance_right = 0;
 		}
 
diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index 101c3cfd3f7c..0b8ce2b9f7d0 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -37,6 +37,7 @@
 #include <linux/workqueue.h>
 #include <linux/security.h>
 #include <linux/sizes.h>
+#include <linux/dynamic_debug.h>
 #include "extent_io.h"
 #include "extent_map.h"
 #include "async-thread.h"
@@ -117,6 +118,7 @@ static inline unsigned long btrfs_chunk_item_size(int num_stripes)
 #define BTRFS_FS_STATE_REMOUNTING	1
 #define BTRFS_FS_STATE_TRANS_ABORTED	2
 #define BTRFS_FS_STATE_DEV_REPLACING	3
+#define BTRFS_FS_STATE_DUMMY_FS_INFO	4
 
 #define BTRFS_BACKREF_REV_MAX		256
 #define BTRFS_BACKREF_REV_SHIFT		56
@@ -144,21 +146,6 @@ struct btrfs_header {
 	u8 level;
 } __attribute__ ((__packed__));
 
-#define BTRFS_NODEPTRS_PER_BLOCK(r) (((r)->nodesize - \
-				      sizeof(struct btrfs_header)) / \
-				     sizeof(struct btrfs_key_ptr))
-#define __BTRFS_LEAF_DATA_SIZE(bs) ((bs) - sizeof(struct btrfs_header))
-#define BTRFS_LEAF_DATA_SIZE(r) (__BTRFS_LEAF_DATA_SIZE(r->nodesize))
-#define BTRFS_FILE_EXTENT_INLINE_DATA_START		\
-		(offsetof(struct btrfs_file_extent_item, disk_bytenr))
-#define BTRFS_MAX_INLINE_DATA_SIZE(r) (BTRFS_LEAF_DATA_SIZE(r) - \
-					sizeof(struct btrfs_item) - \
-					BTRFS_FILE_EXTENT_INLINE_DATA_START)
-#define BTRFS_MAX_XATTR_SIZE(r)	(BTRFS_LEAF_DATA_SIZE(r) - \
-				 sizeof(struct btrfs_item) -\
-				 sizeof(struct btrfs_dir_item))
-
-
 /*
  * this is a very generous portion of the super block, giving us
  * room to translate 14 chunks with 3 stripes each.
@@ -265,7 +252,8 @@ struct btrfs_super_block {
 #define BTRFS_FEATURE_COMPAT_SAFE_CLEAR		0ULL
 
 #define BTRFS_FEATURE_COMPAT_RO_SUPP			\
-	(BTRFS_FEATURE_COMPAT_RO_FREE_SPACE_TREE)
+	(BTRFS_FEATURE_COMPAT_RO_FREE_SPACE_TREE |	\
+	 BTRFS_FEATURE_COMPAT_RO_FREE_SPACE_TREE_VALID)
 
 #define BTRFS_FEATURE_COMPAT_RO_SAFE_SET	0ULL
 #define BTRFS_FEATURE_COMPAT_RO_SAFE_CLEAR	0ULL
@@ -439,6 +427,9 @@ struct btrfs_space_info {
 	struct list_head list;
 	/* Protected by the spinlock 'lock'. */
 	struct list_head ro_bgs;
+	struct list_head priority_tickets;
+	struct list_head tickets;
+	u64 tickets_id;
 
 	struct rw_semaphore groups_sem;
 	/* for block groups in our same type */
@@ -687,9 +678,25 @@ struct btrfs_device;
 struct btrfs_fs_devices;
 struct btrfs_balance_control;
 struct btrfs_delayed_root;
+
+#define BTRFS_FS_BARRIER			1
+#define BTRFS_FS_CLOSING_START			2
+#define BTRFS_FS_CLOSING_DONE			3
+#define BTRFS_FS_LOG_RECOVERING			4
+#define BTRFS_FS_OPEN				5
+#define BTRFS_FS_QUOTA_ENABLED			6
+#define BTRFS_FS_QUOTA_ENABLING			7
+#define BTRFS_FS_QUOTA_DISABLING		8
+#define BTRFS_FS_UPDATE_UUID_TREE_GEN		9
+#define BTRFS_FS_CREATING_FREE_SPACE_TREE	10
+#define BTRFS_FS_BTREE_ERR			11
+#define BTRFS_FS_LOG1_ERR			12
+#define BTRFS_FS_LOG2_ERR			13
+
 struct btrfs_fs_info {
 	u8 fsid[BTRFS_FSID_SIZE];
 	u8 chunk_tree_uuid[BTRFS_UUID_SIZE];
+	unsigned long flags;
 	struct btrfs_root *extent_root;
 	struct btrfs_root *tree_root;
 	struct btrfs_root *chunk_root;
@@ -918,10 +925,6 @@ struct btrfs_fs_info {
 	int thread_pool_size;
 
 	struct kobject *space_info_kobj;
-	int do_barriers;
-	int closing;
-	int log_root_recovering;
-	int open;
 
 	u64 total_pinned;
 
@@ -998,17 +1001,6 @@ struct btrfs_fs_info {
 #ifdef CONFIG_BTRFS_FS_CHECK_INTEGRITY
 	u32 check_integrity_print_mask;
 #endif
-	/*
-	 * quota information
-	 */
-	unsigned int quota_enabled:1;
-
-	/*
-	 * quota_enabled only changes state after a commit. This holds the
-	 * next state.
-	 */
-	unsigned int pending_quota_state:1;
-
 	/* is qgroup tracking in a consistent state? */
 	u64 qgroup_flags;
 
@@ -1040,6 +1032,7 @@ struct btrfs_fs_info {
 	struct btrfs_workqueue *qgroup_rescan_workers;
 	struct completion qgroup_rescan_completion;
 	struct btrfs_work qgroup_rescan_work;
+	bool qgroup_rescan_running;	/* protected by qgroup_rescan_lock */
 
 	/* filesystem state */
 	unsigned long fs_state;
@@ -1071,7 +1064,6 @@ struct btrfs_fs_info {
 	wait_queue_head_t replace_wait;
 
 	struct semaphore uuid_tree_rescan_sem;
-	unsigned int update_uuid_tree_gen:1;
 
 	/* Used to reclaim the metadata space in the background. */
 	struct work_struct async_reclaim_work;
@@ -1090,7 +1082,8 @@ struct btrfs_fs_info {
 	 */
 	struct list_head pinned_chunks;
 
-	int creating_free_space_tree;
+	/* Used to record internally whether fs has been frozen */
+	int fs_frozen;
 };
 
 struct btrfs_subvolume_writers {
@@ -1112,12 +1105,11 @@ struct btrfs_subvolume_writers {
 #define BTRFS_ROOT_REF_COWS		1
 #define BTRFS_ROOT_TRACK_DIRTY		2
 #define BTRFS_ROOT_IN_RADIX		3
-#define BTRFS_ROOT_DUMMY_ROOT		4
-#define BTRFS_ROOT_ORPHAN_ITEM_INSERTED	5
-#define BTRFS_ROOT_DEFRAG_RUNNING	6
-#define BTRFS_ROOT_FORCE_COW		7
-#define BTRFS_ROOT_MULTI_LOG_TASKS	8
-#define BTRFS_ROOT_DIRTY		9
+#define BTRFS_ROOT_ORPHAN_ITEM_INSERTED	4
+#define BTRFS_ROOT_DEFRAG_RUNNING	5
+#define BTRFS_ROOT_FORCE_COW		6
+#define BTRFS_ROOT_MULTI_LOG_TASKS	7
+#define BTRFS_ROOT_DIRTY		8
 
 /*
  * in ram representation of the tree.  extent_root is used for all allocations
@@ -1179,8 +1171,10 @@ struct btrfs_root {
 
 	u64 highest_objectid;
 
+#ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
 	/* only used with CONFIG_BTRFS_FS_RUN_SANITY_TESTS is enabled */
 	u64 alloc_bytenr;
+#endif
 
 	u64 defrag_trans_start;
 	struct btrfs_key defrag_progress;
@@ -1257,6 +1251,39 @@ struct btrfs_root {
 	atomic_t qgroup_meta_rsv;
 };
 
+static inline u32 __BTRFS_LEAF_DATA_SIZE(u32 blocksize)
+{
+	return blocksize - sizeof(struct btrfs_header);
+}
+
+static inline u32 BTRFS_LEAF_DATA_SIZE(const struct btrfs_root *root)
+{
+	return __BTRFS_LEAF_DATA_SIZE(root->nodesize);
+}
+
+static inline u32 BTRFS_MAX_ITEM_SIZE(const struct btrfs_root *root)
+{
+	return BTRFS_LEAF_DATA_SIZE(root) - sizeof(struct btrfs_item);
+}
+
+static inline u32 BTRFS_NODEPTRS_PER_BLOCK(const struct btrfs_root *root)
+{
+	return BTRFS_LEAF_DATA_SIZE(root) / sizeof(struct btrfs_key_ptr);
+}
+
+#define BTRFS_FILE_EXTENT_INLINE_DATA_START		\
+		(offsetof(struct btrfs_file_extent_item, disk_bytenr))
+static inline u32 BTRFS_MAX_INLINE_DATA_SIZE(const struct btrfs_root *root)
+{
+	return BTRFS_MAX_ITEM_SIZE(root) -
+	       BTRFS_FILE_EXTENT_INLINE_DATA_START;
+}
+
+static inline u32 BTRFS_MAX_XATTR_SIZE(const struct btrfs_root *root)
+{
+	return BTRFS_MAX_ITEM_SIZE(root) - sizeof(struct btrfs_dir_item);
+}
+
 /*
  * Flags for mount options.
  *
@@ -1297,21 +1324,21 @@ struct btrfs_root {
 #define btrfs_clear_opt(o, opt)		((o) &= ~BTRFS_MOUNT_##opt)
 #define btrfs_set_opt(o, opt)		((o) |= BTRFS_MOUNT_##opt)
 #define btrfs_raw_test_opt(o, opt)	((o) & BTRFS_MOUNT_##opt)
-#define btrfs_test_opt(root, opt)	((root)->fs_info->mount_opt & \
+#define btrfs_test_opt(fs_info, opt)	((fs_info)->mount_opt & \
 					 BTRFS_MOUNT_##opt)
 
-#define btrfs_set_and_info(root, opt, fmt, args...)			\
+#define btrfs_set_and_info(fs_info, opt, fmt, args...)			\
 {									\
-	if (!btrfs_test_opt(root, opt))					\
-		btrfs_info(root->fs_info, fmt, ##args);			\
-	btrfs_set_opt(root->fs_info->mount_opt, opt);			\
+	if (!btrfs_test_opt(fs_info, opt))				\
+		btrfs_info(fs_info, fmt, ##args);			\
+	btrfs_set_opt(fs_info->mount_opt, opt);				\
 }
 
-#define btrfs_clear_and_info(root, opt, fmt, args...)			\
+#define btrfs_clear_and_info(fs_info, opt, fmt, args...)		\
 {									\
-	if (btrfs_test_opt(root, opt))					\
-		btrfs_info(root->fs_info, fmt, ##args);			\
-	btrfs_clear_opt(root->fs_info->mount_opt, opt);			\
+	if (btrfs_test_opt(fs_info, opt))				\
+		btrfs_info(fs_info, fmt, ##args);			\
+	btrfs_clear_opt(fs_info->mount_opt, opt);			\
 }
 
 #ifdef CONFIG_BTRFS_DEBUG
@@ -1319,9 +1346,9 @@ static inline int
 btrfs_should_fragment_free_space(struct btrfs_root *root,
 				 struct btrfs_block_group_cache *block_group)
 {
-	return (btrfs_test_opt(root, FRAGMENT_METADATA) &&
+	return (btrfs_test_opt(root->fs_info, FRAGMENT_METADATA) &&
 		block_group->flags & BTRFS_BLOCK_GROUP_METADATA) ||
-	       (btrfs_test_opt(root, FRAGMENT_DATA) &&
+	       (btrfs_test_opt(root->fs_info, FRAGMENT_DATA) &&
 		block_group->flags &  BTRFS_BLOCK_GROUP_DATA);
 }
 #endif
@@ -1409,13 +1436,13 @@ static inline void btrfs_init_map_token (struct btrfs_map_token *token)
 #define cpu_to_le8(v) (v)
 #define __le8 u8
 
-#define read_eb_member(eb, ptr, type, member, result) (			\
+#define read_eb_member(eb, ptr, type, member, result) (\
 	read_extent_buffer(eb, (char *)(result),			\
 			   ((unsigned long)(ptr)) +			\
 			    offsetof(type, member),			\
 			   sizeof(((type *)0)->member)))
 
-#define write_eb_member(eb, ptr, type, member, result) (		\
+#define write_eb_member(eb, ptr, type, member, result) (\
 	write_extent_buffer(eb, (char *)(result),			\
 			   ((unsigned long)(ptr)) +			\
 			    offsetof(type, member),			\
@@ -2267,6 +2294,21 @@ static inline unsigned long btrfs_leaf_data(struct extent_buffer *l)
 	return offsetof(struct btrfs_leaf, items);
 }
 
+/*
+ * The leaf data grows from end-to-front in the node.
+ * this returns the address of the start of the last item,
+ * which is the stop of the leaf data stack
+ */
+static inline unsigned int leaf_data_end(struct btrfs_root *root,
+					 struct extent_buffer *leaf)
+{
+	u32 nr = btrfs_header_nritems(leaf);
+
+	if (nr == 0)
+		return BTRFS_LEAF_DATA_SIZE(root);
+	return btrfs_item_offset_nr(leaf, nr - 1);
+}
+
 /* struct btrfs_file_extent_item */
 BTRFS_SETGET_FUNCS(file_extent_type, struct btrfs_file_extent_item, type, 8);
 BTRFS_SETGET_STACK_FUNCS(stack_file_extent_disk_bytenr,
@@ -2518,7 +2560,7 @@ 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);
 int btrfs_async_run_delayed_refs(struct btrfs_root *root,
-				 unsigned long count, int wait);
+				 unsigned long count, u64 transid, int wait);
 int btrfs_lookup_data_extent(struct btrfs_root *root, u64 start, u64 len);
 int btrfs_lookup_extent_info(struct btrfs_trans_handle *trans,
 			     struct btrfs_root *root, u64 bytenr,
@@ -2556,7 +2598,7 @@ int btrfs_alloc_logged_file_extent(struct btrfs_trans_handle *trans,
 				   struct btrfs_root *root,
 				   u64 root_objectid, u64 owner, u64 offset,
 				   struct btrfs_key *ins);
-int btrfs_reserve_extent(struct btrfs_root *root, u64 num_bytes,
+int btrfs_reserve_extent(struct btrfs_root *root, u64 ram_bytes, u64 num_bytes,
 			 u64 min_alloc_size, u64 empty_size, u64 hint_byte,
 			 struct btrfs_key *ins, int is_data, int delalloc);
 int btrfs_inc_ref(struct btrfs_trans_handle *trans, struct btrfs_root *root,
@@ -2624,6 +2666,15 @@ enum btrfs_reserve_flush_enum {
 	BTRFS_RESERVE_FLUSH_ALL,
 };
 
+enum btrfs_flush_state {
+	FLUSH_DELAYED_ITEMS_NR	=	1,
+	FLUSH_DELAYED_ITEMS	=	2,
+	FLUSH_DELALLOC		=	3,
+	FLUSH_DELALLOC_WAIT	=	4,
+	ALLOC_CHUNK		=	5,
+	COMMIT_TRANS		=	6,
+};
+
 int btrfs_check_data_free_space(struct inode *inode, u64 start, u64 len);
 int btrfs_alloc_data_chunk_ondemand(struct inode *inode, u64 bytes);
 void btrfs_free_reserved_data_space(struct inode *inode, u64 start, u64 len);
@@ -2661,8 +2712,8 @@ int btrfs_block_rsv_refill(struct btrfs_root *root,
 			   struct btrfs_block_rsv *block_rsv, u64 min_reserved,
 			   enum btrfs_reserve_flush_enum flush);
 int btrfs_block_rsv_migrate(struct btrfs_block_rsv *src_rsv,
-			    struct btrfs_block_rsv *dst_rsv,
-			    u64 num_bytes);
+			    struct btrfs_block_rsv *dst_rsv, u64 num_bytes,
+			    int update_size);
 int btrfs_cond_migrate_bytes(struct btrfs_fs_info *fs_info,
 			     struct btrfs_block_rsv *dest, u64 num_bytes,
 			     int min_factor);
@@ -2832,10 +2883,14 @@ int btrfs_drop_subtree(struct btrfs_trans_handle *trans,
 static inline int btrfs_fs_closing(struct btrfs_fs_info *fs_info)
 {
 	/*
-	 * Get synced with close_ctree()
+	 * Do it this way so we only ever do one test_bit in the normal case.
 	 */
-	smp_mb();
-	return fs_info->closing;
+	if (test_bit(BTRFS_FS_CLOSING_START, &fs_info->flags)) {
+		if (test_bit(BTRFS_FS_CLOSING_DONE, &fs_info->flags))
+			return 2;
+		return 1;
+	}
+	return 0;
 }
 
 /*
@@ -2875,9 +2930,6 @@ void btrfs_put_tree_mod_seq(struct btrfs_fs_info *fs_info,
 int btrfs_old_root_level(struct btrfs_root *root, u64 time_seq);
 
 /* root-item.c */
-int btrfs_find_root_ref(struct btrfs_root *tree_root,
-			struct btrfs_path *path,
-			u64 root_id, u64 ref_id);
 int btrfs_add_root_ref(struct btrfs_trans_handle *trans,
 		       struct btrfs_root *tree_root,
 		       u64 root_id, u64 ref_id, u64 dirid, u64 sequence,
@@ -3086,12 +3138,12 @@ int btrfs_start_delalloc_inodes(struct btrfs_root *root, int delay_iput);
 int btrfs_start_delalloc_roots(struct btrfs_fs_info *fs_info, int delay_iput,
 			       int nr);
 int btrfs_set_extent_delalloc(struct inode *inode, u64 start, u64 end,
-			      struct extent_state **cached_state);
+			      struct extent_state **cached_state, int dedupe);
 int btrfs_create_subvol_root(struct btrfs_trans_handle *trans,
 			     struct btrfs_root *new_root,
 			     struct btrfs_root *parent_root,
 			     u64 new_dirid);
-int btrfs_merge_bio_hook(int rw, struct page *page, unsigned long offset,
+int btrfs_merge_bio_hook(struct page *page, unsigned long offset,
 			 size_t size, struct bio *bio,
 			 unsigned long bio_flags);
 int btrfs_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf);
@@ -3129,7 +3181,6 @@ int btrfs_prealloc_file_range_trans(struct inode *inode,
 				    struct btrfs_trans_handle *trans, int mode,
 				    u64 start, u64 num_bytes, u64 min_size,
 				    loff_t actual_len, u64 *alloc_hint);
-int btrfs_inode_check_errors(struct inode *inode);
 extern const struct dentry_operations btrfs_dentry_operations;
 #ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
 void btrfs_test_inode_set_ops(struct inode *inode);
@@ -3205,14 +3256,17 @@ int btrfs_parse_options(struct btrfs_root *root, char *options,
 			unsigned long new_flags);
 int btrfs_sync_fs(struct super_block *sb, int wait);
 
+static inline __printf(2, 3)
+void btrfs_no_printk(const struct btrfs_fs_info *fs_info, const char *fmt, ...)
+{
+}
+
 #ifdef CONFIG_PRINTK
 __printf(2, 3)
 void btrfs_printk(const struct btrfs_fs_info *fs_info, const char *fmt, ...);
 #else
-static inline __printf(2, 3)
-void btrfs_printk(const struct btrfs_fs_info *fs_info, const char *fmt, ...)
-{
-}
+#define btrfs_printk(fs_info, fmt, args...) \
+	btrfs_no_printk(fs_info, fmt, ##args)
 #endif
 
 #define btrfs_emerg(fs_info, fmt, args...) \
@@ -3283,7 +3337,35 @@ void btrfs_printk(const struct btrfs_fs_info *fs_info, const char *fmt, ...)
 	btrfs_printk_ratelimited(fs_info, KERN_NOTICE fmt, ##args)
 #define btrfs_info_rl(fs_info, fmt, args...) \
 	btrfs_printk_ratelimited(fs_info, KERN_INFO fmt, ##args)
-#ifdef DEBUG
+
+#if defined(CONFIG_DYNAMIC_DEBUG)
+#define btrfs_debug(fs_info, fmt, args...)				\
+do {									\
+        DEFINE_DYNAMIC_DEBUG_METADATA(descriptor, fmt);         	\
+        if (unlikely(descriptor.flags & _DPRINTK_FLAGS_PRINT))  	\
+		btrfs_printk(fs_info, KERN_DEBUG fmt, ##args);		\
+} while (0)
+#define btrfs_debug_in_rcu(fs_info, fmt, args...) 			\
+do {									\
+        DEFINE_DYNAMIC_DEBUG_METADATA(descriptor, fmt); 	        \
+        if (unlikely(descriptor.flags & _DPRINTK_FLAGS_PRINT)) 		\
+		btrfs_printk_in_rcu(fs_info, KERN_DEBUG fmt, ##args);	\
+} while (0)
+#define btrfs_debug_rl_in_rcu(fs_info, fmt, args...)			\
+do {									\
+        DEFINE_DYNAMIC_DEBUG_METADATA(descriptor, fmt);         	\
+        if (unlikely(descriptor.flags & _DPRINTK_FLAGS_PRINT))  	\
+		btrfs_printk_rl_in_rcu(fs_info, KERN_DEBUG fmt,		\
+				       ##args);\
+} while (0)
+#define btrfs_debug_rl(fs_info, fmt, args...) 				\
+do {									\
+        DEFINE_DYNAMIC_DEBUG_METADATA(descriptor, fmt);         	\
+        if (unlikely(descriptor.flags & _DPRINTK_FLAGS_PRINT))  	\
+		btrfs_printk_ratelimited(fs_info, KERN_DEBUG fmt,	\
+					 ##args);			\
+} while (0)
+#elif defined(DEBUG)
 #define btrfs_debug(fs_info, fmt, args...) \
 	btrfs_printk(fs_info, KERN_DEBUG fmt, ##args)
 #define btrfs_debug_in_rcu(fs_info, fmt, args...) \
@@ -3294,13 +3376,13 @@ void btrfs_printk(const struct btrfs_fs_info *fs_info, const char *fmt, ...)
 	btrfs_printk_ratelimited(fs_info, KERN_DEBUG fmt, ##args)
 #else
 #define btrfs_debug(fs_info, fmt, args...) \
-    no_printk(KERN_DEBUG fmt, ##args)
+	btrfs_no_printk(fs_info, KERN_DEBUG fmt, ##args)
 #define btrfs_debug_in_rcu(fs_info, fmt, args...) \
-	no_printk(KERN_DEBUG fmt, ##args)
+	btrfs_no_printk(fs_info, KERN_DEBUG fmt, ##args)
 #define btrfs_debug_rl_in_rcu(fs_info, fmt, args...) \
-	no_printk(KERN_DEBUG fmt, ##args)
+	btrfs_no_printk(fs_info, KERN_DEBUG fmt, ##args)
 #define btrfs_debug_rl(fs_info, fmt, args...) \
-	no_printk(KERN_DEBUG fmt, ##args)
+	btrfs_no_printk(fs_info, KERN_DEBUG fmt, ##args)
 #endif
 
 #define btrfs_printk_in_rcu(fs_info, fmt, args...)	\
@@ -3331,7 +3413,7 @@ do {								\
 __cold
 static inline void assfail(char *expr, char *file, int line)
 {
-	pr_err("BTRFS: assertion failed: %s, file: %s, line: %d",
+	pr_err("assertion failed: %s, file: %s, line: %d\n",
 	       expr, file, line);
 	BUG();
 }
@@ -3351,23 +3433,23 @@ const char *btrfs_decode_error(int errno);
 
 __cold
 void __btrfs_abort_transaction(struct btrfs_trans_handle *trans,
-			       struct btrfs_root *root, const char *function,
+			       const char *function,
 			       unsigned int line, int errno);
 
 /*
  * Call btrfs_abort_transaction as early as possible when an error condition is
  * detected, that way the exact line number is reported.
  */
-#define btrfs_abort_transaction(trans, root, errno)		\
+#define btrfs_abort_transaction(trans, errno)		\
 do {								\
 	/* Report first abort since mount */			\
 	if (!test_and_set_bit(BTRFS_FS_STATE_TRANS_ABORTED,	\
-			&((root)->fs_info->fs_state))) {	\
+			&((trans)->fs_info->fs_state))) {	\
 		WARN(1, KERN_DEBUG				\
 		"BTRFS: Transaction aborted (error %d)\n",	\
 		(errno));					\
 	}							\
-	__btrfs_abort_transaction((trans), (root), __func__,	\
+	__btrfs_abort_transaction((trans), __func__,		\
 				  __LINE__, (errno));		\
 } while (0)
 
@@ -3599,13 +3681,13 @@ static inline int btrfs_defrag_cancelled(struct btrfs_fs_info *fs_info)
 void btrfs_test_destroy_inode(struct inode *inode);
 #endif
 
-static inline int btrfs_test_is_dummy_root(struct btrfs_root *root)
+static inline int btrfs_is_testing(struct btrfs_fs_info *fs_info)
 {
 #ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
-	if (unlikely(test_bit(BTRFS_ROOT_DUMMY_ROOT, &root->state)))
+	if (unlikely(test_bit(BTRFS_FS_STATE_DUMMY_FS_INFO,
+			      &fs_info->fs_state)))
 		return 1;
 #endif
 	return 0;
 }
-
 #endif
diff --git a/fs/btrfs/dedupe.h b/fs/btrfs/dedupe.h
new file mode 100644
index 000000000000..83ebfe28da9e
--- /dev/null
+++ b/fs/btrfs/dedupe.h
@@ -0,0 +1,24 @@
+/*
+ * Copyright (C) 2016 Fujitsu.  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.
+ */
+
+#ifndef __BTRFS_DEDUPE__
+#define __BTRFS_DEDUPE__
+
+/* later in-band dedupe will expand this struct */
+struct btrfs_dedupe_hash;
+#endif
diff --git a/fs/btrfs/delayed-inode.c b/fs/btrfs/delayed-inode.c
index 61561c2a3f96..0fcf5f25d524 100644
--- a/fs/btrfs/delayed-inode.c
+++ b/fs/btrfs/delayed-inode.c
@@ -34,7 +34,7 @@ int __init btrfs_delayed_inode_init(void)
 	delayed_node_cache = kmem_cache_create("btrfs_delayed_node",
 					sizeof(struct btrfs_delayed_node),
 					0,
-					SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD,
+					SLAB_MEM_SPREAD,
 					NULL);
 	if (!delayed_node_cache)
 		return -ENOMEM;
@@ -385,11 +385,8 @@ static struct btrfs_delayed_item *__btrfs_lookup_delayed_insertion_item(
 					struct btrfs_delayed_node *delayed_node,
 					struct btrfs_key *key)
 {
-	struct btrfs_delayed_item *item;
-
-	item = __btrfs_lookup_delayed_item(&delayed_node->ins_root, key,
+	return __btrfs_lookup_delayed_item(&delayed_node->ins_root, key,
 					   NULL, NULL);
-	return item;
 }
 
 static int __btrfs_add_delayed_item(struct btrfs_delayed_node *delayed_node,
@@ -553,7 +550,7 @@ static int btrfs_delayed_item_reserve_metadata(struct btrfs_trans_handle *trans,
 	dst_rsv = &root->fs_info->delayed_block_rsv;
 
 	num_bytes = btrfs_calc_trans_metadata_size(root, 1);
-	ret = btrfs_block_rsv_migrate(src_rsv, dst_rsv, num_bytes);
+	ret = btrfs_block_rsv_migrate(src_rsv, dst_rsv, num_bytes, 1);
 	if (!ret) {
 		trace_btrfs_space_reservation(root->fs_info, "delayed_item",
 					      item->key.objectid,
@@ -598,6 +595,29 @@ static int btrfs_delayed_inode_reserve_metadata(
 	num_bytes = btrfs_calc_trans_metadata_size(root, 1);
 
 	/*
+	 * If our block_rsv is the delalloc block reserve then check and see if
+	 * we have our extra reservation for updating the inode.  If not fall
+	 * through and try to reserve space quickly.
+	 *
+	 * We used to try and steal from the delalloc block rsv or the global
+	 * reserve, but we'd steal a full reservation, which isn't kind.  We are
+	 * here through delalloc which means we've likely just cowed down close
+	 * to the leaf that contains the inode, so we would steal less just
+	 * doing the fallback inode update, so if we do end up having to steal
+	 * from the global block rsv we hopefully only steal one or two blocks
+	 * worth which is less likely to hurt us.
+	 */
+	if (src_rsv && src_rsv->type == BTRFS_BLOCK_RSV_DELALLOC) {
+		spin_lock(&BTRFS_I(inode)->lock);
+		if (test_and_clear_bit(BTRFS_INODE_DELALLOC_META_RESERVED,
+				       &BTRFS_I(inode)->runtime_flags))
+			release = true;
+		else
+			src_rsv = NULL;
+		spin_unlock(&BTRFS_I(inode)->lock);
+	}
+
+	/*
 	 * btrfs_dirty_inode will update the inode under btrfs_join_transaction
 	 * which doesn't reserve space for speed.  This is a problem since we
 	 * still need to reserve space for this update, so try to reserve the
@@ -626,51 +646,10 @@ static int btrfs_delayed_inode_reserve_metadata(
 						      num_bytes, 1);
 		}
 		return ret;
-	} else if (src_rsv->type == BTRFS_BLOCK_RSV_DELALLOC) {
-		spin_lock(&BTRFS_I(inode)->lock);
-		if (test_and_clear_bit(BTRFS_INODE_DELALLOC_META_RESERVED,
-				       &BTRFS_I(inode)->runtime_flags)) {
-			spin_unlock(&BTRFS_I(inode)->lock);
-			release = true;
-			goto migrate;
-		}
-		spin_unlock(&BTRFS_I(inode)->lock);
-
-		/* Ok we didn't have space pre-reserved.  This shouldn't happen
-		 * too often but it can happen if we do delalloc to an existing
-		 * inode which gets dirtied because of the time update, and then
-		 * isn't touched again until after the transaction commits and
-		 * then we try to write out the data.  First try to be nice and
-		 * reserve something strictly for us.  If not be a pain and try
-		 * to steal from the delalloc block rsv.
-		 */
-		ret = btrfs_block_rsv_add(root, dst_rsv, num_bytes,
-					  BTRFS_RESERVE_NO_FLUSH);
-		if (!ret)
-			goto out;
-
-		ret = btrfs_block_rsv_migrate(src_rsv, dst_rsv, num_bytes);
-		if (!ret)
-			goto out;
-
-		if (btrfs_test_opt(root, ENOSPC_DEBUG)) {
-			btrfs_debug(root->fs_info,
-				    "block rsv migrate returned %d", ret);
-			WARN_ON(1);
-		}
-		/*
-		 * Ok this is a problem, let's just steal from the global rsv
-		 * since this really shouldn't happen that often.
-		 */
-		ret = btrfs_block_rsv_migrate(&root->fs_info->global_block_rsv,
-					      dst_rsv, num_bytes);
-		goto out;
 	}
 
-migrate:
-	ret = btrfs_block_rsv_migrate(src_rsv, dst_rsv, num_bytes);
+	ret = btrfs_block_rsv_migrate(src_rsv, dst_rsv, num_bytes, 1);
 
-out:
 	/*
 	 * Migrate only takes a reservation, it doesn't touch the size of the
 	 * block_rsv.  This is to simplify people who don't normally have things
@@ -1188,7 +1167,7 @@ static int __btrfs_run_delayed_items(struct btrfs_trans_handle *trans,
 		if (ret) {
 			btrfs_release_delayed_node(curr_node);
 			curr_node = NULL;
-			btrfs_abort_transaction(trans, root, ret);
+			btrfs_abort_transaction(trans, ret);
 			break;
 		}
 
@@ -1499,11 +1478,10 @@ int btrfs_insert_delayed_dir_index(struct btrfs_trans_handle *trans,
 	mutex_lock(&delayed_node->mutex);
 	ret = __btrfs_add_delayed_insertion_item(delayed_node, delayed_item);
 	if (unlikely(ret)) {
-		btrfs_err(root->fs_info, "err add delayed dir index item(name: %.*s) "
-				"into the insertion tree of the delayed node"
-				"(root id: %llu, inode id: %llu, errno: %d)",
-				name_len, name, delayed_node->root->objectid,
-				delayed_node->inode_id, ret);
+		btrfs_err(root->fs_info,
+			  "err add delayed dir index item(name: %.*s) into the insertion tree of the delayed node(root id: %llu, inode id: %llu, errno: %d)",
+			  name_len, name, delayed_node->root->objectid,
+			  delayed_node->inode_id, ret);
 		BUG();
 	}
 	mutex_unlock(&delayed_node->mutex);
@@ -1571,11 +1549,9 @@ int btrfs_delete_delayed_dir_index(struct btrfs_trans_handle *trans,
 	mutex_lock(&node->mutex);
 	ret = __btrfs_add_delayed_deletion_item(node, item);
 	if (unlikely(ret)) {
-		btrfs_err(root->fs_info, "err add delayed dir index item(index: %llu) "
-				"into the deletion tree of the delayed node"
-				"(root id: %llu, inode id: %llu, errno: %d)",
-				index, node->root->objectid, node->inode_id,
-				ret);
+		btrfs_err(root->fs_info,
+			  "err add delayed dir index item(index: %llu) into the deletion tree of the delayed node(root id: %llu, inode id: %llu, errno: %d)",
+			  index, node->root->objectid, node->inode_id, ret);
 		BUG();
 	}
 	mutex_unlock(&node->mutex);
@@ -1606,15 +1582,23 @@ int btrfs_inode_delayed_dir_index_count(struct inode *inode)
 	return 0;
 }
 
-void btrfs_get_delayed_items(struct inode *inode, struct list_head *ins_list,
-			     struct list_head *del_list)
+bool btrfs_readdir_get_delayed_items(struct inode *inode,
+				     struct list_head *ins_list,
+				     struct list_head *del_list)
 {
 	struct btrfs_delayed_node *delayed_node;
 	struct btrfs_delayed_item *item;
 
 	delayed_node = btrfs_get_delayed_node(inode);
 	if (!delayed_node)
-		return;
+		return false;
+
+	/*
+	 * We can only do one readdir with delayed items at a time because of
+	 * item->readdir_list.
+	 */
+	inode_unlock_shared(inode);
+	inode_lock(inode);
 
 	mutex_lock(&delayed_node->mutex);
 	item = __btrfs_first_delayed_insertion_item(delayed_node);
@@ -1641,10 +1625,13 @@ void btrfs_get_delayed_items(struct inode *inode, struct list_head *ins_list,
 	 * requeue or dequeue this delayed node.
 	 */
 	atomic_dec(&delayed_node->refs);
+
+	return true;
 }
 
-void btrfs_put_delayed_items(struct list_head *ins_list,
-			     struct list_head *del_list)
+void btrfs_readdir_put_delayed_items(struct inode *inode,
+				     struct list_head *ins_list,
+				     struct list_head *del_list)
 {
 	struct btrfs_delayed_item *curr, *next;
 
@@ -1659,6 +1646,12 @@ void btrfs_put_delayed_items(struct list_head *ins_list,
 		if (atomic_dec_and_test(&curr->refs))
 			kfree(curr);
 	}
+
+	/*
+	 * The VFS is going to do up_read(), so we need to downgrade back to a
+	 * read lock.
+	 */
+	downgrade_write(&inode->i_rwsem);
 }
 
 int btrfs_should_delete_dir_index(struct list_head *del_list,
@@ -1875,7 +1868,8 @@ int btrfs_delayed_delete_inode_ref(struct inode *inode)
 	 * leads to enospc problems.  This means we also can't do
 	 * delayed inode refs
 	 */
-	if (BTRFS_I(inode)->root->fs_info->log_root_recovering)
+	if (test_bit(BTRFS_FS_LOG_RECOVERING,
+		     &BTRFS_I(inode)->root->fs_info->flags))
 		return -EAGAIN;
 
 	delayed_node = btrfs_get_or_create_delayed_node(inode);
diff --git a/fs/btrfs/delayed-inode.h b/fs/btrfs/delayed-inode.h
index 0167853c84ae..2495b3d4075f 100644
--- a/fs/btrfs/delayed-inode.h
+++ b/fs/btrfs/delayed-inode.h
@@ -137,10 +137,12 @@ void btrfs_kill_all_delayed_nodes(struct btrfs_root *root);
 void btrfs_destroy_delayed_inodes(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);
+bool btrfs_readdir_get_delayed_items(struct inode *inode,
+				     struct list_head *ins_list,
+				     struct list_head *del_list);
+void btrfs_readdir_put_delayed_items(struct inode *inode,
+				     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 dir_context *ctx,
diff --git a/fs/btrfs/delayed-ref.c b/fs/btrfs/delayed-ref.c
index 430b3689b112..8d93854a4b4f 100644
--- a/fs/btrfs/delayed-ref.c
+++ b/fs/btrfs/delayed-ref.c
@@ -322,10 +322,11 @@ int btrfs_check_delayed_seq(struct btrfs_fs_info *fs_info,
 		elem = list_first_entry(&fs_info->tree_mod_seq_list,
 					struct seq_list, list);
 		if (seq >= elem->seq) {
-			pr_debug("holding back delayed_ref %#x.%x, lowest is %#x.%x (%p)\n",
-				 (u32)(seq >> 32), (u32)seq,
-				 (u32)(elem->seq >> 32), (u32)elem->seq,
-				 delayed_refs);
+			btrfs_debug(fs_info,
+				"holding back delayed_ref %#x.%x, lowest is %#x.%x (%p)",
+				(u32)(seq >> 32), (u32)seq,
+				(u32)(elem->seq >> 32), (u32)elem->seq,
+				delayed_refs);
 			ret = 1;
 		}
 	}
@@ -541,7 +542,6 @@ add_delayed_ref_head(struct btrfs_fs_info *fs_info,
 	struct btrfs_delayed_ref_head *existing;
 	struct btrfs_delayed_ref_head *head_ref = NULL;
 	struct btrfs_delayed_ref_root *delayed_refs;
-	struct btrfs_qgroup_extent_record *qexisting;
 	int count_mod = 1;
 	int must_insert_reserved = 0;
 
@@ -606,16 +606,15 @@ add_delayed_ref_head(struct btrfs_fs_info *fs_info,
 		qrecord->num_bytes = num_bytes;
 		qrecord->old_roots = NULL;
 
-		qexisting = btrfs_qgroup_insert_dirty_extent(delayed_refs,
-							     qrecord);
-		if (qexisting)
+		if(btrfs_qgroup_insert_dirty_extent_nolock(fs_info,
+					delayed_refs, qrecord))
 			kfree(qrecord);
 	}
 
 	spin_lock_init(&head_ref->lock);
 	mutex_init(&head_ref->mutex);
 
-	trace_add_delayed_ref_head(ref, head_ref, action);
+	trace_add_delayed_ref_head(fs_info, ref, head_ref, action);
 
 	existing = htree_insert(&delayed_refs->href_root,
 				&head_ref->href_node);
@@ -682,7 +681,7 @@ add_delayed_tree_ref(struct btrfs_fs_info *fs_info,
 		ref->type = BTRFS_TREE_BLOCK_REF_KEY;
 	full_ref->level = level;
 
-	trace_add_delayed_tree_ref(ref, full_ref, action);
+	trace_add_delayed_tree_ref(fs_info, ref, full_ref, action);
 
 	ret = add_delayed_ref_tail_merge(trans, delayed_refs, head_ref, ref);
 
@@ -739,7 +738,7 @@ add_delayed_data_ref(struct btrfs_fs_info *fs_info,
 	full_ref->objectid = owner;
 	full_ref->offset = offset;
 
-	trace_add_delayed_data_ref(ref, full_ref, action);
+	trace_add_delayed_data_ref(fs_info, ref, full_ref, action);
 
 	ret = add_delayed_ref_tail_merge(trans, delayed_refs, head_ref, ref);
 
@@ -772,7 +771,8 @@ int btrfs_add_delayed_tree_ref(struct btrfs_fs_info *fs_info,
 	if (!head_ref)
 		goto free_ref;
 
-	if (fs_info->quota_enabled && is_fstree(ref_root)) {
+	if (test_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags) &&
+	    is_fstree(ref_root)) {
 		record = kmalloc(sizeof(*record), GFP_NOFS);
 		if (!record)
 			goto free_head_ref;
@@ -830,7 +830,8 @@ int btrfs_add_delayed_data_ref(struct btrfs_fs_info *fs_info,
 		return -ENOMEM;
 	}
 
-	if (fs_info->quota_enabled && is_fstree(ref_root)) {
+	if (test_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags) &&
+	    is_fstree(ref_root)) {
 		record = kmalloc(sizeof(*record), GFP_NOFS);
 		if (!record) {
 			kmem_cache_free(btrfs_delayed_data_ref_cachep, ref);
@@ -861,33 +862,6 @@ int btrfs_add_delayed_data_ref(struct btrfs_fs_info *fs_info,
 	return 0;
 }
 
-int btrfs_add_delayed_qgroup_reserve(struct btrfs_fs_info *fs_info,
-				     struct btrfs_trans_handle *trans,
-				     u64 ref_root, u64 bytenr, u64 num_bytes)
-{
-	struct btrfs_delayed_ref_root *delayed_refs;
-	struct btrfs_delayed_ref_head *ref_head;
-	int ret = 0;
-
-	if (!fs_info->quota_enabled || !is_fstree(ref_root))
-		return 0;
-
-	delayed_refs = &trans->transaction->delayed_refs;
-
-	spin_lock(&delayed_refs->lock);
-	ref_head = find_ref_head(&delayed_refs->href_root, bytenr, 0);
-	if (!ref_head) {
-		ret = -ENOENT;
-		goto out;
-	}
-	WARN_ON(ref_head->qgroup_reserved || ref_head->qgroup_ref_root);
-	ref_head->qgroup_ref_root = ref_root;
-	ref_head->qgroup_reserved = num_bytes;
-out:
-	spin_unlock(&delayed_refs->lock);
-	return ret;
-}
-
 int btrfs_add_delayed_extent_op(struct btrfs_fs_info *fs_info,
 				struct btrfs_trans_handle *trans,
 				u64 bytenr, u64 num_bytes,
@@ -940,28 +914,28 @@ int btrfs_delayed_ref_init(void)
 	btrfs_delayed_ref_head_cachep = kmem_cache_create(
 				"btrfs_delayed_ref_head",
 				sizeof(struct btrfs_delayed_ref_head), 0,
-				SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD, NULL);
+				SLAB_MEM_SPREAD, NULL);
 	if (!btrfs_delayed_ref_head_cachep)
 		goto fail;
 
 	btrfs_delayed_tree_ref_cachep = kmem_cache_create(
 				"btrfs_delayed_tree_ref",
 				sizeof(struct btrfs_delayed_tree_ref), 0,
-				SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD, NULL);
+				SLAB_MEM_SPREAD, NULL);
 	if (!btrfs_delayed_tree_ref_cachep)
 		goto fail;
 
 	btrfs_delayed_data_ref_cachep = kmem_cache_create(
 				"btrfs_delayed_data_ref",
 				sizeof(struct btrfs_delayed_data_ref), 0,
-				SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD, NULL);
+				SLAB_MEM_SPREAD, NULL);
 	if (!btrfs_delayed_data_ref_cachep)
 		goto fail;
 
 	btrfs_delayed_extent_op_cachep = kmem_cache_create(
 				"btrfs_delayed_extent_op",
 				sizeof(struct btrfs_delayed_extent_op), 0,
-				SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD, NULL);
+				SLAB_MEM_SPREAD, NULL);
 	if (!btrfs_delayed_extent_op_cachep)
 		goto fail;
 
diff --git a/fs/btrfs/delayed-ref.h b/fs/btrfs/delayed-ref.h
index 5fca9534a271..43f3629760e9 100644
--- a/fs/btrfs/delayed-ref.h
+++ b/fs/btrfs/delayed-ref.h
@@ -250,9 +250,6 @@ int btrfs_add_delayed_data_ref(struct btrfs_fs_info *fs_info,
 			       u64 parent, u64 ref_root,
 			       u64 owner, u64 offset, u64 reserved, int action,
 			       struct btrfs_delayed_extent_op *extent_op);
-int btrfs_add_delayed_qgroup_reserve(struct btrfs_fs_info *fs_info,
-				     struct btrfs_trans_handle *trans,
-				     u64 ref_root, u64 bytenr, u64 num_bytes);
 int btrfs_add_delayed_extent_op(struct btrfs_fs_info *fs_info,
 				struct btrfs_trans_handle *trans,
 				u64 bytenr, u64 num_bytes,
diff --git a/fs/btrfs/dev-replace.c b/fs/btrfs/dev-replace.c
index 63ef9cdf0144..05169ef30596 100644
--- a/fs/btrfs/dev-replace.c
+++ b/fs/btrfs/dev-replace.c
@@ -142,7 +142,7 @@ no_valid_dev_replace_entry_found:
 		 * missing
 		 */
 		if (!dev_replace->srcdev &&
-		    !btrfs_test_opt(dev_root, DEGRADED)) {
+		    !btrfs_test_opt(dev_root->fs_info, DEGRADED)) {
 			ret = -EIO;
 			btrfs_warn(fs_info,
 			   "cannot mount because device replace operation is ongoing and");
@@ -151,7 +151,7 @@ no_valid_dev_replace_entry_found:
 			   src_devid);
 		}
 		if (!dev_replace->tgtdev &&
-		    !btrfs_test_opt(dev_root, DEGRADED)) {
+		    !btrfs_test_opt(dev_root->fs_info, DEGRADED)) {
 			ret = -EIO;
 			btrfs_warn(fs_info,
 			   "cannot mount because device replace operation is ongoing and");
@@ -218,8 +218,9 @@ int btrfs_run_dev_replace(struct btrfs_trans_handle *trans,
 	}
 	ret = btrfs_search_slot(trans, dev_root, &key, path, -1, 1);
 	if (ret < 0) {
-		btrfs_warn(fs_info, "error %d while searching for dev_replace item!",
-			ret);
+		btrfs_warn(fs_info,
+			   "error %d while searching for dev_replace item!",
+			   ret);
 		goto out;
 	}
 
@@ -238,8 +239,9 @@ int btrfs_run_dev_replace(struct btrfs_trans_handle *trans,
 		 */
 		ret = btrfs_del_item(trans, dev_root, path);
 		if (ret != 0) {
-			btrfs_warn(fs_info, "delete too small dev_replace item failed %d!",
-				ret);
+			btrfs_warn(fs_info,
+				   "delete too small dev_replace item failed %d!",
+				   ret);
 			goto out;
 		}
 		ret = 1;
@@ -251,8 +253,8 @@ int btrfs_run_dev_replace(struct btrfs_trans_handle *trans,
 		ret = btrfs_insert_empty_item(trans, dev_root, path,
 					      &key, sizeof(*ptr));
 		if (ret < 0) {
-			btrfs_warn(fs_info, "insert dev_replace item failed %d!",
-				ret);
+			btrfs_warn(fs_info,
+				   "insert dev_replace item failed %d!", ret);
 			goto out;
 		}
 	}
@@ -383,7 +385,7 @@ int btrfs_dev_replace_start(struct btrfs_root *root, char *tgtdev_name,
 
 	ret = btrfs_sysfs_add_device_link(tgt_device->fs_devices, tgt_device);
 	if (ret)
-		btrfs_err(fs_info, "kobj add dev failed %d\n", ret);
+		btrfs_err(fs_info, "kobj add dev failed %d", ret);
 
 	btrfs_wait_ordered_roots(root->fs_info, -1, 0, (u64)-1);
 
@@ -772,9 +774,10 @@ int btrfs_resume_dev_replace_async(struct btrfs_fs_info *fs_info)
 		break;
 	}
 	if (!dev_replace->tgtdev || !dev_replace->tgtdev->bdev) {
-		btrfs_info(fs_info, "cannot continue dev_replace, tgtdev is missing");
 		btrfs_info(fs_info,
-			"you may cancel the operation after 'mount -o degraded'");
+			   "cannot continue dev_replace, tgtdev is missing");
+		btrfs_info(fs_info,
+			   "you may cancel the operation after 'mount -o degraded'");
 		btrfs_dev_replace_unlock(dev_replace, 1);
 		return 0;
 	}
diff --git a/fs/btrfs/dir-item.c b/fs/btrfs/dir-item.c
index 1752625fb4dd..0dc1a033275e 100644
--- a/fs/btrfs/dir-item.c
+++ b/fs/btrfs/dir-item.c
@@ -472,9 +472,10 @@ int verify_dir_item(struct btrfs_root *root,
 	/* BTRFS_MAX_XATTR_SIZE is the same for all dir items */
 	if ((btrfs_dir_data_len(leaf, dir_item) +
 	     btrfs_dir_name_len(leaf, dir_item)) > BTRFS_MAX_XATTR_SIZE(root)) {
-		btrfs_crit(root->fs_info, "invalid dir item name + data len: %u + %u",
-		       (unsigned)btrfs_dir_name_len(leaf, dir_item),
-		       (unsigned)btrfs_dir_data_len(leaf, dir_item));
+		btrfs_crit(root->fs_info,
+			   "invalid dir item name + data len: %u + %u",
+			   (unsigned)btrfs_dir_name_len(leaf, dir_item),
+			   (unsigned)btrfs_dir_data_len(leaf, dir_item));
 		return 1;
 	}
 
diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
index 6628fca9f4ed..3a57f99d96aa 100644
--- a/fs/btrfs/disk-io.c
+++ b/fs/btrfs/disk-io.c
@@ -101,7 +101,7 @@ int __init btrfs_end_io_wq_init(void)
 	btrfs_end_io_wq_cache = kmem_cache_create("btrfs_end_io_wq",
 					sizeof(struct btrfs_end_io_wq),
 					0,
-					SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD,
+					SLAB_MEM_SPREAD,
 					NULL);
 	if (!btrfs_end_io_wq_cache)
 		return -ENOMEM;
@@ -124,7 +124,6 @@ struct async_submit_bio {
 	struct list_head list;
 	extent_submit_bio_hook_t *submit_bio_start;
 	extent_submit_bio_hook_t *submit_bio_done;
-	int rw;
 	int mirror_num;
 	unsigned long bio_flags;
 	/*
@@ -327,8 +326,7 @@ static int csum_tree_block(struct btrfs_fs_info *fs_info,
 
 			read_extent_buffer(buf, &val, 0, csum_size);
 			btrfs_warn_rl(fs_info,
-				"%s checksum verify failed on %llu wanted %X found %X "
-				"level %d",
+				"%s checksum verify failed on %llu wanted %X found %X level %d",
 				fs_info->sb->s_id, buf->start,
 				val, found, btrfs_header_level(buf));
 			if (result != (char *)&inline_result)
@@ -403,7 +401,8 @@ out:
  * Return 0 if the superblock checksum type matches the checksum value of that
  * algorithm. Pass the raw disk superblock data.
  */
-static int btrfs_check_super_csum(char *raw_disk_sb)
+static int btrfs_check_super_csum(struct btrfs_fs_info *fs_info,
+				  char *raw_disk_sb)
 {
 	struct btrfs_super_block *disk_sb =
 		(struct btrfs_super_block *)raw_disk_sb;
@@ -429,7 +428,7 @@ static int btrfs_check_super_csum(char *raw_disk_sb)
 	}
 
 	if (csum_type >= ARRAY_SIZE(btrfs_csum_sizes)) {
-		printk(KERN_ERR "BTRFS: unsupported checksum algorithm %u\n",
+		btrfs_err(fs_info, "unsupported checksum algorithm %u",
 				csum_type);
 		ret = 1;
 	}
@@ -443,7 +442,7 @@ static int btrfs_check_super_csum(char *raw_disk_sb)
  */
 static int btree_read_extent_buffer_pages(struct btrfs_root *root,
 					  struct extent_buffer *eb,
-					  u64 start, u64 parent_transid)
+					  u64 parent_transid)
 {
 	struct extent_io_tree *io_tree;
 	int failed = 0;
@@ -455,8 +454,7 @@ static int btree_read_extent_buffer_pages(struct btrfs_root *root,
 	clear_bit(EXTENT_BUFFER_CORRUPT, &eb->bflags);
 	io_tree = &BTRFS_I(root->fs_info->btree_inode)->io_tree;
 	while (1) {
-		ret = read_extent_buffer_pages(io_tree, eb, start,
-					       WAIT_COMPLETE,
+		ret = read_extent_buffer_pages(io_tree, eb, WAIT_COMPLETE,
 					       btree_get_extent, mirror_num);
 		if (!ret) {
 			if (!verify_parent_transid(io_tree, eb,
@@ -548,9 +546,10 @@ static int check_tree_block_fsid(struct btrfs_fs_info *fs_info,
 }
 
 #define CORRUPT(reason, eb, root, slot)				\
-	btrfs_crit(root->fs_info, "corrupt leaf, %s: block=%llu,"	\
-		   "root=%llu, slot=%d", reason,			\
-	       btrfs_header_bytenr(eb),	root->objectid, slot)
+	btrfs_crit(root->fs_info, "corrupt %s, %s: block=%llu,"	\
+		   " root=%llu, slot=%d",			\
+		   btrfs_header_level(eb) == 0 ? "leaf" : "node",\
+		   reason, btrfs_header_bytenr(eb), root->objectid, slot)
 
 static noinline int check_leaf(struct btrfs_root *root,
 			       struct extent_buffer *leaf)
@@ -560,8 +559,29 @@ static noinline int check_leaf(struct btrfs_root *root,
 	u32 nritems = btrfs_header_nritems(leaf);
 	int slot;
 
-	if (nritems == 0)
+	if (nritems == 0) {
+		struct btrfs_root *check_root;
+
+		key.objectid = btrfs_header_owner(leaf);
+		key.type = BTRFS_ROOT_ITEM_KEY;
+		key.offset = (u64)-1;
+
+		check_root = btrfs_get_fs_root(root->fs_info, &key, false);
+		/*
+		 * The only reason we also check NULL here is that during
+		 * open_ctree() some roots has not yet been set up.
+		 */
+		if (!IS_ERR_OR_NULL(check_root)) {
+			/* if leaf is the root, then it's fine */
+			if (leaf->start !=
+			    btrfs_root_bytenr(&check_root->root_item)) {
+				CORRUPT("non-root leaf's nritems is 0",
+					leaf, root, 0);
+				return -EIO;
+			}
+		}
 		return 0;
+	}
 
 	/* Check the 0 item */
 	if (btrfs_item_offset_nr(leaf, 0) + btrfs_item_size_nr(leaf, 0) !=
@@ -613,6 +633,42 @@ static noinline int check_leaf(struct btrfs_root *root,
 	return 0;
 }
 
+static int check_node(struct btrfs_root *root, struct extent_buffer *node)
+{
+	unsigned long nr = btrfs_header_nritems(node);
+	struct btrfs_key key, next_key;
+	int slot;
+	u64 bytenr;
+	int ret = 0;
+
+	if (nr == 0 || nr > BTRFS_NODEPTRS_PER_BLOCK(root)) {
+		btrfs_crit(root->fs_info,
+			   "corrupt node: block %llu root %llu nritems %lu",
+			   node->start, root->objectid, nr);
+		return -EIO;
+	}
+
+	for (slot = 0; slot < nr - 1; slot++) {
+		bytenr = btrfs_node_blockptr(node, slot);
+		btrfs_node_key_to_cpu(node, &key, slot);
+		btrfs_node_key_to_cpu(node, &next_key, slot + 1);
+
+		if (!bytenr) {
+			CORRUPT("invalid item slot", node, root, slot);
+			ret = -EIO;
+			goto out;
+		}
+
+		if (btrfs_comp_cpu_keys(&key, &next_key) >= 0) {
+			CORRUPT("bad key order", node, root, slot);
+			ret = -EIO;
+			goto out;
+		}
+	}
+out:
+	return ret;
+}
+
 static int btree_readpage_end_io_hook(struct btrfs_io_bio *io_bio,
 				      u64 phy_offset, struct page *page,
 				      u64 start, u64 end, int mirror)
@@ -683,6 +739,9 @@ static int btree_readpage_end_io_hook(struct btrfs_io_bio *io_bio,
 		ret = -EIO;
 	}
 
+	if (found_level > 0 && check_node(root, eb))
+		ret = -EIO;
+
 	if (!ret)
 		set_extent_buffer_uptodate(eb);
 err:
@@ -727,7 +786,7 @@ static void end_workqueue_bio(struct bio *bio)
 	fs_info = end_io_wq->info;
 	end_io_wq->error = bio->bi_error;
 
-	if (bio->bi_rw & REQ_WRITE) {
+	if (bio_op(bio) == REQ_OP_WRITE) {
 		if (end_io_wq->metadata == BTRFS_WQ_ENDIO_METADATA) {
 			wq = fs_info->endio_meta_write_workers;
 			func = btrfs_endio_meta_write_helper;
@@ -797,7 +856,7 @@ static void run_one_async_start(struct btrfs_work *work)
 	int ret;
 
 	async = container_of(work, struct  async_submit_bio, work);
-	ret = async->submit_bio_start(async->inode, async->rw, async->bio,
+	ret = async->submit_bio_start(async->inode, async->bio,
 				      async->mirror_num, async->bio_flags,
 				      async->bio_offset);
 	if (ret)
@@ -830,9 +889,8 @@ static void run_one_async_done(struct btrfs_work *work)
 		return;
 	}
 
-	async->submit_bio_done(async->inode, async->rw, async->bio,
-			       async->mirror_num, async->bio_flags,
-			       async->bio_offset);
+	async->submit_bio_done(async->inode, async->bio, async->mirror_num,
+			       async->bio_flags, async->bio_offset);
 }
 
 static void run_one_async_free(struct btrfs_work *work)
@@ -844,7 +902,7 @@ static void run_one_async_free(struct btrfs_work *work)
 }
 
 int btrfs_wq_submit_bio(struct btrfs_fs_info *fs_info, struct inode *inode,
-			int rw, struct bio *bio, int mirror_num,
+			struct bio *bio, int mirror_num,
 			unsigned long bio_flags,
 			u64 bio_offset,
 			extent_submit_bio_hook_t *submit_bio_start,
@@ -857,7 +915,6 @@ int btrfs_wq_submit_bio(struct btrfs_fs_info *fs_info, struct inode *inode,
 		return -ENOMEM;
 
 	async->inode = inode;
-	async->rw = rw;
 	async->bio = bio;
 	async->mirror_num = mirror_num;
 	async->submit_bio_start = submit_bio_start;
@@ -873,7 +930,7 @@ int btrfs_wq_submit_bio(struct btrfs_fs_info *fs_info, struct inode *inode,
 
 	atomic_inc(&fs_info->nr_async_submits);
 
-	if (rw & REQ_SYNC)
+	if (bio->bi_opf & REQ_SYNC)
 		btrfs_set_work_high_priority(&async->work);
 
 	btrfs_queue_work(fs_info->workers, &async->work);
@@ -903,9 +960,8 @@ static int btree_csum_one_bio(struct bio *bio)
 	return ret;
 }
 
-static int __btree_submit_bio_start(struct inode *inode, int rw,
-				    struct bio *bio, int mirror_num,
-				    unsigned long bio_flags,
+static int __btree_submit_bio_start(struct inode *inode, struct bio *bio,
+				    int mirror_num, unsigned long bio_flags,
 				    u64 bio_offset)
 {
 	/*
@@ -915,7 +971,7 @@ static int __btree_submit_bio_start(struct inode *inode, int rw,
 	return btree_csum_one_bio(bio);
 }
 
-static int __btree_submit_bio_done(struct inode *inode, int rw, struct bio *bio,
+static int __btree_submit_bio_done(struct inode *inode, struct bio *bio,
 				 int mirror_num, unsigned long bio_flags,
 				 u64 bio_offset)
 {
@@ -925,7 +981,7 @@ static int __btree_submit_bio_done(struct inode *inode, int rw, struct bio *bio,
 	 * when we're called for a write, we're already in the async
 	 * submission context.  Just jump into btrfs_map_bio
 	 */
-	ret = btrfs_map_bio(BTRFS_I(inode)->root, rw, bio, mirror_num, 1);
+	ret = btrfs_map_bio(BTRFS_I(inode)->root, bio, mirror_num, 1);
 	if (ret) {
 		bio->bi_error = ret;
 		bio_endio(bio);
@@ -944,14 +1000,14 @@ static int check_async_write(struct inode *inode, unsigned long bio_flags)
 	return 1;
 }
 
-static int btree_submit_bio_hook(struct inode *inode, int rw, struct bio *bio,
+static int btree_submit_bio_hook(struct inode *inode, struct bio *bio,
 				 int mirror_num, unsigned long bio_flags,
 				 u64 bio_offset)
 {
 	int async = check_async_write(inode, bio_flags);
 	int ret;
 
-	if (!(rw & REQ_WRITE)) {
+	if (bio_op(bio) != REQ_OP_WRITE) {
 		/*
 		 * called for a read, do the setup so that checksum validation
 		 * can happen in the async kernel threads
@@ -960,21 +1016,19 @@ static int btree_submit_bio_hook(struct inode *inode, int rw, struct bio *bio,
 					  bio, BTRFS_WQ_ENDIO_METADATA);
 		if (ret)
 			goto out_w_error;
-		ret = btrfs_map_bio(BTRFS_I(inode)->root, rw, bio,
-				    mirror_num, 0);
+		ret = btrfs_map_bio(BTRFS_I(inode)->root, bio, mirror_num, 0);
 	} else if (!async) {
 		ret = btree_csum_one_bio(bio);
 		if (ret)
 			goto out_w_error;
-		ret = btrfs_map_bio(BTRFS_I(inode)->root, rw, bio,
-				    mirror_num, 0);
+		ret = btrfs_map_bio(BTRFS_I(inode)->root, bio, mirror_num, 0);
 	} else {
 		/*
 		 * kthread helpers are used to submit writes so that
 		 * checksumming can happen in parallel across all CPUs
 		 */
 		ret = btrfs_wq_submit_bio(BTRFS_I(inode)->root->fs_info,
-					  inode, rw, bio, mirror_num, 0,
+					  inode, bio, mirror_num, 0,
 					  bio_offset,
 					  __btree_submit_bio_start,
 					  __btree_submit_bio_done);
@@ -1098,10 +1152,10 @@ void readahead_tree_block(struct btrfs_root *root, u64 bytenr)
 	struct inode *btree_inode = root->fs_info->btree_inode;
 
 	buf = btrfs_find_create_tree_block(root, bytenr);
-	if (!buf)
+	if (IS_ERR(buf))
 		return;
 	read_extent_buffer_pages(&BTRFS_I(btree_inode)->io_tree,
-				 buf, 0, WAIT_NONE, btree_get_extent, 0);
+				 buf, WAIT_NONE, btree_get_extent, 0);
 	free_extent_buffer(buf);
 }
 
@@ -1114,12 +1168,12 @@ int reada_tree_block_flagged(struct btrfs_root *root, u64 bytenr,
 	int ret;
 
 	buf = btrfs_find_create_tree_block(root, bytenr);
-	if (!buf)
+	if (IS_ERR(buf))
 		return 0;
 
 	set_bit(EXTENT_BUFFER_READAHEAD, &buf->bflags);
 
-	ret = read_extent_buffer_pages(io_tree, buf, 0, WAIT_PAGE_LOCK,
+	ret = read_extent_buffer_pages(io_tree, buf, WAIT_PAGE_LOCK,
 				       btree_get_extent, mirror_num);
 	if (ret) {
 		free_extent_buffer(buf);
@@ -1146,8 +1200,9 @@ struct extent_buffer *btrfs_find_tree_block(struct btrfs_fs_info *fs_info,
 struct extent_buffer *btrfs_find_create_tree_block(struct btrfs_root *root,
 						 u64 bytenr)
 {
-	if (btrfs_test_is_dummy_root(root))
-		return alloc_test_extent_buffer(root->fs_info, bytenr);
+	if (btrfs_is_testing(root->fs_info))
+		return alloc_test_extent_buffer(root->fs_info, bytenr,
+				root->nodesize);
 	return alloc_extent_buffer(root->fs_info, bytenr);
 }
 
@@ -1171,10 +1226,10 @@ struct extent_buffer *read_tree_block(struct btrfs_root *root, u64 bytenr,
 	int ret;
 
 	buf = btrfs_find_create_tree_block(root, bytenr);
-	if (!buf)
-		return ERR_PTR(-ENOMEM);
+	if (IS_ERR(buf))
+		return buf;
 
-	ret = btree_read_extent_buffer_pages(root, buf, 0, parent_transid);
+	ret = btree_read_extent_buffer_pages(root, buf, parent_transid);
 	if (ret) {
 		free_extent_buffer(buf);
 		return ERR_PTR(ret);
@@ -1232,6 +1287,7 @@ static void __setup_root(u32 nodesize, u32 sectorsize, u32 stripesize,
 			 struct btrfs_root *root, struct btrfs_fs_info *fs_info,
 			 u64 objectid)
 {
+	bool dummy = test_bit(BTRFS_FS_STATE_DUMMY_FS_INFO, &fs_info->fs_state);
 	root->node = NULL;
 	root->commit_root = NULL;
 	root->sectorsize = sectorsize;
@@ -1286,14 +1342,14 @@ static void __setup_root(u32 nodesize, u32 sectorsize, u32 stripesize,
 	root->log_transid = 0;
 	root->log_transid_committed = -1;
 	root->last_log_commit = 0;
-	if (fs_info)
+	if (!dummy)
 		extent_io_tree_init(&root->dirty_log_pages,
 				     fs_info->btree_inode->i_mapping);
 
 	memset(&root->root_key, 0, sizeof(root->root_key));
 	memset(&root->root_item, 0, sizeof(root->root_item));
 	memset(&root->defrag_progress, 0, sizeof(root->defrag_progress));
-	if (fs_info)
+	if (!dummy)
 		root->defrag_trans_start = fs_info->generation;
 	else
 		root->defrag_trans_start = 0;
@@ -1314,15 +1370,20 @@ static struct btrfs_root *btrfs_alloc_root(struct btrfs_fs_info *fs_info,
 
 #ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
 /* Should only be used by the testing infrastructure */
-struct btrfs_root *btrfs_alloc_dummy_root(void)
+struct btrfs_root *btrfs_alloc_dummy_root(struct btrfs_fs_info *fs_info,
+					  u32 sectorsize, u32 nodesize)
 {
 	struct btrfs_root *root;
 
-	root = btrfs_alloc_root(NULL, GFP_KERNEL);
+	if (!fs_info)
+		return ERR_PTR(-EINVAL);
+
+	root = btrfs_alloc_root(fs_info, GFP_KERNEL);
 	if (!root)
 		return ERR_PTR(-ENOMEM);
-	__setup_root(4096, 4096, 4096, root, NULL, 1);
-	set_bit(BTRFS_ROOT_DUMMY_ROOT, &root->state);
+	/* We don't use the stripesize in selftest, set it as sectorsize */
+	__setup_root(nodesize, sectorsize, sectorsize, root, fs_info,
+			BTRFS_ROOT_TREE_OBJECTID);
 	root->alloc_bytenr = 0;
 
 	return root;
@@ -1597,14 +1658,14 @@ int btrfs_init_fs_root(struct btrfs_root *root)
 
 	ret = get_anon_bdev(&root->anon_dev);
 	if (ret)
-		goto free_writers;
+		goto fail;
 
 	mutex_lock(&root->objectid_mutex);
 	ret = btrfs_find_highest_objectid(root,
 					&root->highest_objectid);
 	if (ret) {
 		mutex_unlock(&root->objectid_mutex);
-		goto free_root_dev;
+		goto fail;
 	}
 
 	ASSERT(root->highest_objectid <= BTRFS_LAST_FREE_OBJECTID);
@@ -1612,19 +1673,13 @@ int btrfs_init_fs_root(struct btrfs_root *root)
 	mutex_unlock(&root->objectid_mutex);
 
 	return 0;
-
-free_root_dev:
-	free_anon_bdev(root->anon_dev);
-free_writers:
-	btrfs_free_subvolume_writers(root->subv_writers);
 fail:
-	kfree(root->free_ino_ctl);
-	kfree(root->free_ino_pinned);
+	/* the caller is responsible to call free_fs_root */
 	return ret;
 }
 
-static struct btrfs_root *btrfs_lookup_fs_root(struct btrfs_fs_info *fs_info,
-					       u64 root_id)
+struct btrfs_root *btrfs_lookup_fs_root(struct btrfs_fs_info *fs_info,
+					u64 root_id)
 {
 	struct btrfs_root *root;
 
@@ -1803,6 +1858,13 @@ static int cleaner_kthread(void *arg)
 		if (btrfs_need_cleaner_sleep(root))
 			goto sleep;
 
+		/*
+		 * Do not do anything if we might cause open_ctree() to block
+		 * before we have finished mounting the filesystem.
+		 */
+		if (!test_bit(BTRFS_FS_OPEN, &root->fs_info->flags))
+			goto sleep;
+
 		if (!mutex_trylock(&root->fs_info->cleaner_mutex))
 			goto sleep;
 
@@ -2293,9 +2355,8 @@ static void btrfs_init_qgroup(struct btrfs_fs_info *fs_info)
 	fs_info->qgroup_op_tree = RB_ROOT;
 	INIT_LIST_HEAD(&fs_info->dirty_qgroups);
 	fs_info->qgroup_seq = 1;
-	fs_info->quota_enabled = 0;
-	fs_info->pending_quota_state = 0;
 	fs_info->qgroup_ulist = NULL;
+	fs_info->qgroup_rescan_running = false;
 	mutex_init(&fs_info->qgroup_rescan_lock);
 }
 
@@ -2306,17 +2367,19 @@ static int btrfs_init_workqueues(struct btrfs_fs_info *fs_info,
 	unsigned int flags = WQ_MEM_RECLAIM | WQ_FREEZABLE | WQ_UNBOUND;
 
 	fs_info->workers =
-		btrfs_alloc_workqueue("worker", flags | WQ_HIGHPRI,
-				      max_active, 16);
+		btrfs_alloc_workqueue(fs_info, "worker",
+				      flags | WQ_HIGHPRI, max_active, 16);
 
 	fs_info->delalloc_workers =
-		btrfs_alloc_workqueue("delalloc", flags, max_active, 2);
+		btrfs_alloc_workqueue(fs_info, "delalloc",
+				      flags, max_active, 2);
 
 	fs_info->flush_workers =
-		btrfs_alloc_workqueue("flush_delalloc", flags, max_active, 0);
+		btrfs_alloc_workqueue(fs_info, "flush_delalloc",
+				      flags, max_active, 0);
 
 	fs_info->caching_workers =
-		btrfs_alloc_workqueue("cache", flags, max_active, 0);
+		btrfs_alloc_workqueue(fs_info, "cache", flags, max_active, 0);
 
 	/*
 	 * a higher idle thresh on the submit workers makes it much more
@@ -2324,41 +2387,48 @@ static int btrfs_init_workqueues(struct btrfs_fs_info *fs_info,
 	 * devices
 	 */
 	fs_info->submit_workers =
-		btrfs_alloc_workqueue("submit", flags,
+		btrfs_alloc_workqueue(fs_info, "submit", flags,
 				      min_t(u64, fs_devices->num_devices,
 					    max_active), 64);
 
 	fs_info->fixup_workers =
-		btrfs_alloc_workqueue("fixup", flags, 1, 0);
+		btrfs_alloc_workqueue(fs_info, "fixup", flags, 1, 0);
 
 	/*
 	 * endios are largely parallel and should have a very
 	 * low idle thresh
 	 */
 	fs_info->endio_workers =
-		btrfs_alloc_workqueue("endio", flags, max_active, 4);
+		btrfs_alloc_workqueue(fs_info, "endio", flags, max_active, 4);
 	fs_info->endio_meta_workers =
-		btrfs_alloc_workqueue("endio-meta", flags, max_active, 4);
+		btrfs_alloc_workqueue(fs_info, "endio-meta", flags,
+				      max_active, 4);
 	fs_info->endio_meta_write_workers =
-		btrfs_alloc_workqueue("endio-meta-write", flags, max_active, 2);
+		btrfs_alloc_workqueue(fs_info, "endio-meta-write", flags,
+				      max_active, 2);
 	fs_info->endio_raid56_workers =
-		btrfs_alloc_workqueue("endio-raid56", flags, max_active, 4);
+		btrfs_alloc_workqueue(fs_info, "endio-raid56", flags,
+				      max_active, 4);
 	fs_info->endio_repair_workers =
-		btrfs_alloc_workqueue("endio-repair", flags, 1, 0);
+		btrfs_alloc_workqueue(fs_info, "endio-repair", flags, 1, 0);
 	fs_info->rmw_workers =
-		btrfs_alloc_workqueue("rmw", flags, max_active, 2);
+		btrfs_alloc_workqueue(fs_info, "rmw", flags, max_active, 2);
 	fs_info->endio_write_workers =
-		btrfs_alloc_workqueue("endio-write", flags, max_active, 2);
+		btrfs_alloc_workqueue(fs_info, "endio-write", flags,
+				      max_active, 2);
 	fs_info->endio_freespace_worker =
-		btrfs_alloc_workqueue("freespace-write", flags, max_active, 0);
+		btrfs_alloc_workqueue(fs_info, "freespace-write", flags,
+				      max_active, 0);
 	fs_info->delayed_workers =
-		btrfs_alloc_workqueue("delayed-meta", flags, max_active, 0);
+		btrfs_alloc_workqueue(fs_info, "delayed-meta", flags,
+				      max_active, 0);
 	fs_info->readahead_workers =
-		btrfs_alloc_workqueue("readahead", flags, max_active, 2);
+		btrfs_alloc_workqueue(fs_info, "readahead", flags,
+				      max_active, 2);
 	fs_info->qgroup_rescan_workers =
-		btrfs_alloc_workqueue("qgroup-rescan", flags, 1, 0);
+		btrfs_alloc_workqueue(fs_info, "qgroup-rescan", flags, 1, 0);
 	fs_info->extent_workers =
-		btrfs_alloc_workqueue("extent-refs", flags,
+		btrfs_alloc_workqueue(fs_info, "extent-refs", flags,
 				      min_t(u64, fs_devices->num_devices,
 					    max_active), 8);
 
@@ -2469,8 +2539,7 @@ static int btrfs_read_roots(struct btrfs_fs_info *fs_info,
 	root = btrfs_read_tree_root(tree_root, &location);
 	if (!IS_ERR(root)) {
 		set_bit(BTRFS_ROOT_TRACK_DIRTY, &root->state);
-		fs_info->quota_enabled = 1;
-		fs_info->pending_quota_state = 1;
+		set_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags);
 		fs_info->quota_root = root;
 	}
 
@@ -2517,7 +2586,7 @@ int open_ctree(struct super_block *sb,
 	int num_backups_tried = 0;
 	int backup_index = 0;
 	int max_active;
-	bool cleaner_mutex_locked = false;
+	int clear_free_space_tree = 0;
 
 	tree_root = fs_info->tree_root = btrfs_alloc_root(fs_info, GFP_KERNEL);
 	chunk_root = fs_info->chunk_root = btrfs_alloc_root(fs_info, GFP_KERNEL);
@@ -2614,6 +2683,7 @@ int open_ctree(struct super_block *sb,
 	atomic_set(&fs_info->qgroup_op_seq, 0);
 	atomic_set(&fs_info->reada_works_cnt, 0);
 	atomic64_set(&fs_info->tree_mod_seq, 0);
+	fs_info->fs_frozen = 0;
 	fs_info->sb = sb;
 	fs_info->max_inline = BTRFS_DEFAULT_MAX_INLINE;
 	fs_info->metadata_ratio = 0;
@@ -2661,8 +2731,7 @@ int open_ctree(struct super_block *sb,
 	extent_io_tree_init(&fs_info->freed_extents[1],
 			     fs_info->btree_inode->i_mapping);
 	fs_info->pinned_extents = &fs_info->freed_extents[0];
-	fs_info->do_barriers = 1;
-
+	set_bit(BTRFS_FS_BARRIER, &fs_info->flags);
 
 	mutex_init(&fs_info->ordered_operations_mutex);
 	mutex_init(&fs_info->tree_log_mutex);
@@ -2713,7 +2782,7 @@ int open_ctree(struct super_block *sb,
 	 * We want to check superblock checksum, the type is stored inside.
 	 * Pass the whole disk block of size BTRFS_SUPER_INFO_SIZE (4k).
 	 */
-	if (btrfs_check_super_csum(bh->b_data)) {
+	if (btrfs_check_super_csum(fs_info, bh->b_data)) {
 		btrfs_err(fs_info, "superblock checksum mismatch");
 		err = -EINVAL;
 		brelse(bh);
@@ -2797,7 +2866,7 @@ int open_ctree(struct super_block *sb,
 
 	nodesize = btrfs_super_nodesize(disk_super);
 	sectorsize = btrfs_super_sectorsize(disk_super);
-	stripesize = btrfs_super_stripesize(disk_super);
+	stripesize = sectorsize;
 	fs_info->dirty_metadata_batch = nodesize * (1 + ilog2(nr_cpu_ids));
 	fs_info->delalloc_batch = sectorsize * 512 * (1 + ilog2(nr_cpu_ids));
 
@@ -2996,13 +3065,6 @@ retry_root_backup:
 		goto fail_sysfs;
 	}
 
-	/*
-	 * Hold the cleaner_mutex thread here so that we don't block
-	 * for a long time on btrfs_recover_relocation.  cleaner_kthread
-	 * will wait for us to finish mounting the filesystem.
-	 */
-	mutex_lock(&fs_info->cleaner_mutex);
-	cleaner_mutex_locked = true;
 	fs_info->cleaner_kthread = kthread_run(cleaner_kthread, tree_root,
 					       "btrfs-cleaner");
 	if (IS_ERR(fs_info->cleaner_kthread))
@@ -3014,8 +3076,8 @@ retry_root_backup:
 	if (IS_ERR(fs_info->transaction_kthread))
 		goto fail_cleaner;
 
-	if (!btrfs_test_opt(tree_root, SSD) &&
-	    !btrfs_test_opt(tree_root, NOSSD) &&
+	if (!btrfs_test_opt(tree_root->fs_info, SSD) &&
+	    !btrfs_test_opt(tree_root->fs_info, NOSSD) &&
 	    !fs_info->fs_devices->rotating) {
 		btrfs_info(fs_info, "detected SSD devices, enabling SSD mode");
 		btrfs_set_opt(fs_info->mount_opt, SSD);
@@ -3028,9 +3090,9 @@ retry_root_backup:
 	btrfs_apply_pending_changes(fs_info);
 
 #ifdef CONFIG_BTRFS_FS_CHECK_INTEGRITY
-	if (btrfs_test_opt(tree_root, CHECK_INTEGRITY)) {
+	if (btrfs_test_opt(tree_root->fs_info, CHECK_INTEGRITY)) {
 		ret = btrfsic_mount(tree_root, fs_devices,
-				    btrfs_test_opt(tree_root,
+				    btrfs_test_opt(tree_root->fs_info,
 					CHECK_INTEGRITY_INCLUDING_EXTENT_DATA) ?
 				    1 : 0,
 				    fs_info->check_integrity_print_mask);
@@ -3046,7 +3108,7 @@ retry_root_backup:
 
 	/* do not make disk changes in broken FS or nologreplay is given */
 	if (btrfs_super_log_root(disk_super) != 0 &&
-	    !btrfs_test_opt(tree_root, NOLOGREPLAY)) {
+	    !btrfs_test_opt(tree_root->fs_info, NOLOGREPLAY)) {
 		ret = btrfs_replay_log(fs_info, fs_devices);
 		if (ret) {
 			err = ret;
@@ -3062,8 +3124,10 @@ retry_root_backup:
 		ret = btrfs_cleanup_fs_roots(fs_info);
 		if (ret)
 			goto fail_qgroup;
-		/* We locked cleaner_mutex before creating cleaner_kthread. */
+
+		mutex_lock(&fs_info->cleaner_mutex);
 		ret = btrfs_recover_relocation(tree_root);
+		mutex_unlock(&fs_info->cleaner_mutex);
 		if (ret < 0) {
 			btrfs_warn(fs_info, "failed to recover relocation: %d",
 					ret);
@@ -3071,8 +3135,6 @@ retry_root_backup:
 			goto fail_qgroup;
 		}
 	}
-	mutex_unlock(&fs_info->cleaner_mutex);
-	cleaner_mutex_locked = false;
 
 	location.objectid = BTRFS_FS_TREE_OBJECTID;
 	location.type = BTRFS_ROOT_ITEM_KEY;
@@ -3087,7 +3149,27 @@ retry_root_backup:
 	if (sb->s_flags & MS_RDONLY)
 		return 0;
 
-	if (btrfs_test_opt(tree_root, FREE_SPACE_TREE) &&
+	if (btrfs_test_opt(fs_info, CLEAR_CACHE) &&
+	    btrfs_fs_compat_ro(fs_info, FREE_SPACE_TREE)) {
+		clear_free_space_tree = 1;
+	} else if (btrfs_fs_compat_ro(fs_info, FREE_SPACE_TREE) &&
+		   !btrfs_fs_compat_ro(fs_info, FREE_SPACE_TREE_VALID)) {
+		btrfs_warn(fs_info, "free space tree is invalid");
+		clear_free_space_tree = 1;
+	}
+
+	if (clear_free_space_tree) {
+		btrfs_info(fs_info, "clearing free space tree");
+		ret = btrfs_clear_free_space_tree(fs_info);
+		if (ret) {
+			btrfs_warn(fs_info,
+				   "failed to clear free space tree: %d", ret);
+			close_ctree(tree_root);
+			return ret;
+		}
+	}
+
+	if (btrfs_test_opt(tree_root->fs_info, FREE_SPACE_TREE) &&
 	    !btrfs_fs_compat_ro(fs_info, FREE_SPACE_TREE)) {
 		btrfs_info(fs_info, "creating free space tree");
 		ret = btrfs_create_free_space_tree(fs_info);
@@ -3124,18 +3206,6 @@ retry_root_backup:
 
 	btrfs_qgroup_rescan_resume(fs_info);
 
-	if (btrfs_test_opt(tree_root, CLEAR_CACHE) &&
-	    btrfs_fs_compat_ro(fs_info, FREE_SPACE_TREE)) {
-		btrfs_info(fs_info, "clearing free space tree");
-		ret = btrfs_clear_free_space_tree(fs_info);
-		if (ret) {
-			btrfs_warn(fs_info,
-				"failed to clear free space tree: %d", ret);
-			close_ctree(tree_root);
-			return ret;
-		}
-	}
-
 	if (!fs_info->uuid_root) {
 		btrfs_info(fs_info, "creating UUID tree");
 		ret = btrfs_create_uuid_tree(fs_info);
@@ -3145,7 +3215,7 @@ retry_root_backup:
 			close_ctree(tree_root);
 			return ret;
 		}
-	} else if (btrfs_test_opt(tree_root, RESCAN_UUID_TREE) ||
+	} else if (btrfs_test_opt(tree_root->fs_info, RESCAN_UUID_TREE) ||
 		   fs_info->generation !=
 				btrfs_super_uuid_tree_generation(disk_super)) {
 		btrfs_info(fs_info, "checking UUID tree");
@@ -3157,10 +3227,9 @@ retry_root_backup:
 			return ret;
 		}
 	} else {
-		fs_info->update_uuid_tree_gen = 1;
+		set_bit(BTRFS_FS_UPDATE_UUID_TREE_GEN, &fs_info->flags);
 	}
-
-	fs_info->open = 1;
+	set_bit(BTRFS_FS_OPEN, &fs_info->flags);
 
 	/*
 	 * backuproot only affect mount behavior, and if open_ctree succeeded,
@@ -3186,10 +3255,6 @@ fail_cleaner:
 	filemap_write_and_wait(fs_info->btree_inode->i_mapping);
 
 fail_sysfs:
-	if (cleaner_mutex_locked) {
-		mutex_unlock(&fs_info->cleaner_mutex);
-		cleaner_mutex_locked = false;
-	}
 	btrfs_sysfs_remove_mounted(fs_info);
 
 fail_fsdev_sysfs:
@@ -3226,7 +3291,7 @@ fail:
 	return err;
 
 recovery_tree_root:
-	if (!btrfs_test_opt(tree_root, USEBACKUPROOT))
+	if (!btrfs_test_opt(tree_root->fs_info, USEBACKUPROOT))
 		goto fail_tree_roots;
 
 	free_root_pointers(fs_info, 0);
@@ -3420,9 +3485,9 @@ static int write_dev_supers(struct btrfs_device *device,
 		 * to go down lazy.
 		 */
 		if (i == 0)
-			ret = btrfsic_submit_bh(WRITE_FUA, bh);
+			ret = btrfsic_submit_bh(REQ_OP_WRITE, WRITE_FUA, bh);
 		else
-			ret = btrfsic_submit_bh(WRITE_SYNC, bh);
+			ret = btrfsic_submit_bh(REQ_OP_WRITE, WRITE_SYNC, bh);
 		if (ret)
 			errors++;
 	}
@@ -3486,12 +3551,13 @@ static int write_dev_flush(struct btrfs_device *device, int wait)
 
 	bio->bi_end_io = btrfs_end_empty_barrier;
 	bio->bi_bdev = device->bdev;
+	bio_set_op_attrs(bio, REQ_OP_WRITE, WRITE_FLUSH);
 	init_completion(&device->flush_wait);
 	bio->bi_private = &device->flush_wait;
 	device->flush_bio = bio;
 
 	bio_get(bio);
-	btrfsic_submit_bio(WRITE_FLUSH, bio);
+	btrfsic_submit_bio(bio);
 
 	return 0;
 }
@@ -3568,7 +3634,7 @@ int btrfs_get_num_tolerated_disk_barrier_failures(u64 flags)
 	}
 
 	if (min_tolerated == INT_MAX) {
-		pr_warn("BTRFS: unknown raid flag: %llu\n", flags);
+		pr_warn("BTRFS: unknown raid flag: %llu", flags);
 		min_tolerated = 0;
 	}
 
@@ -3641,7 +3707,7 @@ static int write_all_supers(struct btrfs_root *root, int max_mirrors)
 	int total_errors = 0;
 	u64 flags;
 
-	do_barriers = !btrfs_test_opt(root, NOBARRIER);
+	do_barriers = !btrfs_test_opt(root->fs_info, NOBARRIER);
 	backup_super_roots(root->fs_info);
 
 	sb = root->fs_info->super_for_commit;
@@ -3739,8 +3805,15 @@ void btrfs_drop_and_free_fs_root(struct btrfs_fs_info *fs_info,
 	if (btrfs_root_refs(&root->root_item) == 0)
 		synchronize_srcu(&fs_info->subvol_srcu);
 
-	if (test_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state))
+	if (test_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state)) {
 		btrfs_free_log(NULL, root);
+		if (root->reloc_root) {
+			free_extent_buffer(root->reloc_root->node);
+			free_extent_buffer(root->reloc_root->commit_root);
+			btrfs_put_fs_root(root->reloc_root);
+			root->reloc_root = NULL;
+		}
+	}
 
 	if (root->free_ino_pinned)
 		__btrfs_remove_free_space_cache(root->free_ino_pinned);
@@ -3847,11 +3920,10 @@ void close_ctree(struct btrfs_root *root)
 	struct btrfs_fs_info *fs_info = root->fs_info;
 	int ret;
 
-	fs_info->closing = 1;
-	smp_mb();
+	set_bit(BTRFS_FS_CLOSING_START, &fs_info->flags);
 
 	/* wait for the qgroup rescan worker to stop */
-	btrfs_qgroup_wait_for_completion(fs_info);
+	btrfs_qgroup_wait_for_completion(fs_info, false);
 
 	/* wait for the uuid_scan task to finish */
 	down(&fs_info->uuid_tree_rescan_sem);
@@ -3893,8 +3965,7 @@ void close_ctree(struct btrfs_root *root)
 	kthread_stop(fs_info->transaction_kthread);
 	kthread_stop(fs_info->cleaner_kthread);
 
-	fs_info->closing = 2;
-	smp_mb();
+	set_bit(BTRFS_FS_CLOSING_DONE, &fs_info->flags);
 
 	btrfs_free_qgroup_config(fs_info);
 
@@ -3919,13 +3990,13 @@ void close_ctree(struct btrfs_root *root)
 	invalidate_inode_pages2(fs_info->btree_inode->i_mapping);
 	btrfs_stop_all_workers(fs_info);
 
-	fs_info->open = 0;
+	clear_bit(BTRFS_FS_OPEN, &fs_info->flags);
 	free_root_pointers(fs_info, 1);
 
 	iput(fs_info->btree_inode);
 
 #ifdef CONFIG_BTRFS_FS_CHECK_INTEGRITY
-	if (btrfs_test_opt(root, CHECK_INTEGRITY))
+	if (btrfs_test_opt(root->fs_info, CHECK_INTEGRITY))
 		btrfsic_unmount(root, fs_info->fs_devices);
 #endif
 
@@ -3990,8 +4061,7 @@ void btrfs_mark_buffer_dirty(struct extent_buffer *buf)
 	root = BTRFS_I(buf->pages[0]->mapping->host)->root;
 	btrfs_assert_tree_locked(buf);
 	if (transid != root->fs_info->generation)
-		WARN(1, KERN_CRIT "btrfs transid mismatch buffer %llu, "
-		       "found %llu running %llu\n",
+		WARN(1, KERN_CRIT "btrfs transid mismatch buffer %llu, found %llu running %llu\n",
 			buf->start, transid, root->fs_info->generation);
 	was_dirty = set_extent_buffer_dirty(buf);
 	if (!was_dirty)
@@ -4042,7 +4112,7 @@ void btrfs_btree_balance_dirty_nodelay(struct btrfs_root *root)
 int btrfs_read_buffer(struct extent_buffer *buf, u64 parent_transid)
 {
 	struct btrfs_root *root = BTRFS_I(buf->pages[0]->mapping->host)->root;
-	return btree_read_extent_buffer_pages(root, buf, 0, parent_transid);
+	return btree_read_extent_buffer_pages(root, buf, parent_transid);
 }
 
 static int btrfs_check_super_valid(struct btrfs_fs_info *fs_info,
@@ -4054,24 +4124,24 @@ static int btrfs_check_super_valid(struct btrfs_fs_info *fs_info,
 	int ret = 0;
 
 	if (btrfs_super_magic(sb) != BTRFS_MAGIC) {
-		printk(KERN_ERR "BTRFS: no valid FS found\n");
+		btrfs_err(fs_info, "no valid FS found");
 		ret = -EINVAL;
 	}
 	if (btrfs_super_flags(sb) & ~BTRFS_SUPER_FLAG_SUPP)
-		printk(KERN_WARNING "BTRFS: unrecognized super flag: %llu\n",
+		btrfs_warn(fs_info, "unrecognized super flag: %llu",
 				btrfs_super_flags(sb) & ~BTRFS_SUPER_FLAG_SUPP);
 	if (btrfs_super_root_level(sb) >= BTRFS_MAX_LEVEL) {
-		printk(KERN_ERR "BTRFS: tree_root level too big: %d >= %d\n",
+		btrfs_err(fs_info, "tree_root level too big: %d >= %d",
 				btrfs_super_root_level(sb), BTRFS_MAX_LEVEL);
 		ret = -EINVAL;
 	}
 	if (btrfs_super_chunk_root_level(sb) >= BTRFS_MAX_LEVEL) {
-		printk(KERN_ERR "BTRFS: chunk_root level too big: %d >= %d\n",
+		btrfs_err(fs_info, "chunk_root level too big: %d >= %d",
 				btrfs_super_chunk_root_level(sb), BTRFS_MAX_LEVEL);
 		ret = -EINVAL;
 	}
 	if (btrfs_super_log_root_level(sb) >= BTRFS_MAX_LEVEL) {
-		printk(KERN_ERR "BTRFS: log_root level too big: %d >= %d\n",
+		btrfs_err(fs_info, "log_root level too big: %d >= %d",
 				btrfs_super_log_root_level(sb), BTRFS_MAX_LEVEL);
 		ret = -EINVAL;
 	}
@@ -4082,47 +4152,48 @@ static int btrfs_check_super_valid(struct btrfs_fs_info *fs_info,
 	 */
 	if (!is_power_of_2(sectorsize) || sectorsize < 4096 ||
 	    sectorsize > BTRFS_MAX_METADATA_BLOCKSIZE) {
-		printk(KERN_ERR "BTRFS: invalid sectorsize %llu\n", sectorsize);
+		btrfs_err(fs_info, "invalid sectorsize %llu", sectorsize);
 		ret = -EINVAL;
 	}
 	/* Only PAGE SIZE is supported yet */
 	if (sectorsize != PAGE_SIZE) {
-		printk(KERN_ERR "BTRFS: sectorsize %llu not supported yet, only support %lu\n",
-				sectorsize, PAGE_SIZE);
+		btrfs_err(fs_info,
+			"sectorsize %llu not supported yet, only support %lu",
+			sectorsize, PAGE_SIZE);
 		ret = -EINVAL;
 	}
 	if (!is_power_of_2(nodesize) || nodesize < sectorsize ||
 	    nodesize > BTRFS_MAX_METADATA_BLOCKSIZE) {
-		printk(KERN_ERR "BTRFS: invalid nodesize %llu\n", nodesize);
+		btrfs_err(fs_info, "invalid nodesize %llu", nodesize);
 		ret = -EINVAL;
 	}
 	if (nodesize != le32_to_cpu(sb->__unused_leafsize)) {
-		printk(KERN_ERR "BTRFS: invalid leafsize %u, should be %llu\n",
-				le32_to_cpu(sb->__unused_leafsize),
-				nodesize);
+		btrfs_err(fs_info, "invalid leafsize %u, should be %llu",
+			  le32_to_cpu(sb->__unused_leafsize), nodesize);
 		ret = -EINVAL;
 	}
 
 	/* Root alignment check */
 	if (!IS_ALIGNED(btrfs_super_root(sb), sectorsize)) {
-		printk(KERN_WARNING "BTRFS: tree_root block unaligned: %llu\n",
-				btrfs_super_root(sb));
+		btrfs_warn(fs_info, "tree_root block unaligned: %llu",
+			   btrfs_super_root(sb));
 		ret = -EINVAL;
 	}
 	if (!IS_ALIGNED(btrfs_super_chunk_root(sb), sectorsize)) {
-		printk(KERN_WARNING "BTRFS: chunk_root block unaligned: %llu\n",
-				btrfs_super_chunk_root(sb));
+		btrfs_warn(fs_info, "chunk_root block unaligned: %llu",
+			   btrfs_super_chunk_root(sb));
 		ret = -EINVAL;
 	}
 	if (!IS_ALIGNED(btrfs_super_log_root(sb), sectorsize)) {
-		printk(KERN_WARNING "BTRFS: log_root block unaligned: %llu\n",
-				btrfs_super_log_root(sb));
+		btrfs_warn(fs_info, "log_root block unaligned: %llu",
+			   btrfs_super_log_root(sb));
 		ret = -EINVAL;
 	}
 
 	if (memcmp(fs_info->fsid, sb->dev_item.fsid, BTRFS_UUID_SIZE) != 0) {
-		printk(KERN_ERR "BTRFS: dev_item UUID does not match fsid: %pU != %pU\n",
-				fs_info->fsid, sb->dev_item.fsid);
+		btrfs_err(fs_info,
+			   "dev_item UUID does not match fsid: %pU != %pU",
+			   fs_info->fsid, sb->dev_item.fsid);
 		ret = -EINVAL;
 	}
 
@@ -4130,17 +4201,27 @@ static int btrfs_check_super_valid(struct btrfs_fs_info *fs_info,
 	 * Hint to catch really bogus numbers, bitflips or so, more exact checks are
 	 * done later
 	 */
+	if (btrfs_super_bytes_used(sb) < 6 * btrfs_super_nodesize(sb)) {
+		btrfs_err(fs_info, "bytes_used is too small %llu",
+			  btrfs_super_bytes_used(sb));
+		ret = -EINVAL;
+	}
+	if (!is_power_of_2(btrfs_super_stripesize(sb))) {
+		btrfs_err(fs_info, "invalid stripesize %u",
+			  btrfs_super_stripesize(sb));
+		ret = -EINVAL;
+	}
 	if (btrfs_super_num_devices(sb) > (1UL << 31))
-		printk(KERN_WARNING "BTRFS: suspicious number of devices: %llu\n",
-				btrfs_super_num_devices(sb));
+		btrfs_warn(fs_info, "suspicious number of devices: %llu",
+			   btrfs_super_num_devices(sb));
 	if (btrfs_super_num_devices(sb) == 0) {
-		printk(KERN_ERR "BTRFS: number of devices is 0\n");
+		btrfs_err(fs_info, "number of devices is 0");
 		ret = -EINVAL;
 	}
 
 	if (btrfs_super_bytenr(sb) != BTRFS_SUPER_INFO_OFFSET) {
-		printk(KERN_ERR "BTRFS: super offset mismatch %llu != %u\n",
-				btrfs_super_bytenr(sb), BTRFS_SUPER_INFO_OFFSET);
+		btrfs_err(fs_info, "super offset mismatch %llu != %u",
+			  btrfs_super_bytenr(sb), BTRFS_SUPER_INFO_OFFSET);
 		ret = -EINVAL;
 	}
 
@@ -4149,17 +4230,17 @@ static int btrfs_check_super_valid(struct btrfs_fs_info *fs_info,
 	 * and one chunk
 	 */
 	if (btrfs_super_sys_array_size(sb) > BTRFS_SYSTEM_CHUNK_ARRAY_SIZE) {
-		printk(KERN_ERR "BTRFS: system chunk array too big %u > %u\n",
-				btrfs_super_sys_array_size(sb),
-				BTRFS_SYSTEM_CHUNK_ARRAY_SIZE);
+		btrfs_err(fs_info, "system chunk array too big %u > %u",
+			  btrfs_super_sys_array_size(sb),
+			  BTRFS_SYSTEM_CHUNK_ARRAY_SIZE);
 		ret = -EINVAL;
 	}
 	if (btrfs_super_sys_array_size(sb) < sizeof(struct btrfs_disk_key)
 			+ sizeof(struct btrfs_chunk)) {
-		printk(KERN_ERR "BTRFS: system chunk array too small %u < %zu\n",
-				btrfs_super_sys_array_size(sb),
-				sizeof(struct btrfs_disk_key)
-				+ sizeof(struct btrfs_chunk));
+		btrfs_err(fs_info, "system chunk array too small %u < %zu",
+			  btrfs_super_sys_array_size(sb),
+			  sizeof(struct btrfs_disk_key)
+			  + sizeof(struct btrfs_chunk));
 		ret = -EINVAL;
 	}
 
@@ -4168,14 +4249,16 @@ static int btrfs_check_super_valid(struct btrfs_fs_info *fs_info,
 	 * but it's still possible that it's the one that's wrong.
 	 */
 	if (btrfs_super_generation(sb) < btrfs_super_chunk_root_generation(sb))
-		printk(KERN_WARNING
-			"BTRFS: suspicious: generation < chunk_root_generation: %llu < %llu\n",
-			btrfs_super_generation(sb), btrfs_super_chunk_root_generation(sb));
+		btrfs_warn(fs_info,
+			"suspicious: generation < chunk_root_generation: %llu < %llu",
+			btrfs_super_generation(sb),
+			btrfs_super_chunk_root_generation(sb));
 	if (btrfs_super_generation(sb) < btrfs_super_cache_generation(sb)
 	    && btrfs_super_cache_generation(sb) != (u64)-1)
-		printk(KERN_WARNING
-			"BTRFS: suspicious: generation < cache_generation: %llu < %llu\n",
-			btrfs_super_generation(sb), btrfs_super_cache_generation(sb));
+		btrfs_warn(fs_info,
+			"suspicious: generation < cache_generation: %llu < %llu",
+			btrfs_super_generation(sb),
+			btrfs_super_cache_generation(sb));
 
 	return ret;
 }
@@ -4419,9 +4502,80 @@ again:
 	return 0;
 }
 
+static void btrfs_cleanup_bg_io(struct btrfs_block_group_cache *cache)
+{
+	struct inode *inode;
+
+	inode = cache->io_ctl.inode;
+	if (inode) {
+		invalidate_inode_pages2(inode->i_mapping);
+		BTRFS_I(inode)->generation = 0;
+		cache->io_ctl.inode = NULL;
+		iput(inode);
+	}
+	btrfs_put_block_group(cache);
+}
+
+void btrfs_cleanup_dirty_bgs(struct btrfs_transaction *cur_trans,
+			     struct btrfs_root *root)
+{
+	struct btrfs_block_group_cache *cache;
+
+	spin_lock(&cur_trans->dirty_bgs_lock);
+	while (!list_empty(&cur_trans->dirty_bgs)) {
+		cache = list_first_entry(&cur_trans->dirty_bgs,
+					 struct btrfs_block_group_cache,
+					 dirty_list);
+		if (!cache) {
+			btrfs_err(root->fs_info,
+				  "orphan block group dirty_bgs list");
+			spin_unlock(&cur_trans->dirty_bgs_lock);
+			return;
+		}
+
+		if (!list_empty(&cache->io_list)) {
+			spin_unlock(&cur_trans->dirty_bgs_lock);
+			list_del_init(&cache->io_list);
+			btrfs_cleanup_bg_io(cache);
+			spin_lock(&cur_trans->dirty_bgs_lock);
+		}
+
+		list_del_init(&cache->dirty_list);
+		spin_lock(&cache->lock);
+		cache->disk_cache_state = BTRFS_DC_ERROR;
+		spin_unlock(&cache->lock);
+
+		spin_unlock(&cur_trans->dirty_bgs_lock);
+		btrfs_put_block_group(cache);
+		spin_lock(&cur_trans->dirty_bgs_lock);
+	}
+	spin_unlock(&cur_trans->dirty_bgs_lock);
+
+	while (!list_empty(&cur_trans->io_bgs)) {
+		cache = list_first_entry(&cur_trans->io_bgs,
+					 struct btrfs_block_group_cache,
+					 io_list);
+		if (!cache) {
+			btrfs_err(root->fs_info,
+				  "orphan block group on io_bgs list");
+			return;
+		}
+
+		list_del_init(&cache->io_list);
+		spin_lock(&cache->lock);
+		cache->disk_cache_state = BTRFS_DC_ERROR;
+		spin_unlock(&cache->lock);
+		btrfs_cleanup_bg_io(cache);
+	}
+}
+
 void btrfs_cleanup_one_transaction(struct btrfs_transaction *cur_trans,
 				   struct btrfs_root *root)
 {
+	btrfs_cleanup_dirty_bgs(cur_trans, root);
+	ASSERT(list_empty(&cur_trans->dirty_bgs));
+	ASSERT(list_empty(&cur_trans->io_bgs));
+
 	btrfs_destroy_delayed_refs(cur_trans, root);
 
 	cur_trans->state = TRANS_STATE_COMMIT_START;
diff --git a/fs/btrfs/disk-io.h b/fs/btrfs/disk-io.h
index 8e79d0070bcf..1a3237e5700f 100644
--- a/fs/btrfs/disk-io.h
+++ b/fs/btrfs/disk-io.h
@@ -68,6 +68,8 @@ struct extent_buffer *btrfs_find_tree_block(struct btrfs_fs_info *fs_info,
 struct btrfs_root *btrfs_read_fs_root(struct btrfs_root *tree_root,
 				      struct btrfs_key *location);
 int btrfs_init_fs_root(struct btrfs_root *root);
+struct btrfs_root *btrfs_lookup_fs_root(struct btrfs_fs_info *fs_info,
+					u64 root_id);
 int btrfs_insert_fs_root(struct btrfs_fs_info *fs_info,
 			 struct btrfs_root *root);
 void btrfs_free_fs_roots(struct btrfs_fs_info *fs_info);
@@ -90,7 +92,8 @@ void btrfs_drop_and_free_fs_root(struct btrfs_fs_info *fs_info,
 void btrfs_free_fs_root(struct btrfs_root *root);
 
 #ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
-struct btrfs_root *btrfs_alloc_dummy_root(void);
+struct btrfs_root *btrfs_alloc_dummy_root(struct btrfs_fs_info *fs_info,
+					  u32 sectorsize, u32 nodesize);
 #endif
 
 /*
@@ -122,7 +125,7 @@ void btrfs_csum_final(u32 crc, char *result);
 int btrfs_bio_wq_end_io(struct btrfs_fs_info *info, struct bio *bio,
 			enum btrfs_wq_endio_type metadata);
 int btrfs_wq_submit_bio(struct btrfs_fs_info *fs_info, struct inode *inode,
-			int rw, struct bio *bio, int mirror_num,
+			struct bio *bio, int mirror_num,
 			unsigned long bio_flags, u64 bio_offset,
 			extent_submit_bio_hook_t *submit_bio_start,
 			extent_submit_bio_hook_t *submit_bio_done);
@@ -133,6 +136,8 @@ int btrfs_init_log_root_tree(struct btrfs_trans_handle *trans,
 			     struct btrfs_fs_info *fs_info);
 int btrfs_add_log_tree(struct btrfs_trans_handle *trans,
 		       struct btrfs_root *root);
+void btrfs_cleanup_dirty_bgs(struct btrfs_transaction *trans,
+			     struct btrfs_root *root);
 void btrfs_cleanup_one_transaction(struct btrfs_transaction *trans,
 				  struct btrfs_root *root);
 struct btrfs_root *btrfs_create_tree(struct btrfs_trans_handle *trans,
diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index a400951e8678..210c94ac8818 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -60,21 +60,6 @@ enum {
 	CHUNK_ALLOC_FORCE = 2,
 };
 
-/*
- * Control how reservations are dealt with.
- *
- * RESERVE_FREE - freeing a reservation.
- * RESERVE_ALLOC - allocating space and we need to update bytes_may_use for
- *   ENOSPC accounting
- * RESERVE_ALLOC_NO_ACCOUNT - allocating space and we should not update
- *   bytes_may_use as the ENOSPC accounting is done elsewhere
- */
-enum {
-	RESERVE_FREE = 0,
-	RESERVE_ALLOC = 1,
-	RESERVE_ALLOC_NO_ACCOUNT = 2,
-};
-
 static int update_block_group(struct btrfs_trans_handle *trans,
 			      struct btrfs_root *root, u64 bytenr,
 			      u64 num_bytes, int alloc);
@@ -102,15 +87,27 @@ static int do_chunk_alloc(struct btrfs_trans_handle *trans,
 			  int force);
 static int find_next_key(struct btrfs_path *path, int level,
 			 struct btrfs_key *key);
-static void dump_space_info(struct btrfs_space_info *info, u64 bytes,
+static void dump_space_info(struct btrfs_fs_info *fs_info,
+			    struct btrfs_space_info *info, u64 bytes,
 			    int dump_block_groups);
-static int btrfs_update_reserved_bytes(struct btrfs_block_group_cache *cache,
-				       u64 num_bytes, int reserve,
-				       int delalloc);
+static int btrfs_add_reserved_bytes(struct btrfs_block_group_cache *cache,
+				    u64 ram_bytes, u64 num_bytes, int delalloc);
+static int btrfs_free_reserved_bytes(struct btrfs_block_group_cache *cache,
+				     u64 num_bytes, int delalloc);
 static int block_rsv_use_bytes(struct btrfs_block_rsv *block_rsv,
 			       u64 num_bytes);
 int btrfs_pin_extent(struct btrfs_root *root,
 		     u64 bytenr, u64 num_bytes, int reserved);
+static int __reserve_metadata_bytes(struct btrfs_root *root,
+				    struct btrfs_space_info *space_info,
+				    u64 orig_bytes,
+				    enum btrfs_reserve_flush_enum flush);
+static void space_info_add_new_bytes(struct btrfs_fs_info *fs_info,
+				     struct btrfs_space_info *space_info,
+				     u64 num_bytes);
+static void space_info_add_old_bytes(struct btrfs_fs_info *fs_info,
+				     struct btrfs_space_info *space_info,
+				     u64 num_bytes);
 
 static noinline int
 block_group_cache_done(struct btrfs_block_group_cache *cache)
@@ -270,9 +267,8 @@ static int exclude_super_stripes(struct btrfs_root *root,
 
 	for (i = 0; i < BTRFS_SUPER_MIRROR_MAX; i++) {
 		bytenr = btrfs_sb_offset(i);
-		ret = btrfs_rmap_block(&root->fs_info->mapping_tree,
-				       cache->key.objectid, bytenr,
-				       0, &logical, &nr, &stripe_len);
+		ret = btrfs_rmap_block(root->fs_info, cache->key.objectid,
+				       bytenr, 0, &logical, &nr, &stripe_len);
 		if (ret)
 			return ret;
 
@@ -734,11 +730,7 @@ static int cache_block_group(struct btrfs_block_group_cache *cache,
 static struct btrfs_block_group_cache *
 btrfs_lookup_first_block_group(struct btrfs_fs_info *info, u64 bytenr)
 {
-	struct btrfs_block_group_cache *cache;
-
-	cache = block_group_cache_tree_search(info, bytenr, 0);
-
-	return cache;
+	return block_group_cache_tree_search(info, bytenr, 0);
 }
 
 /*
@@ -748,11 +740,7 @@ struct btrfs_block_group_cache *btrfs_lookup_block_group(
 						 struct btrfs_fs_info *info,
 						 u64 bytenr)
 {
-	struct btrfs_block_group_cache *cache;
-
-	cache = block_group_cache_tree_search(info, bytenr, 1);
-
-	return cache;
+	return block_group_cache_tree_search(info, bytenr, 1);
 }
 
 static struct btrfs_space_info *__find_space_info(struct btrfs_fs_info *info,
@@ -2042,8 +2030,13 @@ int btrfs_discard_extent(struct btrfs_root *root, u64 bytenr,
 	struct btrfs_bio *bbio = NULL;
 
 
+	/*
+	 * Avoid races with device replace and make sure our bbio has devices
+	 * associated to its stripes that don't go away while we are discarding.
+	 */
+	btrfs_bio_counter_inc_blocked(root->fs_info);
 	/* Tell the block device(s) that the sectors can be discarded */
-	ret = btrfs_map_block(root->fs_info, REQ_DISCARD,
+	ret = btrfs_map_block(root->fs_info, REQ_OP_DISCARD,
 			      bytenr, &num_bytes, &bbio, 0);
 	/* Error condition is -ENOMEM */
 	if (!ret) {
@@ -2074,6 +2067,7 @@ int btrfs_discard_extent(struct btrfs_root *root, u64 bytenr,
 		}
 		btrfs_put_bbio(bbio);
 	}
+	btrfs_bio_counter_dec(root->fs_info);
 
 	if (actual_bytes)
 		*actual_bytes = discarded_bytes;
@@ -2164,7 +2158,7 @@ static int __btrfs_inc_extent_ref(struct btrfs_trans_handle *trans,
 				    path, bytenr, parent, root_objectid,
 				    owner, offset, refs_to_add);
 	if (ret)
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 out:
 	btrfs_free_path(path);
 	return ret;
@@ -2188,7 +2182,7 @@ static int run_delayed_data_ref(struct btrfs_trans_handle *trans,
 	ins.type = BTRFS_EXTENT_ITEM_KEY;
 
 	ref = btrfs_delayed_node_to_data_ref(node);
-	trace_run_delayed_data_ref(node, ref, node->action);
+	trace_run_delayed_data_ref(root->fs_info, node, ref, node->action);
 
 	if (node->type == BTRFS_SHARED_DATA_REF_KEY)
 		parent = ref->parent;
@@ -2343,7 +2337,7 @@ static int run_delayed_tree_ref(struct btrfs_trans_handle *trans,
 						 SKINNY_METADATA);
 
 	ref = btrfs_delayed_node_to_tree_ref(node);
-	trace_run_delayed_tree_ref(node, ref, node->action);
+	trace_run_delayed_tree_ref(root->fs_info, node, ref, node->action);
 
 	if (node->type == BTRFS_SHARED_BLOCK_REF_KEY)
 		parent = ref->parent;
@@ -2358,7 +2352,13 @@ static int run_delayed_tree_ref(struct btrfs_trans_handle *trans,
 		ins.type = BTRFS_EXTENT_ITEM_KEY;
 	}
 
-	BUG_ON(node->ref_mod != 1);
+	if (node->ref_mod != 1) {
+		btrfs_err(root->fs_info,
+	"btree block(%llu) has %d references rather than 1: action %d ref_root %llu parent %llu",
+			  node->bytenr, node->ref_mod, node->action, ref_root,
+			  parent);
+		return -EIO;
+	}
 	if (node->action == BTRFS_ADD_DELAYED_REF && insert_reserved) {
 		BUG_ON(!extent_op || !extent_op->update_flags);
 		ret = alloc_reserved_tree_block(trans, root,
@@ -2407,7 +2407,8 @@ static int run_one_delayed_ref(struct btrfs_trans_handle *trans,
 		 */
 		BUG_ON(extent_op);
 		head = btrfs_delayed_node_to_head(node);
-		trace_run_delayed_ref_head(node, head, node->action);
+		trace_run_delayed_ref_head(root->fs_info, node, head,
+					   node->action);
 
 		if (insert_reserved) {
 			btrfs_pin_extent(root, node->bytenr,
@@ -2587,7 +2588,9 @@ static noinline int __btrfs_run_delayed_refs(struct btrfs_trans_handle *trans,
 					if (must_insert_reserved)
 						locked_ref->must_insert_reserved = 1;
 					locked_ref->processing = 0;
-					btrfs_debug(fs_info, "run_delayed_extent_op returned %d", ret);
+					btrfs_debug(fs_info,
+						    "run_delayed_extent_op returned %d",
+						    ret);
 					btrfs_delayed_ref_unlock(locked_ref);
 					return ret;
 				}
@@ -2647,7 +2650,8 @@ static noinline int __btrfs_run_delayed_refs(struct btrfs_trans_handle *trans,
 			locked_ref->processing = 0;
 			btrfs_delayed_ref_unlock(locked_ref);
 			btrfs_put_delayed_ref(ref);
-			btrfs_debug(fs_info, "run_one_delayed_ref returned %d", ret);
+			btrfs_debug(fs_info, "run_one_delayed_ref returned %d",
+				    ret);
 			return ret;
 		}
 
@@ -2762,7 +2766,7 @@ u64 btrfs_csum_bytes_to_leaves(struct btrfs_root *root, u64 csum_bytes)
 	u64 num_csums_per_leaf;
 	u64 num_csums;
 
-	csum_size = BTRFS_LEAF_DATA_SIZE(root) - sizeof(struct btrfs_item);
+	csum_size = BTRFS_MAX_ITEM_SIZE(root);
 	num_csums_per_leaf = div64_u64(csum_size,
 			(u64)btrfs_super_csum_size(root->fs_info->super_copy));
 	num_csums = div64_u64(csum_bytes, root->sectorsize);
@@ -2829,6 +2833,7 @@ int btrfs_should_throttle_delayed_refs(struct btrfs_trans_handle *trans,
 
 struct async_delayed_refs {
 	struct btrfs_root *root;
+	u64 transid;
 	int count;
 	int error;
 	int sync;
@@ -2844,6 +2849,10 @@ static void delayed_ref_async_start(struct btrfs_work *work)
 
 	async = container_of(work, struct async_delayed_refs, work);
 
+	/* if the commit is already started, we don't need to wait here */
+	if (btrfs_transaction_blocked(async->root->fs_info))
+		goto done;
+
 	trans = btrfs_join_transaction(async->root);
 	if (IS_ERR(trans)) {
 		async->error = PTR_ERR(trans);
@@ -2855,10 +2864,15 @@ static void delayed_ref_async_start(struct btrfs_work *work)
 	 * wait on delayed refs
 	 */
 	trans->sync = true;
+
+	/* Don't bother flushing if we got into a different transaction */
+	if (trans->transid > async->transid)
+		goto end;
+
 	ret = btrfs_run_delayed_refs(trans, async->root, async->count);
 	if (ret)
 		async->error = ret;
-
+end:
 	ret = btrfs_end_transaction(trans, async->root);
 	if (ret && !async->error)
 		async->error = ret;
@@ -2870,7 +2884,7 @@ done:
 }
 
 int btrfs_async_run_delayed_refs(struct btrfs_root *root,
-				 unsigned long count, int wait)
+				 unsigned long count, u64 transid, int wait)
 {
 	struct async_delayed_refs *async;
 	int ret;
@@ -2882,6 +2896,7 @@ int btrfs_async_run_delayed_refs(struct btrfs_root *root,
 	async->root = root->fs_info->tree_root;
 	async->count = count;
 	async->error = 0;
+	async->transid = transid;
 	if (wait)
 		async->sync = 1;
 	else
@@ -2926,7 +2941,7 @@ int btrfs_run_delayed_refs(struct btrfs_trans_handle *trans,
 	if (trans->aborted)
 		return 0;
 
-	if (root->fs_info->creating_free_space_tree)
+	if (test_bit(BTRFS_FS_CREATING_FREE_SPACE_TREE, &root->fs_info->flags))
 		return 0;
 
 	if (root == root->fs_info->extent_root)
@@ -2943,7 +2958,7 @@ again:
 	trans->can_flush_pending_bgs = false;
 	ret = __btrfs_run_delayed_refs(trans, root, count);
 	if (ret < 0) {
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 		return ret;
 	}
 
@@ -2957,7 +2972,6 @@ again:
 			spin_unlock(&delayed_refs->lock);
 			goto out;
 		}
-		count = (unsigned long)-1;
 
 		while (node) {
 			head = rb_entry(node, struct btrfs_delayed_ref_head,
@@ -3207,7 +3221,7 @@ static int __btrfs_mod_ref(struct btrfs_trans_handle *trans,
 			    u64, u64, u64, u64, u64, u64);
 
 
-	if (btrfs_test_is_dummy_root(root))
+	if (btrfs_is_testing(root->fs_info))
 		return 0;
 
 	ref_root = btrfs_header_owner(buf);
@@ -3402,7 +3416,7 @@ again:
 		 * transaction, this only happens in really bad situations
 		 * anyway.
 		 */
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 		goto out_put;
 	}
 	WARN_ON(ret);
@@ -3420,7 +3434,7 @@ again:
 
 	spin_lock(&block_group->lock);
 	if (block_group->cached != BTRFS_CACHE_FINISHED ||
-	    !btrfs_test_opt(root, SPACE_CACHE)) {
+	    !btrfs_test_opt(root->fs_info, SPACE_CACHE)) {
 		/*
 		 * don't bother trying to write stuff out _if_
 		 * a) we're not cached,
@@ -3473,7 +3487,6 @@ again:
 		dcs = BTRFS_DC_SETUP;
 	else if (ret == -ENOSPC)
 		set_bit(BTRFS_TRANS_CACHE_ENOSPC, &trans->transaction->flags);
-	btrfs_free_reserved_data_space(inode, 0, num_pages);
 
 out_put:
 	iput(inode);
@@ -3497,7 +3510,7 @@ int btrfs_setup_space_cache(struct btrfs_trans_handle *trans,
 	struct btrfs_path *path;
 
 	if (list_empty(&cur_trans->dirty_bgs) ||
-	    !btrfs_test_opt(root, SPACE_CACHE))
+	    !btrfs_test_opt(root->fs_info, SPACE_CACHE))
 		return 0;
 
 	path = btrfs_alloc_path();
@@ -3642,7 +3655,7 @@ again:
 				}
 				spin_unlock(&cur_trans->dirty_bgs_lock);
 			} else if (ret) {
-				btrfs_abort_transaction(trans, root, ret);
+				btrfs_abort_transaction(trans, ret);
 			}
 		}
 
@@ -3681,6 +3694,8 @@ again:
 			goto again;
 		}
 		spin_unlock(&cur_trans->dirty_bgs_lock);
+	} else if (ret < 0) {
+		btrfs_cleanup_dirty_bgs(cur_trans, root);
 	}
 
 	btrfs_free_path(path);
@@ -3788,7 +3803,7 @@ int btrfs_write_dirty_block_groups(struct btrfs_trans_handle *trans,
 							    cache);
 			}
 			if (ret)
-				btrfs_abort_transaction(trans, root, ret);
+				btrfs_abort_transaction(trans, ret);
 		}
 
 		/* if its not on the io list, we need to put the block group */
@@ -3896,6 +3911,7 @@ static const char *alloc_name(u64 flags)
 
 static int update_space_info(struct btrfs_fs_info *info, u64 flags,
 			     u64 total_bytes, u64 bytes_used,
+			     u64 bytes_readonly,
 			     struct btrfs_space_info **space_info)
 {
 	struct btrfs_space_info *found;
@@ -3916,8 +3932,11 @@ static int update_space_info(struct btrfs_fs_info *info, u64 flags,
 		found->disk_total += total_bytes * factor;
 		found->bytes_used += bytes_used;
 		found->disk_used += bytes_used * factor;
+		found->bytes_readonly += bytes_readonly;
 		if (total_bytes > 0)
 			found->full = 0;
+		space_info_add_new_bytes(info, found, total_bytes -
+					 bytes_used - bytes_readonly);
 		spin_unlock(&found->lock);
 		*space_info = found;
 		return 0;
@@ -3943,7 +3962,7 @@ static int update_space_info(struct btrfs_fs_info *info, u64 flags,
 	found->disk_used = bytes_used * factor;
 	found->bytes_pinned = 0;
 	found->bytes_reserved = 0;
-	found->bytes_readonly = 0;
+	found->bytes_readonly = bytes_readonly;
 	found->bytes_may_use = 0;
 	found->full = 0;
 	found->max_extent_size = 0;
@@ -3952,6 +3971,8 @@ static int update_space_info(struct btrfs_fs_info *info, u64 flags,
 	found->flush = 0;
 	init_waitqueue_head(&found->wait);
 	INIT_LIST_HEAD(&found->ro_bgs);
+	INIT_LIST_HEAD(&found->tickets);
+	INIT_LIST_HEAD(&found->priority_tickets);
 
 	ret = kobject_init_and_add(&found->kobj, &space_info_ktype,
 				    info->space_info_kobj, "%s",
@@ -4252,13 +4273,10 @@ int btrfs_check_data_free_space(struct inode *inode, u64 start, u64 len)
 	if (ret < 0)
 		return ret;
 
-	/*
-	 * Use new btrfs_qgroup_reserve_data to reserve precious data space
-	 *
-	 * TODO: Find a good method to avoid reserve data space for NOCOW
-	 * range, but don't impact performance on quota disable case.
-	 */
+	/* Use new btrfs_qgroup_reserve_data to reserve precious data space. */
 	ret = btrfs_qgroup_reserve_data(inode, start, len);
+	if (ret)
+		btrfs_free_reserved_data_space_noquota(inode, start, len);
 	return ret;
 }
 
@@ -4410,10 +4428,10 @@ void check_system_chunk(struct btrfs_trans_handle *trans,
 	thresh = btrfs_calc_trunc_metadata_size(root, num_devs) +
 		btrfs_calc_trans_metadata_size(root, 1);
 
-	if (left < thresh && btrfs_test_opt(root, ENOSPC_DEBUG)) {
+	if (left < thresh && btrfs_test_opt(root->fs_info, ENOSPC_DEBUG)) {
 		btrfs_info(root->fs_info, "left=%llu, need=%llu, flags=%llu",
 			left, thresh, type);
-		dump_space_info(info, 0, 0);
+		dump_space_info(root->fs_info, info, 0, 0);
 	}
 
 	if (left < thresh) {
@@ -4438,6 +4456,15 @@ void check_system_chunk(struct btrfs_trans_handle *trans,
 	}
 }
 
+/*
+ * If force is CHUNK_ALLOC_FORCE:
+ *    - return 1 if it successfully allocates a chunk,
+ *    - return errors including -ENOSPC otherwise.
+ * If force is NOT CHUNK_ALLOC_FORCE:
+ *    - return 0 if it doesn't need to allocate a new chunk,
+ *    - return 1 if it successfully allocates a chunk,
+ *    - return errors including -ENOSPC otherwise.
+ */
 static int do_chunk_alloc(struct btrfs_trans_handle *trans,
 			  struct btrfs_root *extent_root, u64 flags, int force)
 {
@@ -4453,7 +4480,7 @@ static int do_chunk_alloc(struct btrfs_trans_handle *trans,
 	space_info = __find_space_info(extent_root->fs_info, flags);
 	if (!space_info) {
 		ret = update_space_info(extent_root->fs_info, flags,
-					0, 0, &space_info);
+					0, 0, 0, &space_info);
 		BUG_ON(ret); /* -ENOMEM */
 	}
 	BUG_ON(!space_info); /* Logic error */
@@ -4555,7 +4582,7 @@ out:
 	 */
 	if (trans->can_flush_pending_bgs &&
 	    trans->chunk_bytes_reserved >= (u64)SZ_2M) {
-		btrfs_create_pending_block_groups(trans, trans->root);
+		btrfs_create_pending_block_groups(trans, extent_root);
 		btrfs_trans_release_chunk_metadata(trans);
 	}
 	return ret;
@@ -4565,12 +4592,19 @@ static int can_overcommit(struct btrfs_root *root,
 			  struct btrfs_space_info *space_info, u64 bytes,
 			  enum btrfs_reserve_flush_enum flush)
 {
-	struct btrfs_block_rsv *global_rsv = &root->fs_info->global_block_rsv;
-	u64 profile = btrfs_get_alloc_profile(root, 0);
+	struct btrfs_block_rsv *global_rsv;
+	u64 profile;
 	u64 space_size;
 	u64 avail;
 	u64 used;
 
+	/* Don't overcommit when in mixed mode. */
+	if (space_info->flags & BTRFS_BLOCK_GROUP_DATA)
+		return 0;
+
+	BUG_ON(root->fs_info == NULL);
+	global_rsv = &root->fs_info->global_block_rsv;
+	profile = btrfs_get_alloc_profile(root, 0);
 	used = space_info->bytes_used + space_info->bytes_reserved +
 		space_info->bytes_pinned + space_info->bytes_readonly;
 
@@ -4722,6 +4756,11 @@ skip_async:
 			spin_unlock(&space_info->lock);
 			break;
 		}
+		if (list_empty(&space_info->tickets) &&
+		    list_empty(&space_info->priority_tickets)) {
+			spin_unlock(&space_info->lock);
+			break;
+		}
 		spin_unlock(&space_info->lock);
 
 		loops++;
@@ -4790,13 +4829,11 @@ commit:
 	return btrfs_commit_transaction(trans, root);
 }
 
-enum flush_state {
-	FLUSH_DELAYED_ITEMS_NR	=	1,
-	FLUSH_DELAYED_ITEMS	=	2,
-	FLUSH_DELALLOC		=	3,
-	FLUSH_DELALLOC_WAIT	=	4,
-	ALLOC_CHUNK		=	5,
-	COMMIT_TRANS		=	6,
+struct reserve_ticket {
+	u64 bytes;
+	int error;
+	struct list_head list;
+	wait_queue_head_t wait;
 };
 
 static int flush_space(struct btrfs_root *root,
@@ -4838,7 +4875,7 @@ static int flush_space(struct btrfs_root *root,
 				     btrfs_get_alloc_profile(root, 0),
 				     CHUNK_ALLOC_NO_FORCE);
 		btrfs_end_transaction(trans, root);
-		if (ret == -ENOSPC)
+		if (ret > 0 || ret == -ENOSPC)
 			ret = 0;
 		break;
 	case COMMIT_TRANS:
@@ -4849,6 +4886,8 @@ static int flush_space(struct btrfs_root *root,
 		break;
 	}
 
+	trace_btrfs_flush_space(root->fs_info, space_info->flags, num_bytes,
+				orig_bytes, state, ret);
 	return ret;
 }
 
@@ -4856,17 +4895,22 @@ static inline u64
 btrfs_calc_reclaim_metadata_size(struct btrfs_root *root,
 				 struct btrfs_space_info *space_info)
 {
+	struct reserve_ticket *ticket;
 	u64 used;
 	u64 expected;
-	u64 to_reclaim;
+	u64 to_reclaim = 0;
+
+	list_for_each_entry(ticket, &space_info->tickets, list)
+		to_reclaim += ticket->bytes;
+	list_for_each_entry(ticket, &space_info->priority_tickets, list)
+		to_reclaim += ticket->bytes;
+	if (to_reclaim)
+		return to_reclaim;
 
 	to_reclaim = min_t(u64, num_online_cpus() * SZ_1M, SZ_16M);
-	spin_lock(&space_info->lock);
 	if (can_overcommit(root, space_info, to_reclaim,
-			   BTRFS_RESERVE_FLUSH_ALL)) {
-		to_reclaim = 0;
-		goto out;
-	}
+			   BTRFS_RESERVE_FLUSH_ALL))
+		return 0;
 
 	used = space_info->bytes_used + space_info->bytes_reserved +
 	       space_info->bytes_pinned + space_info->bytes_readonly +
@@ -4882,14 +4926,11 @@ btrfs_calc_reclaim_metadata_size(struct btrfs_root *root,
 		to_reclaim = 0;
 	to_reclaim = min(to_reclaim, space_info->bytes_may_use +
 				     space_info->bytes_reserved);
-out:
-	spin_unlock(&space_info->lock);
-
 	return to_reclaim;
 }
 
 static inline int need_do_async_reclaim(struct btrfs_space_info *space_info,
-					struct btrfs_fs_info *fs_info, u64 used)
+					struct btrfs_root *root, u64 used)
 {
 	u64 thresh = div_factor_fine(space_info->total_bytes, 98);
 
@@ -4897,73 +4938,176 @@ static inline int need_do_async_reclaim(struct btrfs_space_info *space_info,
 	if ((space_info->bytes_used + space_info->bytes_reserved) >= thresh)
 		return 0;
 
-	return (used >= thresh && !btrfs_fs_closing(fs_info) &&
-		!test_bit(BTRFS_FS_STATE_REMOUNTING, &fs_info->fs_state));
+	if (!btrfs_calc_reclaim_metadata_size(root, space_info))
+		return 0;
+
+	return (used >= thresh && !btrfs_fs_closing(root->fs_info) &&
+		!test_bit(BTRFS_FS_STATE_REMOUNTING,
+			  &root->fs_info->fs_state));
 }
 
-static int btrfs_need_do_async_reclaim(struct btrfs_space_info *space_info,
-				       struct btrfs_fs_info *fs_info,
-				       int flush_state)
+static void wake_all_tickets(struct list_head *head)
 {
-	u64 used;
+	struct reserve_ticket *ticket;
 
-	spin_lock(&space_info->lock);
-	/*
-	 * We run out of space and have not got any free space via flush_space,
-	 * so don't bother doing async reclaim.
-	 */
-	if (flush_state > COMMIT_TRANS && space_info->full) {
-		spin_unlock(&space_info->lock);
-		return 0;
-	}
-
-	used = space_info->bytes_used + space_info->bytes_reserved +
-	       space_info->bytes_pinned + space_info->bytes_readonly +
-	       space_info->bytes_may_use;
-	if (need_do_async_reclaim(space_info, fs_info, used)) {
-		spin_unlock(&space_info->lock);
-		return 1;
+	while (!list_empty(head)) {
+		ticket = list_first_entry(head, struct reserve_ticket, list);
+		list_del_init(&ticket->list);
+		ticket->error = -ENOSPC;
+		wake_up(&ticket->wait);
 	}
-	spin_unlock(&space_info->lock);
-
-	return 0;
 }
 
+/*
+ * This is for normal flushers, we can wait all goddamned day if we want to.  We
+ * will loop and continuously try to flush as long as we are making progress.
+ * We count progress as clearing off tickets each time we have to loop.
+ */
 static void btrfs_async_reclaim_metadata_space(struct work_struct *work)
 {
 	struct btrfs_fs_info *fs_info;
 	struct btrfs_space_info *space_info;
 	u64 to_reclaim;
 	int flush_state;
+	int commit_cycles = 0;
+	u64 last_tickets_id;
 
 	fs_info = container_of(work, struct btrfs_fs_info, async_reclaim_work);
 	space_info = __find_space_info(fs_info, BTRFS_BLOCK_GROUP_METADATA);
 
+	spin_lock(&space_info->lock);
 	to_reclaim = btrfs_calc_reclaim_metadata_size(fs_info->fs_root,
 						      space_info);
-	if (!to_reclaim)
+	if (!to_reclaim) {
+		space_info->flush = 0;
+		spin_unlock(&space_info->lock);
 		return;
+	}
+	last_tickets_id = space_info->tickets_id;
+	spin_unlock(&space_info->lock);
 
 	flush_state = FLUSH_DELAYED_ITEMS_NR;
 	do {
+		struct reserve_ticket *ticket;
+		int ret;
+
+		ret = flush_space(fs_info->fs_root, space_info, to_reclaim,
+			    to_reclaim, flush_state);
+		spin_lock(&space_info->lock);
+		if (list_empty(&space_info->tickets)) {
+			space_info->flush = 0;
+			spin_unlock(&space_info->lock);
+			return;
+		}
+		to_reclaim = btrfs_calc_reclaim_metadata_size(fs_info->fs_root,
+							      space_info);
+		ticket = list_first_entry(&space_info->tickets,
+					  struct reserve_ticket, list);
+		if (last_tickets_id == space_info->tickets_id) {
+			flush_state++;
+		} else {
+			last_tickets_id = space_info->tickets_id;
+			flush_state = FLUSH_DELAYED_ITEMS_NR;
+			if (commit_cycles)
+				commit_cycles--;
+		}
+
+		if (flush_state > COMMIT_TRANS) {
+			commit_cycles++;
+			if (commit_cycles > 2) {
+				wake_all_tickets(&space_info->tickets);
+				space_info->flush = 0;
+			} else {
+				flush_state = FLUSH_DELAYED_ITEMS_NR;
+			}
+		}
+		spin_unlock(&space_info->lock);
+	} while (flush_state <= COMMIT_TRANS);
+}
+
+void btrfs_init_async_reclaim_work(struct work_struct *work)
+{
+	INIT_WORK(work, btrfs_async_reclaim_metadata_space);
+}
+
+static void priority_reclaim_metadata_space(struct btrfs_fs_info *fs_info,
+					    struct btrfs_space_info *space_info,
+					    struct reserve_ticket *ticket)
+{
+	u64 to_reclaim;
+	int flush_state = FLUSH_DELAYED_ITEMS_NR;
+
+	spin_lock(&space_info->lock);
+	to_reclaim = btrfs_calc_reclaim_metadata_size(fs_info->fs_root,
+						      space_info);
+	if (!to_reclaim) {
+		spin_unlock(&space_info->lock);
+		return;
+	}
+	spin_unlock(&space_info->lock);
+
+	do {
 		flush_space(fs_info->fs_root, space_info, to_reclaim,
 			    to_reclaim, flush_state);
 		flush_state++;
-		if (!btrfs_need_do_async_reclaim(space_info, fs_info,
-						 flush_state))
+		spin_lock(&space_info->lock);
+		if (ticket->bytes == 0) {
+			spin_unlock(&space_info->lock);
 			return;
+		}
+		spin_unlock(&space_info->lock);
+
+		/*
+		 * Priority flushers can't wait on delalloc without
+		 * deadlocking.
+		 */
+		if (flush_state == FLUSH_DELALLOC ||
+		    flush_state == FLUSH_DELALLOC_WAIT)
+			flush_state = ALLOC_CHUNK;
 	} while (flush_state < COMMIT_TRANS);
 }
 
-void btrfs_init_async_reclaim_work(struct work_struct *work)
+static int wait_reserve_ticket(struct btrfs_fs_info *fs_info,
+			       struct btrfs_space_info *space_info,
+			       struct reserve_ticket *ticket, u64 orig_bytes)
+
 {
-	INIT_WORK(work, btrfs_async_reclaim_metadata_space);
+	DEFINE_WAIT(wait);
+	int ret = 0;
+
+	spin_lock(&space_info->lock);
+	while (ticket->bytes > 0 && ticket->error == 0) {
+		ret = prepare_to_wait_event(&ticket->wait, &wait, TASK_KILLABLE);
+		if (ret) {
+			ret = -EINTR;
+			break;
+		}
+		spin_unlock(&space_info->lock);
+
+		schedule();
+
+		finish_wait(&ticket->wait, &wait);
+		spin_lock(&space_info->lock);
+	}
+	if (!ret)
+		ret = ticket->error;
+	if (!list_empty(&ticket->list))
+		list_del_init(&ticket->list);
+	if (ticket->bytes && ticket->bytes < orig_bytes) {
+		u64 num_bytes = orig_bytes - ticket->bytes;
+		space_info->bytes_may_use -= num_bytes;
+		trace_btrfs_space_reservation(fs_info, "space_info",
+					      space_info->flags, num_bytes, 0);
+	}
+	spin_unlock(&space_info->lock);
+
+	return ret;
 }
 
 /**
  * reserve_metadata_bytes - try to reserve bytes from the block_rsv's space
  * @root - the root we're allocating for
- * @block_rsv - the block_rsv we're allocating for
+ * @space_info - the space info we want to allocate from
  * @orig_bytes - the number of bytes we want
  * @flush - whether or not we can flush to make our reservation
  *
@@ -4974,81 +5118,36 @@ void btrfs_init_async_reclaim_work(struct work_struct *work)
  * regain reservations will be made and this will fail if there is not enough
  * space already.
  */
-static int reserve_metadata_bytes(struct btrfs_root *root,
-				  struct btrfs_block_rsv *block_rsv,
-				  u64 orig_bytes,
-				  enum btrfs_reserve_flush_enum flush)
+static int __reserve_metadata_bytes(struct btrfs_root *root,
+				    struct btrfs_space_info *space_info,
+				    u64 orig_bytes,
+				    enum btrfs_reserve_flush_enum flush)
 {
-	struct btrfs_space_info *space_info = block_rsv->space_info;
+	struct reserve_ticket ticket;
 	u64 used;
-	u64 num_bytes = orig_bytes;
-	int flush_state = FLUSH_DELAYED_ITEMS_NR;
 	int ret = 0;
-	bool flushing = false;
 
-again:
-	ret = 0;
-	spin_lock(&space_info->lock);
-	/*
-	 * We only want to wait if somebody other than us is flushing and we
-	 * are actually allowed to flush all things.
-	 */
-	while (flush == BTRFS_RESERVE_FLUSH_ALL && !flushing &&
-	       space_info->flush) {
-		spin_unlock(&space_info->lock);
-		/*
-		 * If we have a trans handle we can't wait because the flusher
-		 * may have to commit the transaction, which would mean we would
-		 * deadlock since we are waiting for the flusher to finish, but
-		 * hold the current transaction open.
-		 */
-		if (current->journal_info)
-			return -EAGAIN;
-		ret = wait_event_killable(space_info->wait, !space_info->flush);
-		/* Must have been killed, return */
-		if (ret)
-			return -EINTR;
-
-		spin_lock(&space_info->lock);
-	}
+	ASSERT(orig_bytes);
+	ASSERT(!current->journal_info || flush != BTRFS_RESERVE_FLUSH_ALL);
 
+	spin_lock(&space_info->lock);
 	ret = -ENOSPC;
 	used = space_info->bytes_used + space_info->bytes_reserved +
 		space_info->bytes_pinned + space_info->bytes_readonly +
 		space_info->bytes_may_use;
 
 	/*
-	 * The idea here is that we've not already over-reserved the block group
-	 * then we can go ahead and save our reservation first and then start
-	 * flushing if we need to.  Otherwise if we've already overcommitted
-	 * lets start flushing stuff first and then come back and try to make
-	 * our reservation.
+	 * If we have enough space then hooray, make our reservation and carry
+	 * on.  If not see if we can overcommit, and if we can, hooray carry on.
+	 * If not things get more complicated.
 	 */
-	if (used <= space_info->total_bytes) {
-		if (used + orig_bytes <= space_info->total_bytes) {
-			space_info->bytes_may_use += orig_bytes;
-			trace_btrfs_space_reservation(root->fs_info,
-				"space_info", space_info->flags, orig_bytes, 1);
-			ret = 0;
-		} else {
-			/*
-			 * Ok set num_bytes to orig_bytes since we aren't
-			 * overocmmitted, this way we only try and reclaim what
-			 * we need.
-			 */
-			num_bytes = orig_bytes;
-		}
-	} else {
-		/*
-		 * Ok we're over committed, set num_bytes to the overcommitted
-		 * amount plus the amount of bytes that we need for this
-		 * reservation.
-		 */
-		num_bytes = used - space_info->total_bytes +
-			(orig_bytes * 2);
-	}
-
-	if (ret && can_overcommit(root, space_info, orig_bytes, flush)) {
+	if (used + orig_bytes <= space_info->total_bytes) {
+		space_info->bytes_may_use += orig_bytes;
+		trace_btrfs_space_reservation(root->fs_info, "space_info",
+					      space_info->flags, orig_bytes,
+					      1);
+		ret = 0;
+	} else if (can_overcommit(root, space_info, orig_bytes, flush)) {
 		space_info->bytes_may_use += orig_bytes;
 		trace_btrfs_space_reservation(root->fs_info, "space_info",
 					      space_info->flags, orig_bytes,
@@ -5057,16 +5156,31 @@ again:
 	}
 
 	/*
-	 * Couldn't make our reservation, save our place so while we're trying
-	 * to reclaim space we can actually use it instead of somebody else
-	 * stealing it from us.
+	 * If we couldn't make a reservation then setup our reservation ticket
+	 * and kick the async worker if it's not already running.
 	 *
-	 * We make the other tasks wait for the flush only when we can flush
-	 * all things.
+	 * If we are a priority flusher then we just need to add our ticket to
+	 * the list and we will do our own flushing further down.
 	 */
 	if (ret && flush != BTRFS_RESERVE_NO_FLUSH) {
-		flushing = true;
-		space_info->flush = 1;
+		ticket.bytes = orig_bytes;
+		ticket.error = 0;
+		init_waitqueue_head(&ticket.wait);
+		if (flush == BTRFS_RESERVE_FLUSH_ALL) {
+			list_add_tail(&ticket.list, &space_info->tickets);
+			if (!space_info->flush) {
+				space_info->flush = 1;
+				trace_btrfs_trigger_flush(root->fs_info,
+							  space_info->flags,
+							  orig_bytes, flush,
+							  "enospc");
+				queue_work(system_unbound_wq,
+					   &root->fs_info->async_reclaim_work);
+			}
+		} else {
+			list_add_tail(&ticket.list,
+				      &space_info->priority_tickets);
+		}
 	} else if (!ret && space_info->flags & BTRFS_BLOCK_GROUP_METADATA) {
 		used += orig_bytes;
 		/*
@@ -5074,40 +5188,68 @@ again:
 		 * which means we won't have fs_info->fs_root set, so don't do
 		 * the async reclaim as we will panic.
 		 */
-		if (!root->fs_info->log_root_recovering &&
-		    need_do_async_reclaim(space_info, root->fs_info, used) &&
-		    !work_busy(&root->fs_info->async_reclaim_work))
+		if (!test_bit(BTRFS_FS_LOG_RECOVERING, &root->fs_info->flags) &&
+		    need_do_async_reclaim(space_info, root, used) &&
+		    !work_busy(&root->fs_info->async_reclaim_work)) {
+			trace_btrfs_trigger_flush(root->fs_info,
+						  space_info->flags,
+						  orig_bytes, flush,
+						  "preempt");
 			queue_work(system_unbound_wq,
 				   &root->fs_info->async_reclaim_work);
+		}
 	}
 	spin_unlock(&space_info->lock);
-
 	if (!ret || flush == BTRFS_RESERVE_NO_FLUSH)
-		goto out;
+		return ret;
 
-	ret = flush_space(root, space_info, num_bytes, orig_bytes,
-			  flush_state);
-	flush_state++;
+	if (flush == BTRFS_RESERVE_FLUSH_ALL)
+		return wait_reserve_ticket(root->fs_info, space_info, &ticket,
+					   orig_bytes);
 
-	/*
-	 * If we are FLUSH_LIMIT, we can not flush delalloc, or the deadlock
-	 * would happen. So skip delalloc flush.
-	 */
-	if (flush == BTRFS_RESERVE_FLUSH_LIMIT &&
-	    (flush_state == FLUSH_DELALLOC ||
-	     flush_state == FLUSH_DELALLOC_WAIT))
-		flush_state = ALLOC_CHUNK;
+	ret = 0;
+	priority_reclaim_metadata_space(root->fs_info, space_info, &ticket);
+	spin_lock(&space_info->lock);
+	if (ticket.bytes) {
+		if (ticket.bytes < orig_bytes) {
+			u64 num_bytes = orig_bytes - ticket.bytes;
+			space_info->bytes_may_use -= num_bytes;
+			trace_btrfs_space_reservation(root->fs_info,
+					"space_info", space_info->flags,
+					num_bytes, 0);
 
-	if (!ret)
-		goto again;
-	else if (flush == BTRFS_RESERVE_FLUSH_LIMIT &&
-		 flush_state < COMMIT_TRANS)
-		goto again;
-	else if (flush == BTRFS_RESERVE_FLUSH_ALL &&
-		 flush_state <= COMMIT_TRANS)
-		goto again;
+		}
+		list_del_init(&ticket.list);
+		ret = -ENOSPC;
+	}
+	spin_unlock(&space_info->lock);
+	ASSERT(list_empty(&ticket.list));
+	return ret;
+}
 
-out:
+/**
+ * reserve_metadata_bytes - try to reserve bytes from the block_rsv's space
+ * @root - the root we're allocating for
+ * @block_rsv - the block_rsv we're allocating for
+ * @orig_bytes - the number of bytes we want
+ * @flush - whether or not we can flush to make our reservation
+ *
+ * This will reserve orgi_bytes number of bytes from the space info associated
+ * with the block_rsv.  If there is not enough space it will make an attempt to
+ * flush out space to make room.  It will do this by flushing delalloc if
+ * possible or committing the transaction.  If flush is 0 then no attempts to
+ * regain reservations will be made and this will fail if there is not enough
+ * space already.
+ */
+static int reserve_metadata_bytes(struct btrfs_root *root,
+				  struct btrfs_block_rsv *block_rsv,
+				  u64 orig_bytes,
+				  enum btrfs_reserve_flush_enum flush)
+{
+	int ret;
+
+	ret = __reserve_metadata_bytes(root, block_rsv->space_info, orig_bytes,
+				       flush);
 	if (ret == -ENOSPC &&
 	    unlikely(root->orphan_cleanup_state == ORPHAN_CLEANUP_STARTED)) {
 		struct btrfs_block_rsv *global_rsv =
@@ -5120,13 +5262,8 @@ out:
 	if (ret == -ENOSPC)
 		trace_btrfs_space_reservation(root->fs_info,
 					      "space_info:enospc",
-					      space_info->flags, orig_bytes, 1);
-	if (flushing) {
-		spin_lock(&space_info->lock);
-		space_info->flush = 0;
-		wake_up_all(&space_info->wait);
-		spin_unlock(&space_info->lock);
-	}
+					      block_rsv->space_info->flags,
+					      orig_bytes, 1);
 	return ret;
 }
 
@@ -5202,6 +5339,110 @@ int btrfs_cond_migrate_bytes(struct btrfs_fs_info *fs_info,
 	return 0;
 }
 
+/*
+ * This is for space we already have accounted in space_info->bytes_may_use, so
+ * basically when we're returning space from block_rsv's.
+ */
+static void space_info_add_old_bytes(struct btrfs_fs_info *fs_info,
+				     struct btrfs_space_info *space_info,
+				     u64 num_bytes)
+{
+	struct reserve_ticket *ticket;
+	struct list_head *head;
+	u64 used;
+	enum btrfs_reserve_flush_enum flush = BTRFS_RESERVE_NO_FLUSH;
+	bool check_overcommit = false;
+
+	spin_lock(&space_info->lock);
+	head = &space_info->priority_tickets;
+
+	/*
+	 * If we are over our limit then we need to check and see if we can
+	 * overcommit, and if we can't then we just need to free up our space
+	 * and not satisfy any requests.
+	 */
+	used = space_info->bytes_used + space_info->bytes_reserved +
+		space_info->bytes_pinned + space_info->bytes_readonly +
+		space_info->bytes_may_use;
+	if (used - num_bytes >= space_info->total_bytes)
+		check_overcommit = true;
+again:
+	while (!list_empty(head) && num_bytes) {
+		ticket = list_first_entry(head, struct reserve_ticket,
+					  list);
+		/*
+		 * We use 0 bytes because this space is already reserved, so
+		 * adding the ticket space would be a double count.
+		 */
+		if (check_overcommit &&
+		    !can_overcommit(fs_info->extent_root, space_info, 0,
+				    flush))
+			break;
+		if (num_bytes >= ticket->bytes) {
+			list_del_init(&ticket->list);
+			num_bytes -= ticket->bytes;
+			ticket->bytes = 0;
+			space_info->tickets_id++;
+			wake_up(&ticket->wait);
+		} else {
+			ticket->bytes -= num_bytes;
+			num_bytes = 0;
+		}
+	}
+
+	if (num_bytes && head == &space_info->priority_tickets) {
+		head = &space_info->tickets;
+		flush = BTRFS_RESERVE_FLUSH_ALL;
+		goto again;
+	}
+	space_info->bytes_may_use -= num_bytes;
+	trace_btrfs_space_reservation(fs_info, "space_info",
+				      space_info->flags, num_bytes, 0);
+	spin_unlock(&space_info->lock);
+}
+
+/*
+ * This is for newly allocated space that isn't accounted in
+ * space_info->bytes_may_use yet.  So if we allocate a chunk or unpin an extent
+ * we use this helper.
+ */
+static void space_info_add_new_bytes(struct btrfs_fs_info *fs_info,
+				     struct btrfs_space_info *space_info,
+				     u64 num_bytes)
+{
+	struct reserve_ticket *ticket;
+	struct list_head *head = &space_info->priority_tickets;
+
+again:
+	while (!list_empty(head) && num_bytes) {
+		ticket = list_first_entry(head, struct reserve_ticket,
+					  list);
+		if (num_bytes >= ticket->bytes) {
+			trace_btrfs_space_reservation(fs_info, "space_info",
+						      space_info->flags,
+						      ticket->bytes, 1);
+			list_del_init(&ticket->list);
+			num_bytes -= ticket->bytes;
+			space_info->bytes_may_use += ticket->bytes;
+			ticket->bytes = 0;
+			space_info->tickets_id++;
+			wake_up(&ticket->wait);
+		} else {
+			trace_btrfs_space_reservation(fs_info, "space_info",
+						      space_info->flags,
+						      num_bytes, 1);
+			space_info->bytes_may_use += num_bytes;
+			ticket->bytes -= num_bytes;
+			num_bytes = 0;
+		}
+	}
+
+	if (num_bytes && head == &space_info->priority_tickets) {
+		head = &space_info->tickets;
+		goto again;
+	}
+}
+
 static void block_rsv_release_bytes(struct btrfs_fs_info *fs_info,
 				    struct btrfs_block_rsv *block_rsv,
 				    struct btrfs_block_rsv *dest, u64 num_bytes)
@@ -5236,18 +5477,15 @@ static void block_rsv_release_bytes(struct btrfs_fs_info *fs_info,
 			}
 			spin_unlock(&dest->lock);
 		}
-		if (num_bytes) {
-			spin_lock(&space_info->lock);
-			space_info->bytes_may_use -= num_bytes;
-			trace_btrfs_space_reservation(fs_info, "space_info",
-					space_info->flags, num_bytes, 0);
-			spin_unlock(&space_info->lock);
-		}
+		if (num_bytes)
+			space_info_add_old_bytes(fs_info, space_info,
+						 num_bytes);
 	}
 }
 
-static int block_rsv_migrate_bytes(struct btrfs_block_rsv *src,
-				   struct btrfs_block_rsv *dst, u64 num_bytes)
+int btrfs_block_rsv_migrate(struct btrfs_block_rsv *src,
+			    struct btrfs_block_rsv *dst, u64 num_bytes,
+			    int update_size)
 {
 	int ret;
 
@@ -5255,7 +5493,7 @@ static int block_rsv_migrate_bytes(struct btrfs_block_rsv *src,
 	if (ret)
 		return ret;
 
-	block_rsv_add_bytes(dst, num_bytes, 1);
+	block_rsv_add_bytes(dst, num_bytes, update_size);
 	return 0;
 }
 
@@ -5362,13 +5600,6 @@ int btrfs_block_rsv_refill(struct btrfs_root *root,
 	return ret;
 }
 
-int btrfs_block_rsv_migrate(struct btrfs_block_rsv *src_rsv,
-			    struct btrfs_block_rsv *dst_rsv,
-			    u64 num_bytes)
-{
-	return block_rsv_migrate_bytes(src_rsv, dst_rsv, num_bytes);
-}
-
 void btrfs_block_rsv_release(struct btrfs_root *root,
 			     struct btrfs_block_rsv *block_rsv,
 			     u64 num_bytes)
@@ -5381,48 +5612,21 @@ void btrfs_block_rsv_release(struct btrfs_root *root,
 				num_bytes);
 }
 
-/*
- * helper to calculate size of global block reservation.
- * the desired value is sum of space used by extent tree,
- * checksum tree and root tree
- */
-static u64 calc_global_metadata_size(struct btrfs_fs_info *fs_info)
-{
-	struct btrfs_space_info *sinfo;
-	u64 num_bytes;
-	u64 meta_used;
-	u64 data_used;
-	int csum_size = btrfs_super_csum_size(fs_info->super_copy);
-
-	sinfo = __find_space_info(fs_info, BTRFS_BLOCK_GROUP_DATA);
-	spin_lock(&sinfo->lock);
-	data_used = sinfo->bytes_used;
-	spin_unlock(&sinfo->lock);
-
-	sinfo = __find_space_info(fs_info, BTRFS_BLOCK_GROUP_METADATA);
-	spin_lock(&sinfo->lock);
-	if (sinfo->flags & BTRFS_BLOCK_GROUP_DATA)
-		data_used = 0;
-	meta_used = sinfo->bytes_used;
-	spin_unlock(&sinfo->lock);
-
-	num_bytes = (data_used >> fs_info->sb->s_blocksize_bits) *
-		    csum_size * 2;
-	num_bytes += div_u64(data_used + meta_used, 50);
-
-	if (num_bytes * 3 > meta_used)
-		num_bytes = div_u64(meta_used, 3);
-
-	return ALIGN(num_bytes, fs_info->extent_root->nodesize << 10);
-}
-
 static void update_global_block_rsv(struct btrfs_fs_info *fs_info)
 {
 	struct btrfs_block_rsv *block_rsv = &fs_info->global_block_rsv;
 	struct btrfs_space_info *sinfo = block_rsv->space_info;
 	u64 num_bytes;
 
-	num_bytes = calc_global_metadata_size(fs_info);
+	/*
+	 * The global block rsv is based on the size of the extent tree, the
+	 * checksum tree and the root tree.  If the fs is empty we want to set
+	 * it to a minimal amount for safety.
+	 */
+	num_bytes = btrfs_root_used(&fs_info->extent_root->root_item) +
+		btrfs_root_used(&fs_info->csum_root->root_item) +
+		btrfs_root_used(&fs_info->tree_root->root_item);
+	num_bytes = max_t(u64, num_bytes, SZ_16M);
 
 	spin_lock(&sinfo->lock);
 	spin_lock(&block_rsv->lock);
@@ -5520,7 +5724,7 @@ void btrfs_trans_release_metadata(struct btrfs_trans_handle *trans,
  */
 void btrfs_trans_release_chunk_metadata(struct btrfs_trans_handle *trans)
 {
-	struct btrfs_fs_info *fs_info = trans->root->fs_info;
+	struct btrfs_fs_info *fs_info = trans->fs_info;
 
 	if (!trans->chunk_bytes_reserved)
 		return;
@@ -5537,7 +5741,13 @@ int btrfs_orphan_reserve_metadata(struct btrfs_trans_handle *trans,
 				  struct inode *inode)
 {
 	struct btrfs_root *root = BTRFS_I(inode)->root;
-	struct btrfs_block_rsv *src_rsv = get_block_rsv(trans, root);
+	/*
+	 * We always use trans->block_rsv here as we will have reserved space
+	 * for our orphan when starting the transaction, using get_block_rsv()
+	 * here will sometimes make us choose the wrong block rsv as we could be
+	 * doing a reloc inode for a non refcounted root.
+	 */
+	struct btrfs_block_rsv *src_rsv = trans->block_rsv;
 	struct btrfs_block_rsv *dst_rsv = root->orphan_block_rsv;
 
 	/*
@@ -5548,7 +5758,7 @@ int btrfs_orphan_reserve_metadata(struct btrfs_trans_handle *trans,
 	u64 num_bytes = btrfs_calc_trans_metadata_size(root, 1);
 	trace_btrfs_space_reservation(root->fs_info, "orphan",
 				      btrfs_ino(inode), num_bytes, 1);
-	return block_rsv_migrate_bytes(src_rsv, dst_rsv, num_bytes);
+	return btrfs_block_rsv_migrate(src_rsv, dst_rsv, num_bytes, 1);
 }
 
 void btrfs_orphan_release_metadata(struct inode *inode)
@@ -5584,7 +5794,7 @@ int btrfs_subvolume_reserve_metadata(struct btrfs_root *root,
 	int ret;
 	struct btrfs_block_rsv *global_rsv = &root->fs_info->global_block_rsv;
 
-	if (root->fs_info->quota_enabled) {
+	if (test_bit(BTRFS_FS_QUOTA_ENABLED, &root->fs_info->flags)) {
 		/* One for parent inode, two for dir entries */
 		num_bytes = 3 * root->nodesize;
 		ret = btrfs_qgroup_reserve_meta(root, num_bytes);
@@ -5603,7 +5813,7 @@ int btrfs_subvolume_reserve_metadata(struct btrfs_root *root,
 				  BTRFS_RESERVE_FLUSH_ALL);
 
 	if (ret == -ENOSPC && use_global_rsv)
-		ret = btrfs_block_rsv_migrate(global_rsv, rsv, num_bytes);
+		ret = btrfs_block_rsv_migrate(global_rsv, rsv, num_bytes, 1);
 
 	if (ret && *qgroup_reserved)
 		btrfs_qgroup_free_meta(root, *qgroup_reserved);
@@ -5713,21 +5923,26 @@ int btrfs_delalloc_reserve_metadata(struct inode *inode, u64 num_bytes)
 	u64 to_reserve = 0;
 	u64 csum_bytes;
 	unsigned nr_extents = 0;
-	int extra_reserve = 0;
 	enum btrfs_reserve_flush_enum flush = BTRFS_RESERVE_FLUSH_ALL;
 	int ret = 0;
 	bool delalloc_lock = true;
 	u64 to_free = 0;
 	unsigned dropped;
+	bool release_extra = false;
 
 	/* If we are a free space inode we need to not flush since we will be in
 	 * the middle of a transaction commit.  We also don't need the delalloc
 	 * mutex since we won't race with anybody.  We need this mostly to make
 	 * lockdep shut its filthy mouth.
+	 *
+	 * If we have a transaction open (can happen if we call truncate_block
+	 * from truncate), then we need FLUSH_LIMIT so we don't deadlock.
 	 */
 	if (btrfs_is_free_space_inode(inode)) {
 		flush = BTRFS_RESERVE_NO_FLUSH;
 		delalloc_lock = false;
+	} else if (current->journal_info) {
+		flush = BTRFS_RESERVE_FLUSH_LIMIT;
 	}
 
 	if (flush != BTRFS_RESERVE_NO_FLUSH &&
@@ -5744,46 +5959,37 @@ int btrfs_delalloc_reserve_metadata(struct inode *inode, u64 num_bytes)
 					 BTRFS_MAX_EXTENT_SIZE - 1,
 					 BTRFS_MAX_EXTENT_SIZE);
 	BTRFS_I(inode)->outstanding_extents += nr_extents;
-	nr_extents = 0;
 
+	nr_extents = 0;
 	if (BTRFS_I(inode)->outstanding_extents >
 	    BTRFS_I(inode)->reserved_extents)
-		nr_extents = BTRFS_I(inode)->outstanding_extents -
+		nr_extents += BTRFS_I(inode)->outstanding_extents -
 			BTRFS_I(inode)->reserved_extents;
 
-	/*
-	 * Add an item to reserve for updating the inode when we complete the
-	 * delalloc io.
-	 */
-	if (!test_bit(BTRFS_INODE_DELALLOC_META_RESERVED,
-		      &BTRFS_I(inode)->runtime_flags)) {
-		nr_extents++;
-		extra_reserve = 1;
-	}
-
-	to_reserve = btrfs_calc_trans_metadata_size(root, nr_extents);
+	/* We always want to reserve a slot for updating the inode. */
+	to_reserve = btrfs_calc_trans_metadata_size(root, nr_extents + 1);
 	to_reserve += calc_csum_metadata_size(inode, num_bytes, 1);
 	csum_bytes = BTRFS_I(inode)->csum_bytes;
 	spin_unlock(&BTRFS_I(inode)->lock);
 
-	if (root->fs_info->quota_enabled) {
+	if (test_bit(BTRFS_FS_QUOTA_ENABLED, &root->fs_info->flags)) {
 		ret = btrfs_qgroup_reserve_meta(root,
 				nr_extents * root->nodesize);
 		if (ret)
 			goto out_fail;
 	}
 
-	ret = reserve_metadata_bytes(root, block_rsv, to_reserve, flush);
+	ret = btrfs_block_rsv_add(root, block_rsv, to_reserve, flush);
 	if (unlikely(ret)) {
 		btrfs_qgroup_free_meta(root, nr_extents * root->nodesize);
 		goto out_fail;
 	}
 
 	spin_lock(&BTRFS_I(inode)->lock);
-	if (extra_reserve) {
-		set_bit(BTRFS_INODE_DELALLOC_META_RESERVED,
-			&BTRFS_I(inode)->runtime_flags);
-		nr_extents--;
+	if (test_and_set_bit(BTRFS_INODE_DELALLOC_META_RESERVED,
+			     &BTRFS_I(inode)->runtime_flags)) {
+		to_reserve -= btrfs_calc_trans_metadata_size(root, 1);
+		release_extra = true;
 	}
 	BTRFS_I(inode)->reserved_extents += nr_extents;
 	spin_unlock(&BTRFS_I(inode)->lock);
@@ -5794,8 +6000,10 @@ int btrfs_delalloc_reserve_metadata(struct inode *inode, u64 num_bytes)
 	if (to_reserve)
 		trace_btrfs_space_reservation(root->fs_info, "delalloc",
 					      btrfs_ino(inode), to_reserve, 1);
-	block_rsv_add_bytes(block_rsv, to_reserve, 1);
-
+	if (release_extra)
+		btrfs_block_rsv_release(root, block_rsv,
+					btrfs_calc_trans_metadata_size(root,
+								       1));
 	return 0;
 
 out_fail:
@@ -5887,7 +6095,7 @@ void btrfs_delalloc_release_metadata(struct inode *inode, u64 num_bytes)
 	if (dropped > 0)
 		to_free += btrfs_calc_trans_metadata_size(root, dropped);
 
-	if (btrfs_test_is_dummy_root(root))
+	if (btrfs_is_testing(root->fs_info))
 		return;
 
 	trace_btrfs_space_reservation(root->fs_info, "delalloc",
@@ -5904,8 +6112,6 @@ void btrfs_delalloc_release_metadata(struct inode *inode, u64 num_bytes)
  * @start: start range we are writing to
  * @len: how long the range we are writing to
  *
- * TODO: This function will finally replace old btrfs_delalloc_reserve_space()
- *
  * This will do the following things
  *
  * o reserve space in data space info for num bytes
@@ -6002,7 +6208,7 @@ static int update_block_group(struct btrfs_trans_handle *trans,
 		spin_lock(&cache->space_info->lock);
 		spin_lock(&cache->lock);
 
-		if (btrfs_test_opt(root, SPACE_CACHE) &&
+		if (btrfs_test_opt(root->fs_info, SPACE_CACHE) &&
 		    cache->disk_cache_state < BTRFS_DC_CLEAR)
 			cache->disk_cache_state = BTRFS_DC_CLEAR;
 
@@ -6027,6 +6233,9 @@ static int update_block_group(struct btrfs_trans_handle *trans,
 			spin_unlock(&cache->lock);
 			spin_unlock(&cache->space_info->lock);
 
+			trace_btrfs_space_reservation(root->fs_info, "pinned",
+						      cache->space_info->flags,
+						      num_bytes, 1);
 			set_extent_dirty(info->pinned_extents,
 					 bytenr, bytenr + num_bytes - 1,
 					 GFP_NOFS | __GFP_NOFAIL);
@@ -6101,10 +6310,10 @@ static int pin_down_extent(struct btrfs_root *root,
 	spin_unlock(&cache->lock);
 	spin_unlock(&cache->space_info->lock);
 
+	trace_btrfs_space_reservation(root->fs_info, "pinned",
+				      cache->space_info->flags, num_bytes, 1);
 	set_extent_dirty(root->fs_info->pinned_extents, bytenr,
 			 bytenr + num_bytes - 1, GFP_NOFS | __GFP_NOFAIL);
-	if (reserved)
-		trace_btrfs_reserved_extent_free(root, bytenr, num_bytes);
 	return 0;
 }
 
@@ -6280,19 +6489,15 @@ void btrfs_wait_block_group_reservations(struct btrfs_block_group_cache *bg)
 }
 
 /**
- * btrfs_update_reserved_bytes - update the block_group and space info counters
+ * btrfs_add_reserved_bytes - update the block_group and space info counters
  * @cache:	The cache we are manipulating
+ * @ram_bytes:  The number of bytes of file content, and will be same to
+ *              @num_bytes except for the compress path.
  * @num_bytes:	The number of bytes in question
- * @reserve:	One of the reservation enums
  * @delalloc:   The blocks are allocated for the delalloc write
  *
- * This is called by the allocator when it reserves space, or by somebody who is
- * freeing space that was never actually used on disk.  For example if you
- * reserve some space for a new leaf in transaction A and before transaction A
- * commits you free that leaf, you call this with reserve set to 0 in order to
- * clear the reservation.
- *
- * Metadata reservations should be called with RESERVE_ALLOC so we do the proper
+ * This is called by the allocator when it reserves space. Metadata
+ * reservations should be called with RESERVE_ALLOC so we do the proper
  * ENOSPC accounting.  For data we handle the reservation through clearing the
  * delalloc bits in the io_tree.  We have to do this since we could end up
  * allocating less disk space for the amount of data we have reserved in the
@@ -6302,44 +6507,63 @@ void btrfs_wait_block_group_reservations(struct btrfs_block_group_cache *bg)
  * make the reservation and return -EAGAIN, otherwise this function always
  * succeeds.
  */
-static int btrfs_update_reserved_bytes(struct btrfs_block_group_cache *cache,
-				       u64 num_bytes, int reserve, int delalloc)
+static int btrfs_add_reserved_bytes(struct btrfs_block_group_cache *cache,
+				    u64 ram_bytes, u64 num_bytes, int delalloc)
 {
 	struct btrfs_space_info *space_info = cache->space_info;
 	int ret = 0;
 
 	spin_lock(&space_info->lock);
 	spin_lock(&cache->lock);
-	if (reserve != RESERVE_FREE) {
-		if (cache->ro) {
-			ret = -EAGAIN;
-		} else {
-			cache->reserved += num_bytes;
-			space_info->bytes_reserved += num_bytes;
-			if (reserve == RESERVE_ALLOC) {
-				trace_btrfs_space_reservation(cache->fs_info,
-						"space_info", space_info->flags,
-						num_bytes, 0);
-				space_info->bytes_may_use -= num_bytes;
-			}
-
-			if (delalloc)
-				cache->delalloc_bytes += num_bytes;
-		}
+	if (cache->ro) {
+		ret = -EAGAIN;
 	} else {
-		if (cache->ro)
-			space_info->bytes_readonly += num_bytes;
-		cache->reserved -= num_bytes;
-		space_info->bytes_reserved -= num_bytes;
+		cache->reserved += num_bytes;
+		space_info->bytes_reserved += num_bytes;
 
+		trace_btrfs_space_reservation(cache->fs_info,
+				"space_info", space_info->flags,
+				ram_bytes, 0);
+		space_info->bytes_may_use -= ram_bytes;
 		if (delalloc)
-			cache->delalloc_bytes -= num_bytes;
+			cache->delalloc_bytes += num_bytes;
 	}
 	spin_unlock(&cache->lock);
 	spin_unlock(&space_info->lock);
 	return ret;
 }
 
+/**
+ * btrfs_free_reserved_bytes - update the block_group and space info counters
+ * @cache:      The cache we are manipulating
+ * @num_bytes:  The number of bytes in question
+ * @delalloc:   The blocks are allocated for the delalloc write
+ *
+ * This is called by somebody who is freeing space that was never actually used
+ * on disk.  For example if you reserve some space for a new leaf in transaction
+ * A and before transaction A commits you free that leaf, you call this with
+ * reserve set to 0 in order to clear the reservation.
+ */
+
+static int btrfs_free_reserved_bytes(struct btrfs_block_group_cache *cache,
+				     u64 num_bytes, int delalloc)
+{
+	struct btrfs_space_info *space_info = cache->space_info;
+	int ret = 0;
+
+	spin_lock(&space_info->lock);
+	spin_lock(&cache->lock);
+	if (cache->ro)
+		space_info->bytes_readonly += num_bytes;
+	cache->reserved -= num_bytes;
+	space_info->bytes_reserved -= num_bytes;
+
+	if (delalloc)
+		cache->delalloc_bytes -= num_bytes;
+	spin_unlock(&cache->lock);
+	spin_unlock(&space_info->lock);
+	return ret;
+}
 void btrfs_prepare_extent_commit(struct btrfs_trans_handle *trans,
 				struct btrfs_root *root)
 {
@@ -6381,7 +6605,7 @@ fetch_cluster_info(struct btrfs_root *root, struct btrfs_space_info *space_info,
 		   u64 *empty_cluster)
 {
 	struct btrfs_free_cluster *ret = NULL;
-	bool ssd = btrfs_test_opt(root, SSD);
+	bool ssd = btrfs_test_opt(root->fs_info, SSD);
 
 	*empty_cluster = 0;
 	if (btrfs_mixed_space_info(space_info))
@@ -6459,6 +6683,9 @@ static int unpin_extent_range(struct btrfs_root *root, u64 start, u64 end,
 		spin_lock(&cache->lock);
 		cache->pinned -= len;
 		space_info->bytes_pinned -= len;
+
+		trace_btrfs_space_reservation(fs_info, "pinned",
+					      space_info->flags, len, 0);
 		space_info->max_extent_size = 0;
 		percpu_counter_add(&space_info->total_bytes_pinned, -len);
 		if (cache->ro) {
@@ -6466,17 +6693,29 @@ static int unpin_extent_range(struct btrfs_root *root, u64 start, u64 end,
 			readonly = true;
 		}
 		spin_unlock(&cache->lock);
-		if (!readonly && global_rsv->space_info == space_info) {
+		if (!readonly && return_free_space &&
+		    global_rsv->space_info == space_info) {
+			u64 to_add = len;
+			WARN_ON(!return_free_space);
 			spin_lock(&global_rsv->lock);
 			if (!global_rsv->full) {
-				len = min(len, global_rsv->size -
-					  global_rsv->reserved);
-				global_rsv->reserved += len;
-				space_info->bytes_may_use += len;
+				to_add = min(len, global_rsv->size -
+					     global_rsv->reserved);
+				global_rsv->reserved += to_add;
+				space_info->bytes_may_use += to_add;
 				if (global_rsv->reserved >= global_rsv->size)
 					global_rsv->full = 1;
+				trace_btrfs_space_reservation(fs_info,
+							      "space_info",
+							      space_info->flags,
+							      to_add, 1);
+				len -= to_add;
 			}
 			spin_unlock(&global_rsv->lock);
+			/* Add to any tickets we may have */
+			if (len)
+				space_info_add_new_bytes(fs_info, space_info,
+							 len);
 		}
 		spin_unlock(&space_info->lock);
 	}
@@ -6511,7 +6750,7 @@ int btrfs_finish_extent_commit(struct btrfs_trans_handle *trans,
 			break;
 		}
 
-		if (btrfs_test_opt(root, DISCARD))
+		if (btrfs_test_opt(root->fs_info, DISCARD))
 			ret = btrfs_discard_extent(root, start,
 						   end + 1 - start, NULL);
 
@@ -6649,7 +6888,7 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
 						    NULL, refs_to_drop,
 						    is_data, &last_ref);
 			if (ret) {
-				btrfs_abort_transaction(trans, extent_root, ret);
+				btrfs_abort_transaction(trans, ret);
 				goto out;
 			}
 			btrfs_release_path(path);
@@ -6691,14 +6930,15 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
 			}
 
 			if (ret) {
-				btrfs_err(info, "umm, got %d back from search, was looking for %llu",
-					ret, bytenr);
+				btrfs_err(info,
+					  "umm, got %d back from search, was looking for %llu",
+					  ret, bytenr);
 				if (ret > 0)
 					btrfs_print_leaf(extent_root,
 							 path->nodes[0]);
 			}
 			if (ret < 0) {
-				btrfs_abort_transaction(trans, extent_root, ret);
+				btrfs_abort_transaction(trans, ret);
 				goto out;
 			}
 			extent_slot = path->slots[0];
@@ -6709,10 +6949,10 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
 			"unable to find ref byte nr %llu parent %llu root %llu  owner %llu offset %llu",
 			bytenr, parent, root_objectid, owner_objectid,
 			owner_offset);
-		btrfs_abort_transaction(trans, extent_root, ret);
+		btrfs_abort_transaction(trans, ret);
 		goto out;
 	} else {
-		btrfs_abort_transaction(trans, extent_root, ret);
+		btrfs_abort_transaction(trans, ret);
 		goto out;
 	}
 
@@ -6724,7 +6964,7 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
 		ret = convert_extent_item_v0(trans, extent_root, path,
 					     owner_objectid, 0);
 		if (ret < 0) {
-			btrfs_abort_transaction(trans, extent_root, ret);
+			btrfs_abort_transaction(trans, ret);
 			goto out;
 		}
 
@@ -6738,12 +6978,13 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
 		ret = btrfs_search_slot(trans, extent_root, &key, path,
 					-1, 1);
 		if (ret) {
-			btrfs_err(info, "umm, got %d back from search, was looking for %llu",
+			btrfs_err(info,
+				  "umm, got %d back from search, was looking for %llu",
 				ret, bytenr);
 			btrfs_print_leaf(extent_root, path->nodes[0]);
 		}
 		if (ret < 0) {
-			btrfs_abort_transaction(trans, extent_root, ret);
+			btrfs_abort_transaction(trans, ret);
 			goto out;
 		}
 
@@ -6765,10 +7006,11 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
 
 	refs = btrfs_extent_refs(leaf, ei);
 	if (refs < refs_to_drop) {
-		btrfs_err(info, "trying to drop %d refs but we only have %Lu "
-			  "for bytenr %Lu", refs_to_drop, refs, bytenr);
+		btrfs_err(info,
+			  "trying to drop %d refs but we only have %Lu for bytenr %Lu",
+			  refs_to_drop, refs, bytenr);
 		ret = -EINVAL;
-		btrfs_abort_transaction(trans, extent_root, ret);
+		btrfs_abort_transaction(trans, ret);
 		goto out;
 	}
 	refs -= refs_to_drop;
@@ -6791,7 +7033,7 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
 						    iref, refs_to_drop,
 						    is_data, &last_ref);
 			if (ret) {
-				btrfs_abort_transaction(trans, extent_root, ret);
+				btrfs_abort_transaction(trans, ret);
 				goto out;
 			}
 		}
@@ -6814,7 +7056,7 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
 		ret = btrfs_del_items(trans, extent_root, path, path->slots[0],
 				      num_to_del);
 		if (ret) {
-			btrfs_abort_transaction(trans, extent_root, ret);
+			btrfs_abort_transaction(trans, ret);
 			goto out;
 		}
 		btrfs_release_path(path);
@@ -6822,7 +7064,7 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
 		if (is_data) {
 			ret = btrfs_del_csums(trans, root, bytenr, num_bytes);
 			if (ret) {
-				btrfs_abort_transaction(trans, extent_root, ret);
+				btrfs_abort_transaction(trans, ret);
 				goto out;
 			}
 		}
@@ -6830,13 +7072,13 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
 		ret = add_to_free_space_tree(trans, root->fs_info, bytenr,
 					     num_bytes);
 		if (ret) {
-			btrfs_abort_transaction(trans, extent_root, ret);
+			btrfs_abort_transaction(trans, ret);
 			goto out;
 		}
 
 		ret = update_block_group(trans, root, bytenr, num_bytes, 0);
 		if (ret) {
-			btrfs_abort_transaction(trans, extent_root, ret);
+			btrfs_abort_transaction(trans, ret);
 			goto out;
 		}
 	}
@@ -6959,7 +7201,7 @@ void btrfs_free_tree_block(struct btrfs_trans_handle *trans,
 		WARN_ON(test_bit(EXTENT_BUFFER_DIRTY, &buf->bflags));
 
 		btrfs_add_free_space(cache, buf->start, buf->len);
-		btrfs_update_reserved_bytes(cache, buf->len, RESERVE_FREE, 0);
+		btrfs_free_reserved_bytes(cache, buf->len, 0);
 		btrfs_put_block_group(cache);
 		trace_btrfs_reserved_extent_free(root, buf->start, buf->len);
 		pin = 0;
@@ -6985,7 +7227,7 @@ int btrfs_free_extent(struct btrfs_trans_handle *trans, struct btrfs_root *root,
 	int ret;
 	struct btrfs_fs_info *fs_info = root->fs_info;
 
-	if (btrfs_test_is_dummy_root(root))
+	if (btrfs_is_testing(fs_info))
 		return 0;
 
 	add_pinned_bytes(root->fs_info, num_bytes, owner, root_objectid);
@@ -7184,9 +7426,9 @@ btrfs_release_block_group(struct btrfs_block_group_cache *cache,
  * the free space extent currently.
  */
 static noinline int find_free_extent(struct btrfs_root *orig_root,
-				     u64 num_bytes, u64 empty_size,
-				     u64 hint_byte, struct btrfs_key *ins,
-				     u64 flags, int delalloc)
+				u64 ram_bytes, u64 num_bytes, u64 empty_size,
+				u64 hint_byte, struct btrfs_key *ins,
+				u64 flags, int delalloc)
 {
 	int ret = 0;
 	struct btrfs_root *root = orig_root->fs_info->extent_root;
@@ -7198,8 +7440,6 @@ static noinline int find_free_extent(struct btrfs_root *orig_root,
 	struct btrfs_space_info *space_info;
 	int loop = 0;
 	int index = __get_raid_index(flags);
-	int alloc_type = (flags & BTRFS_BLOCK_GROUP_DATA) ?
-		RESERVE_ALLOC_NO_ACCOUNT : RESERVE_ALLOC;
 	bool failed_cluster_refill = false;
 	bool failed_alloc = false;
 	bool use_cluster = true;
@@ -7531,8 +7771,8 @@ checks:
 					     search_start - offset);
 		BUG_ON(offset > search_start);
 
-		ret = btrfs_update_reserved_bytes(block_group, num_bytes,
-						  alloc_type, delalloc);
+		ret = btrfs_add_reserved_bytes(block_group, ram_bytes,
+				num_bytes, delalloc);
 		if (ret == -EAGAIN) {
 			btrfs_add_free_space(block_group, offset, num_bytes);
 			goto loop;
@@ -7620,8 +7860,7 @@ loop:
 			 * can do more things.
 			 */
 			if (ret < 0 && ret != -ENOSPC)
-				btrfs_abort_transaction(trans,
-							root, ret);
+				btrfs_abort_transaction(trans, ret);
 			else
 				ret = 0;
 			if (!exist)
@@ -7665,23 +7904,24 @@ out:
 	return ret;
 }
 
-static void dump_space_info(struct btrfs_space_info *info, u64 bytes,
+static void dump_space_info(struct btrfs_fs_info *fs_info,
+			    struct btrfs_space_info *info, u64 bytes,
 			    int dump_block_groups)
 {
 	struct btrfs_block_group_cache *cache;
 	int index = 0;
 
 	spin_lock(&info->lock);
-	printk(KERN_INFO "BTRFS: space_info %llu has %llu free, is %sfull\n",
-	       info->flags,
-	       info->total_bytes - info->bytes_used - info->bytes_pinned -
-	       info->bytes_reserved - info->bytes_readonly,
-	       (info->full) ? "" : "not ");
-	printk(KERN_INFO "BTRFS: space_info total=%llu, used=%llu, pinned=%llu, "
-	       "reserved=%llu, may_use=%llu, readonly=%llu\n",
-	       info->total_bytes, info->bytes_used, info->bytes_pinned,
-	       info->bytes_reserved, info->bytes_may_use,
-	       info->bytes_readonly);
+	btrfs_info(fs_info, "space_info %llu has %llu free, is %sfull",
+		   info->flags,
+		   info->total_bytes - info->bytes_used - info->bytes_pinned -
+		   info->bytes_reserved - info->bytes_readonly -
+		   info->bytes_may_use, (info->full) ? "" : "not ");
+	btrfs_info(fs_info,
+		"space_info total=%llu, used=%llu, pinned=%llu, reserved=%llu, may_use=%llu, readonly=%llu",
+		info->total_bytes, info->bytes_used, info->bytes_pinned,
+		info->bytes_reserved, info->bytes_may_use,
+		info->bytes_readonly);
 	spin_unlock(&info->lock);
 
 	if (!dump_block_groups)
@@ -7691,12 +7931,11 @@ static void dump_space_info(struct btrfs_space_info *info, u64 bytes,
 again:
 	list_for_each_entry(cache, &info->block_groups[index], list) {
 		spin_lock(&cache->lock);
-		printk(KERN_INFO "BTRFS: "
-			   "block group %llu has %llu bytes, "
-			   "%llu used %llu pinned %llu reserved %s\n",
-		       cache->key.objectid, cache->key.offset,
-		       btrfs_block_group_used(&cache->item), cache->pinned,
-		       cache->reserved, cache->ro ? "[readonly]" : "");
+		btrfs_info(fs_info,
+			"block group %llu has %llu bytes, %llu used %llu pinned %llu reserved %s",
+			cache->key.objectid, cache->key.offset,
+			btrfs_block_group_used(&cache->item), cache->pinned,
+			cache->reserved, cache->ro ? "[readonly]" : "");
 		btrfs_dump_free_space(cache, bytes);
 		spin_unlock(&cache->lock);
 	}
@@ -7705,11 +7944,12 @@ again:
 	up_read(&info->groups_sem);
 }
 
-int btrfs_reserve_extent(struct btrfs_root *root,
+int btrfs_reserve_extent(struct btrfs_root *root, u64 ram_bytes,
 			 u64 num_bytes, u64 min_alloc_size,
 			 u64 empty_size, u64 hint_byte,
 			 struct btrfs_key *ins, int is_data, int delalloc)
 {
+	struct btrfs_fs_info *fs_info = root->fs_info;
 	bool final_tried = num_bytes == min_alloc_size;
 	u64 flags;
 	int ret;
@@ -7717,27 +7957,28 @@ int btrfs_reserve_extent(struct btrfs_root *root,
 	flags = btrfs_get_alloc_profile(root, is_data);
 again:
 	WARN_ON(num_bytes < root->sectorsize);
-	ret = find_free_extent(root, num_bytes, empty_size, hint_byte, ins,
-			       flags, delalloc);
+	ret = find_free_extent(root, ram_bytes, num_bytes, empty_size,
+			       hint_byte, ins, flags, delalloc);
 	if (!ret && !is_data) {
-		btrfs_dec_block_group_reservations(root->fs_info,
-						   ins->objectid);
+		btrfs_dec_block_group_reservations(fs_info, ins->objectid);
 	} else if (ret == -ENOSPC) {
 		if (!final_tried && ins->offset) {
 			num_bytes = min(num_bytes >> 1, ins->offset);
 			num_bytes = round_down(num_bytes, root->sectorsize);
 			num_bytes = max(num_bytes, min_alloc_size);
+			ram_bytes = num_bytes;
 			if (num_bytes == min_alloc_size)
 				final_tried = true;
 			goto again;
-		} else if (btrfs_test_opt(root, ENOSPC_DEBUG)) {
+		} else if (btrfs_test_opt(fs_info, ENOSPC_DEBUG)) {
 			struct btrfs_space_info *sinfo;
 
-			sinfo = __find_space_info(root->fs_info, flags);
-			btrfs_err(root->fs_info, "allocation failed flags %llu, wanted %llu",
-				flags, num_bytes);
+			sinfo = __find_space_info(fs_info, flags);
+			btrfs_err(root->fs_info,
+				  "allocation failed flags %llu, wanted %llu",
+				  flags, num_bytes);
 			if (sinfo)
-				dump_space_info(sinfo, num_bytes, 1);
+				dump_space_info(fs_info, sinfo, num_bytes, 1);
 		}
 	}
 
@@ -7761,16 +8002,14 @@ static int __btrfs_free_reserved_extent(struct btrfs_root *root,
 	if (pin)
 		pin_down_extent(root, cache, start, len, 1);
 	else {
-		if (btrfs_test_opt(root, DISCARD))
+		if (btrfs_test_opt(root->fs_info, DISCARD))
 			ret = btrfs_discard_extent(root, start, len, NULL);
 		btrfs_add_free_space(cache, start, len);
-		btrfs_update_reserved_bytes(cache, len, RESERVE_FREE, delalloc);
+		btrfs_free_reserved_bytes(cache, len, delalloc);
+		trace_btrfs_reserved_extent_free(root, start, len);
 	}
 
 	btrfs_put_block_group(cache);
-
-	trace_btrfs_reserved_extent_free(root, start, len);
-
 	return ret;
 }
 
@@ -7979,6 +8218,7 @@ int btrfs_alloc_logged_file_extent(struct btrfs_trans_handle *trans,
 {
 	int ret;
 	struct btrfs_block_group_cache *block_group;
+	struct btrfs_space_info *space_info;
 
 	/*
 	 * Mixed block groups will exclude before processing the log so we only
@@ -7994,9 +8234,14 @@ int btrfs_alloc_logged_file_extent(struct btrfs_trans_handle *trans,
 	if (!block_group)
 		return -EINVAL;
 
-	ret = btrfs_update_reserved_bytes(block_group, ins->offset,
-					  RESERVE_ALLOC_NO_ACCOUNT, 0);
-	BUG_ON(ret); /* logic error */
+	space_info = block_group->space_info;
+	spin_lock(&space_info->lock);
+	spin_lock(&block_group->lock);
+	space_info->bytes_reserved += ins->offset;
+	block_group->reserved += ins->offset;
+	spin_unlock(&block_group->lock);
+	spin_unlock(&space_info->lock);
+
 	ret = alloc_reserved_file_extent(trans, root, 0, root_objectid,
 					 0, owner, offset, ins, 1);
 	btrfs_put_block_group(block_group);
@@ -8010,8 +8255,9 @@ btrfs_init_new_buffer(struct btrfs_trans_handle *trans, struct btrfs_root *root,
 	struct extent_buffer *buf;
 
 	buf = btrfs_find_create_tree_block(root, bytenr);
-	if (!buf)
-		return ERR_PTR(-ENOMEM);
+	if (IS_ERR(buf))
+		return buf;
+
 	btrfs_set_header_generation(buf, trans->transid);
 	btrfs_set_buffer_lockdep_class(root->root_key.objectid, buf, level);
 	btrfs_tree_lock(buf);
@@ -8038,7 +8284,7 @@ btrfs_init_new_buffer(struct btrfs_trans_handle *trans, struct btrfs_root *root,
 		set_extent_dirty(&trans->transaction->dirty_pages, buf->start,
 			 buf->start + buf->len - 1, GFP_NOFS);
 	}
-	trans->blocks_used++;
+	trans->dirty = true;
 	/* this returns a buffer locked for blocking */
 	return buf;
 }
@@ -8070,7 +8316,7 @@ again:
 		goto again;
 	}
 
-	if (btrfs_test_opt(root, ENOSPC_DEBUG)) {
+	if (btrfs_test_opt(root->fs_info, ENOSPC_DEBUG)) {
 		static DEFINE_RATELIMIT_STATE(_rs,
 				DEFAULT_RATELIMIT_INTERVAL * 10,
 				/*DEFAULT_RATELIMIT_BURST*/ 1);
@@ -8124,19 +8370,21 @@ struct extent_buffer *btrfs_alloc_tree_block(struct btrfs_trans_handle *trans,
 	bool skinny_metadata = btrfs_fs_incompat(root->fs_info,
 						 SKINNY_METADATA);
 
-	if (btrfs_test_is_dummy_root(root)) {
+#ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
+	if (btrfs_is_testing(root->fs_info)) {
 		buf = btrfs_init_new_buffer(trans, root, root->alloc_bytenr,
 					    level);
 		if (!IS_ERR(buf))
 			root->alloc_bytenr += blocksize;
 		return buf;
 	}
+#endif
 
 	block_rsv = use_block_rsv(trans, root, blocksize);
 	if (IS_ERR(block_rsv))
 		return ERR_CAST(block_rsv);
 
-	ret = btrfs_reserve_extent(root, blocksize, blocksize,
+	ret = btrfs_reserve_extent(root, blocksize, blocksize, blocksize,
 				   empty_size, hint, &ins, 0, 0);
 	if (ret)
 		goto out_unuse;
@@ -8218,7 +8466,6 @@ static noinline void reada_walk_down(struct btrfs_trans_handle *trans,
 	u64 refs;
 	u64 flags;
 	u32 nritems;
-	u32 blocksize;
 	struct btrfs_key key;
 	struct extent_buffer *eb;
 	int ret;
@@ -8236,7 +8483,6 @@ static noinline void reada_walk_down(struct btrfs_trans_handle *trans,
 
 	eb = path->nodes[wc->level];
 	nritems = btrfs_header_nritems(eb);
-	blocksize = root->nodesize;
 
 	for (slot = path->slots[wc->level]; slot < nritems; slot++) {
 		if (nread >= wc->reada_count)
@@ -8289,34 +8535,6 @@ reada:
 	wc->reada_slot = slot;
 }
 
-/*
- * These may not be seen by the usual inc/dec ref code so we have to
- * add them here.
- */
-static int record_one_subtree_extent(struct btrfs_trans_handle *trans,
-				     struct btrfs_root *root, u64 bytenr,
-				     u64 num_bytes)
-{
-	struct btrfs_qgroup_extent_record *qrecord;
-	struct btrfs_delayed_ref_root *delayed_refs;
-
-	qrecord = kmalloc(sizeof(*qrecord), GFP_NOFS);
-	if (!qrecord)
-		return -ENOMEM;
-
-	qrecord->bytenr = bytenr;
-	qrecord->num_bytes = num_bytes;
-	qrecord->old_roots = NULL;
-
-	delayed_refs = &trans->transaction->delayed_refs;
-	spin_lock(&delayed_refs->lock);
-	if (btrfs_qgroup_insert_dirty_extent(delayed_refs, qrecord))
-		kfree(qrecord);
-	spin_unlock(&delayed_refs->lock);
-
-	return 0;
-}
-
 static int account_leaf_items(struct btrfs_trans_handle *trans,
 			      struct btrfs_root *root,
 			      struct extent_buffer *eb)
@@ -8328,7 +8546,7 @@ static int account_leaf_items(struct btrfs_trans_handle *trans,
 	u64 bytenr, num_bytes;
 
 	/* We can be called directly from walk_up_proc() */
-	if (!root->fs_info->quota_enabled)
+	if (!test_bit(BTRFS_FS_QUOTA_ENABLED, &root->fs_info->flags))
 		return 0;
 
 	for (i = 0; i < nr; i++) {
@@ -8350,7 +8568,8 @@ static int account_leaf_items(struct btrfs_trans_handle *trans,
 
 		num_bytes = btrfs_file_extent_disk_num_bytes(eb, fi);
 
-		ret = record_one_subtree_extent(trans, root, bytenr, num_bytes);
+		ret = btrfs_qgroup_insert_dirty_extent(trans, root->fs_info,
+				bytenr, num_bytes, GFP_NOFS);
 		if (ret)
 			return ret;
 	}
@@ -8436,7 +8655,7 @@ static int account_shared_subtree(struct btrfs_trans_handle *trans,
 	BUG_ON(root_level < 0 || root_level > BTRFS_MAX_LEVEL);
 	BUG_ON(root_eb == NULL);
 
-	if (!root->fs_info->quota_enabled)
+	if (!test_bit(BTRFS_FS_QUOTA_ENABLED, &root->fs_info->flags))
 		return 0;
 
 	if (!extent_buffer_uptodate(root_eb)) {
@@ -8499,8 +8718,9 @@ walk_down:
 			btrfs_set_lock_blocking_rw(eb, BTRFS_READ_LOCK);
 			path->locks[level] = BTRFS_READ_LOCK_BLOCKING;
 
-			ret = record_one_subtree_extent(trans, root, child_bytenr,
-							root->nodesize);
+			ret = btrfs_qgroup_insert_dirty_extent(trans,
+					root->fs_info, child_bytenr,
+					root->nodesize, GFP_NOFS);
 			if (ret)
 				goto out;
 		}
@@ -8653,8 +8873,9 @@ static noinline int do_walk_down(struct btrfs_trans_handle *trans,
 	next = btrfs_find_tree_block(root->fs_info, bytenr);
 	if (!next) {
 		next = btrfs_find_create_tree_block(root, bytenr);
-		if (!next)
-			return -ENOMEM;
+		if (IS_ERR(next))
+			return PTR_ERR(next);
+
 		btrfs_set_buffer_lockdep_class(root->root_key.objectid, next,
 					       level - 1);
 		reada = 1;
@@ -8665,14 +8886,13 @@ static noinline int do_walk_down(struct btrfs_trans_handle *trans,
 	ret = btrfs_lookup_extent_info(trans, root, bytenr, level - 1, 1,
 				       &wc->refs[level - 1],
 				       &wc->flags[level - 1]);
-	if (ret < 0) {
-		btrfs_tree_unlock(next);
-		return ret;
-	}
+	if (ret < 0)
+		goto out_unlock;
 
 	if (unlikely(wc->refs[level - 1] == 0)) {
 		btrfs_err(root->fs_info, "Missing references.");
-		BUG();
+		ret = -EIO;
+		goto out_unlock;
 	}
 	*lookup_info = 0;
 
@@ -8724,7 +8944,12 @@ static noinline int do_walk_down(struct btrfs_trans_handle *trans,
 	}
 
 	level--;
-	BUG_ON(level != btrfs_header_level(next));
+	ASSERT(level == btrfs_header_level(next));
+	if (level != btrfs_header_level(next)) {
+		btrfs_err(root->fs_info, "mismatched level");
+		ret = -EIO;
+		goto out_unlock;
+	}
 	path->nodes[level] = next;
 	path->slots[level] = 0;
 	path->locks[level] = BTRFS_WRITE_LOCK_BLOCKING;
@@ -8739,8 +8964,15 @@ skip:
 		if (wc->flags[level] & BTRFS_BLOCK_FLAG_FULL_BACKREF) {
 			parent = path->nodes[level]->start;
 		} else {
-			BUG_ON(root->root_key.objectid !=
+			ASSERT(root->root_key.objectid ==
 			       btrfs_header_owner(path->nodes[level]));
+			if (root->root_key.objectid !=
+			    btrfs_header_owner(path->nodes[level])) {
+				btrfs_err(root->fs_info,
+						"mismatched block owner");
+				ret = -EIO;
+				goto out_unlock;
+			}
 			parent = 0;
 		}
 
@@ -8749,20 +8981,24 @@ skip:
 						     generation, level - 1);
 			if (ret) {
 				btrfs_err_rl(root->fs_info,
-					"Error "
-					"%d accounting shared subtree. Quota "
-					"is out of sync, rescan required.",
-					ret);
+					     "Error %d accounting shared subtree. Quota is out of sync, rescan required.",
+					     ret);
 			}
 		}
 		ret = btrfs_free_extent(trans, root, bytenr, blocksize, parent,
 				root->root_key.objectid, level - 1, 0);
-		BUG_ON(ret); /* -ENOMEM */
+		if (ret)
+			goto out_unlock;
 	}
+
+	*lookup_info = 1;
+	ret = 1;
+
+out_unlock:
 	btrfs_tree_unlock(next);
 	free_extent_buffer(next);
-	*lookup_info = 1;
-	return 1;
+
+	return ret;
 }
 
 /*
@@ -8842,10 +9078,8 @@ static noinline int walk_up_proc(struct btrfs_trans_handle *trans,
 			ret = account_leaf_items(trans, root, eb);
 			if (ret) {
 				btrfs_err_rl(root->fs_info,
-					"error "
-					"%d accounting leaf items. Quota "
-					"is out of sync, rescan required.",
-					ret);
+					     "error %d accounting leaf items. Quota is out of sync, rescan required.",
+					     ret);
 			}
 		}
 		/* make block locked assertion in clean_tree_block happy */
@@ -8961,9 +9195,10 @@ int btrfs_drop_snapshot(struct btrfs_root *root,
 			 struct btrfs_block_rsv *block_rsv, int update_ref,
 			 int for_reloc)
 {
+	struct btrfs_fs_info *fs_info = root->fs_info;
 	struct btrfs_path *path;
 	struct btrfs_trans_handle *trans;
-	struct btrfs_root *tree_root = root->fs_info->tree_root;
+	struct btrfs_root *tree_root = fs_info->tree_root;
 	struct btrfs_root_item *root_item = &root->root_item;
 	struct walk_control *wc;
 	struct btrfs_key key;
@@ -8972,7 +9207,7 @@ int btrfs_drop_snapshot(struct btrfs_root *root,
 	int level;
 	bool root_dropped = false;
 
-	btrfs_debug(root->fs_info, "Drop subvolume %llu", root->objectid);
+	btrfs_debug(fs_info, "Drop subvolume %llu", root->objectid);
 
 	path = btrfs_alloc_path();
 	if (!path) {
@@ -9094,14 +9329,15 @@ int btrfs_drop_snapshot(struct btrfs_root *root,
 						&root->root_key,
 						root_item);
 			if (ret) {
-				btrfs_abort_transaction(trans, tree_root, ret);
+				btrfs_abort_transaction(trans, ret);
 				err = ret;
 				goto out_end_trans;
 			}
 
 			btrfs_end_transaction_throttle(trans, tree_root);
 			if (!for_reloc && btrfs_need_cleaner_sleep(root)) {
-				pr_debug("BTRFS: drop snapshot early exit\n");
+				btrfs_debug(fs_info,
+					    "drop snapshot early exit");
 				err = -EAGAIN;
 				goto out_free;
 			}
@@ -9121,7 +9357,7 @@ int btrfs_drop_snapshot(struct btrfs_root *root,
 
 	ret = btrfs_del_root(trans, tree_root, &root->root_key);
 	if (ret) {
-		btrfs_abort_transaction(trans, tree_root, ret);
+		btrfs_abort_transaction(trans, ret);
 		goto out_end_trans;
 	}
 
@@ -9129,7 +9365,7 @@ int btrfs_drop_snapshot(struct btrfs_root *root,
 		ret = btrfs_find_root(tree_root, &root->root_key, path,
 				      NULL, NULL);
 		if (ret < 0) {
-			btrfs_abort_transaction(trans, tree_root, ret);
+			btrfs_abort_transaction(trans, ret);
 			err = ret;
 			goto out_end_trans;
 		} else if (ret > 0) {
@@ -9167,7 +9403,7 @@ out:
 	if (!for_reloc && root_dropped == false)
 		btrfs_add_dead_root(root);
 	if (err && err != -EAGAIN)
-		btrfs_handle_fs_error(root->fs_info, err, NULL);
+		btrfs_handle_fs_error(fs_info, err, NULL);
 	return err;
 }
 
@@ -9500,7 +9736,7 @@ int btrfs_can_relocate(struct btrfs_root *root, u64 bytenr)
 	int full = 0;
 	int ret = 0;
 
-	debug = btrfs_test_opt(root, ENOSPC_DEBUG);
+	debug = btrfs_test_opt(root->fs_info, ENOSPC_DEBUG);
 
 	block_group = btrfs_lookup_block_group(root->fs_info, bytenr);
 
@@ -9656,7 +9892,23 @@ static int find_first_block_group(struct btrfs_root *root,
 
 		if (found_key.objectid >= key->objectid &&
 		    found_key.type == BTRFS_BLOCK_GROUP_ITEM_KEY) {
-			ret = 0;
+			struct extent_map_tree *em_tree;
+			struct extent_map *em;
+
+			em_tree = &root->fs_info->mapping_tree.map_tree;
+			read_lock(&em_tree->lock);
+			em = lookup_extent_mapping(em_tree, found_key.objectid,
+						   found_key.offset);
+			read_unlock(&em_tree->lock);
+			if (!em) {
+				btrfs_err(root->fs_info,
+			"logical %llu len %llu found bg but no related chunk",
+					  found_key.objectid, found_key.offset);
+				ret = -ENOENT;
+			} else {
+				ret = 0;
+			}
+			free_extent_map(em);
 			goto out;
 		}
 		path->slots[0]++;
@@ -9693,6 +9945,7 @@ void btrfs_put_block_group_cache(struct btrfs_fs_info *info)
 		block_group->iref = 0;
 		block_group->inode = NULL;
 		spin_unlock(&block_group->lock);
+		ASSERT(block_group->io_ctl.inode == NULL);
 		iput(inode);
 		last = block_group->key.objectid + block_group->key.offset;
 		btrfs_put_block_group(block_group);
@@ -9750,6 +10003,10 @@ int btrfs_free_block_groups(struct btrfs_fs_info *info)
 			free_excluded_extents(info->extent_root, block_group);
 
 		btrfs_remove_free_space_cache(block_group);
+		ASSERT(list_empty(&block_group->dirty_list));
+		ASSERT(list_empty(&block_group->io_list));
+		ASSERT(list_empty(&block_group->bg_list));
+		ASSERT(atomic_read(&block_group->count) == 1);
 		btrfs_put_block_group(block_group);
 
 		spin_lock(&info->block_group_cache_lock);
@@ -9772,13 +10029,15 @@ int btrfs_free_block_groups(struct btrfs_fs_info *info)
 		space_info = list_entry(info->space_info.next,
 					struct btrfs_space_info,
 					list);
-		if (btrfs_test_opt(info->tree_root, ENOSPC_DEBUG)) {
-			if (WARN_ON(space_info->bytes_pinned > 0 ||
+
+		/*
+		 * Do not hide this behind enospc_debug, this is actually
+		 * important and indicates a real bug if this happens.
+		 */
+		if (WARN_ON(space_info->bytes_pinned > 0 ||
 			    space_info->bytes_reserved > 0 ||
-			    space_info->bytes_may_use > 0)) {
-				dump_space_info(space_info, 0, 0);
-			}
-		}
+			    space_info->bytes_may_use > 0))
+			dump_space_info(info, space_info, 0, 0);
 		list_del(&space_info->list);
 		for (i = 0; i < BTRFS_NR_RAID_TYPES; i++) {
 			struct kobject *kobj;
@@ -9827,7 +10086,8 @@ static void __link_block_group(struct btrfs_space_info *space_info,
 
 	return;
 out_err:
-	pr_warn("BTRFS: failed to add kobject for block cache. ignoring.\n");
+	btrfs_warn(cache->fs_info,
+		   "failed to add kobject for block cache, ignoring");
 }
 
 static struct btrfs_block_group_cache *
@@ -9885,6 +10145,11 @@ int btrfs_read_block_groups(struct btrfs_root *root)
 	struct extent_buffer *leaf;
 	int need_clear = 0;
 	u64 cache_gen;
+	u64 feature;
+	int mixed;
+
+	feature = btrfs_super_incompat_flags(info->super_copy);
+	mixed = !!(feature & BTRFS_FEATURE_INCOMPAT_MIXED_GROUPS);
 
 	root = info->extent_root;
 	key.objectid = 0;
@@ -9896,10 +10161,10 @@ int btrfs_read_block_groups(struct btrfs_root *root)
 	path->reada = READA_FORWARD;
 
 	cache_gen = btrfs_super_cache_generation(root->fs_info->super_copy);
-	if (btrfs_test_opt(root, SPACE_CACHE) &&
+	if (btrfs_test_opt(root->fs_info, SPACE_CACHE) &&
 	    btrfs_super_generation(root->fs_info->super_copy) != cache_gen)
 		need_clear = 1;
-	if (btrfs_test_opt(root, CLEAR_CACHE))
+	if (btrfs_test_opt(root->fs_info, CLEAR_CACHE))
 		need_clear = 1;
 
 	while (1) {
@@ -9930,7 +10195,7 @@ int btrfs_read_block_groups(struct btrfs_root *root)
 			 * b) Setting 'dirty flag' makes sure that we flush
 			 *    the new space cache info onto disk.
 			 */
-			if (btrfs_test_opt(root, SPACE_CACHE))
+			if (btrfs_test_opt(root->fs_info, SPACE_CACHE))
 				cache->disk_cache_state = BTRFS_DC_CLEAR;
 		}
 
@@ -9938,6 +10203,15 @@ int btrfs_read_block_groups(struct btrfs_root *root)
 				   btrfs_item_ptr_offset(leaf, path->slots[0]),
 				   sizeof(cache->item));
 		cache->flags = btrfs_block_group_flags(&cache->item);
+		if (!mixed &&
+		    ((cache->flags & BTRFS_BLOCK_GROUP_METADATA) &&
+		    (cache->flags & BTRFS_BLOCK_GROUP_DATA))) {
+			btrfs_err(info,
+"bg %llu is a mixed block group but filesystem hasn't enabled mixed block groups",
+				  cache->key.objectid);
+			ret = -EINVAL;
+			goto error;
+		}
 
 		key.objectid = found_key.objectid + found_key.offset;
 		btrfs_release_path(path);
@@ -9986,9 +10260,10 @@ int btrfs_read_block_groups(struct btrfs_root *root)
 			goto error;
 		}
 
+		trace_btrfs_add_block_group(root->fs_info, cache, 0);
 		ret = update_space_info(info, cache->flags, found_key.offset,
 					btrfs_block_group_used(&cache->item),
-					&space_info);
+					cache->bytes_super, &space_info);
 		if (ret) {
 			btrfs_remove_free_space_cache(cache);
 			spin_lock(&info->block_group_cache_lock);
@@ -10001,9 +10276,6 @@ int btrfs_read_block_groups(struct btrfs_root *root)
 		}
 
 		cache->space_info = space_info;
-		spin_lock(&cache->space_info->lock);
-		cache->space_info->bytes_readonly += cache->bytes_super;
-		spin_unlock(&cache->space_info->lock);
 
 		__link_block_group(space_info, cache);
 
@@ -10074,11 +10346,11 @@ void btrfs_create_pending_block_groups(struct btrfs_trans_handle *trans,
 		ret = btrfs_insert_item(trans, extent_root, &key, &item,
 					sizeof(item));
 		if (ret)
-			btrfs_abort_transaction(trans, extent_root, ret);
+			btrfs_abort_transaction(trans, ret);
 		ret = btrfs_finish_chunk_alloc(trans, extent_root,
 					       key.objectid, key.offset);
 		if (ret)
-			btrfs_abort_transaction(trans, extent_root, ret);
+			btrfs_abort_transaction(trans, ret);
 		add_block_group_free_space(trans, root->fs_info, block_group);
 		/* already aborted the transaction if it failed. */
 next:
@@ -10095,7 +10367,6 @@ int btrfs_make_block_group(struct btrfs_trans_handle *trans,
 	int ret;
 	struct btrfs_root *extent_root;
 	struct btrfs_block_group_cache *cache;
-
 	extent_root = root->fs_info->extent_root;
 
 	btrfs_set_log_full_commit(root->fs_info, trans);
@@ -10141,7 +10412,7 @@ int btrfs_make_block_group(struct btrfs_trans_handle *trans,
 	 * assigned to our block group, but don't update its counters just yet.
 	 * We want our bg to be added to the rbtree with its ->space_info set.
 	 */
-	ret = update_space_info(root->fs_info, cache->flags, 0, 0,
+	ret = update_space_info(root->fs_info, cache->flags, 0, 0, 0,
 				&cache->space_info);
 	if (ret) {
 		btrfs_remove_free_space_cache(cache);
@@ -10160,8 +10431,9 @@ int btrfs_make_block_group(struct btrfs_trans_handle *trans,
 	 * Now that our block group has its ->space_info set and is inserted in
 	 * the rbtree, update the space info's counters.
 	 */
+	trace_btrfs_add_block_group(root->fs_info, cache, 1);
 	ret = update_space_info(root->fs_info, cache->flags, size, bytes_used,
-				&cache->space_info);
+				cache->bytes_super, &cache->space_info);
 	if (ret) {
 		btrfs_remove_free_space_cache(cache);
 		spin_lock(&root->fs_info->block_group_cache_lock);
@@ -10174,16 +10446,11 @@ int btrfs_make_block_group(struct btrfs_trans_handle *trans,
 	}
 	update_global_block_rsv(root->fs_info);
 
-	spin_lock(&cache->space_info->lock);
-	cache->space_info->bytes_readonly += cache->bytes_super;
-	spin_unlock(&cache->space_info->lock);
-
 	__link_block_group(cache->space_info, cache);
 
 	list_add_tail(&cache->bg_list, &trans->new_bgs);
 
 	set_avail_alloc_bits(extent_root->fs_info, type);
-
 	return 0;
 }
 
@@ -10396,7 +10663,7 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans,
 	spin_lock(&block_group->space_info->lock);
 	list_del_init(&block_group->ro_list);
 
-	if (btrfs_test_opt(root, ENOSPC_DEBUG)) {
+	if (btrfs_test_opt(root->fs_info, ENOSPC_DEBUG)) {
 		WARN_ON(block_group->space_info->total_bytes
 			< block_group->key.offset);
 		WARN_ON(block_group->space_info->bytes_readonly
@@ -10554,7 +10821,7 @@ void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info)
 	struct btrfs_trans_handle *trans;
 	int ret = 0;
 
-	if (!fs_info->open)
+	if (!test_bit(BTRFS_FS_OPEN, &fs_info->flags))
 		return;
 
 	spin_lock(&fs_info->unused_bgs_lock);
@@ -10664,7 +10931,7 @@ void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info)
 		spin_unlock(&space_info->lock);
 
 		/* DISCARD can flip during remount */
-		trimming = btrfs_test_opt(root, DISCARD);
+		trimming = btrfs_test_opt(root->fs_info, DISCARD);
 
 		/* Implicit trim during transaction commit. */
 		if (trimming)
@@ -10728,21 +10995,21 @@ int btrfs_init_space_info(struct btrfs_fs_info *fs_info)
 		mixed = 1;
 
 	flags = BTRFS_BLOCK_GROUP_SYSTEM;
-	ret = update_space_info(fs_info, flags, 0, 0, &space_info);
+	ret = update_space_info(fs_info, flags, 0, 0, 0, &space_info);
 	if (ret)
 		goto out;
 
 	if (mixed) {
 		flags = BTRFS_BLOCK_GROUP_METADATA | BTRFS_BLOCK_GROUP_DATA;
-		ret = update_space_info(fs_info, flags, 0, 0, &space_info);
+		ret = update_space_info(fs_info, flags, 0, 0, 0, &space_info);
 	} else {
 		flags = BTRFS_BLOCK_GROUP_METADATA;
-		ret = update_space_info(fs_info, flags, 0, 0, &space_info);
+		ret = update_space_info(fs_info, flags, 0, 0, 0, &space_info);
 		if (ret)
 			goto out;
 
 		flags = BTRFS_BLOCK_GROUP_DATA;
-		ret = update_space_info(fs_info, flags, 0, 0, &space_info);
+		ret = update_space_info(fs_info, flags, 0, 0, 0, &space_info);
 	}
 out:
 	return ret;
diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c
index 3cd57825c75f..66a755150056 100644
--- a/fs/btrfs/extent_io.c
+++ b/fs/btrfs/extent_io.c
@@ -20,6 +20,7 @@
 #include "locking.h"
 #include "rcu-string.h"
 #include "backref.h"
+#include "transaction.h"
 
 static struct kmem_cache *extent_state_cache;
 static struct kmem_cache *extent_buffer_cache;
@@ -74,8 +75,7 @@ void btrfs_leak_debug_check(void)
 
 	while (!list_empty(&buffers)) {
 		eb = list_entry(buffers.next, struct extent_buffer, leak_list);
-		printk(KERN_ERR "BTRFS: buffer leak start %llu len %lu "
-		       "refs %d\n",
+		pr_err("BTRFS: buffer leak start %llu len %lu refs %d\n",
 		       eb->start, eb->len, atomic_read(&eb->refs));
 		list_del(&eb->leak_list);
 		kmem_cache_free(extent_buffer_cache, eb);
@@ -163,13 +163,13 @@ int __init extent_io_init(void)
 {
 	extent_state_cache = kmem_cache_create("btrfs_extent_state",
 			sizeof(struct extent_state), 0,
-			SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD, NULL);
+			SLAB_MEM_SPREAD, NULL);
 	if (!extent_state_cache)
 		return -ENOMEM;
 
 	extent_buffer_cache = kmem_cache_create("btrfs_extent_buffer",
 			sizeof(struct extent_buffer), 0,
-			SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD, NULL);
+			SLAB_MEM_SPREAD, NULL);
 	if (!extent_buffer_cache)
 		goto free_state_cache;
 
@@ -460,8 +460,7 @@ static int insert_state(struct extent_io_tree *tree,
 	if (node) {
 		struct extent_state *found;
 		found = rb_entry(node, struct extent_state, rb_node);
-		printk(KERN_ERR "BTRFS: found node %llu %llu on insert of "
-		       "%llu %llu\n",
+		pr_err("BTRFS: found node %llu %llu on insert of %llu %llu\n",
 		       found->start, found->end, start, end);
 		return -EEXIST;
 	}
@@ -572,9 +571,8 @@ alloc_extent_state_atomic(struct extent_state *prealloc)
 
 static void extent_io_tree_panic(struct extent_io_tree *tree, int err)
 {
-	btrfs_panic(tree_fs_info(tree), err, "Locking error: "
-		    "Extent tree was modified by another "
-		    "thread while locked.");
+	btrfs_panic(tree_fs_info(tree), err,
+		    "Locking error: Extent tree was modified by another thread while locked.");
 }
 
 /*
@@ -1729,7 +1727,7 @@ out_failed:
 }
 
 void extent_clear_unlock_delalloc(struct inode *inode, u64 start, u64 end,
-				 struct page *locked_page,
+				 u64 delalloc_end, struct page *locked_page,
 				 unsigned clear_bits,
 				 unsigned long page_ops)
 {
@@ -2025,9 +2023,16 @@ int repair_io_failure(struct inode *inode, u64 start, u64 length, u64 logical,
 	bio->bi_iter.bi_size = 0;
 	map_length = length;
 
+	/*
+	 * Avoid races with device replace and make sure our bbio has devices
+	 * associated to its stripes that don't go away while we are doing the
+	 * read repair operation.
+	 */
+	btrfs_bio_counter_inc_blocked(fs_info);
 	ret = btrfs_map_block(fs_info, WRITE, logical,
 			      &map_length, &bbio, mirror_num);
 	if (ret) {
+		btrfs_bio_counter_dec(fs_info);
 		bio_put(bio);
 		return -EIO;
 	}
@@ -2037,14 +2042,17 @@ int repair_io_failure(struct inode *inode, u64 start, u64 length, u64 logical,
 	dev = bbio->stripes[mirror_num-1].dev;
 	btrfs_put_bbio(bbio);
 	if (!dev || !dev->bdev || !dev->writeable) {
+		btrfs_bio_counter_dec(fs_info);
 		bio_put(bio);
 		return -EIO;
 	}
 	bio->bi_bdev = dev->bdev;
+	bio_set_op_attrs(bio, REQ_OP_WRITE, WRITE_SYNC);
 	bio_add_page(bio, page, length, pg_offset);
 
-	if (btrfsic_submit_bio_wait(WRITE_SYNC, bio)) {
+	if (btrfsic_submit_bio_wait(bio)) {
 		/* try to remap that extent elsewhere? */
+		btrfs_bio_counter_dec(fs_info);
 		bio_put(bio);
 		btrfs_dev_stat_inc_and_print(dev, BTRFS_DEV_STAT_WRITE_ERRS);
 		return -EIO;
@@ -2054,6 +2062,7 @@ int repair_io_failure(struct inode *inode, u64 start, u64 length, u64 logical,
 		"read error corrected: ino %llu off %llu (dev %s sector %llu)",
 				  btrfs_ino(inode), start,
 				  rcu_str_deref(dev->name), sector);
+	btrfs_bio_counter_dec(fs_info);
 	bio_put(bio);
 	return 0;
 }
@@ -2111,8 +2120,9 @@ int clean_io_failure(struct inode *inode, u64 start, struct page *page,
 
 	if (failrec->in_validation) {
 		/* there was no real error, just free the record */
-		pr_debug("clean_io_failure: freeing dummy error at %llu\n",
-			 failrec->start);
+		btrfs_debug(fs_info,
+			"clean_io_failure: freeing dummy error at %llu",
+			failrec->start);
 		goto out;
 	}
 	if (fs_info->sb->s_flags & MS_RDONLY)
@@ -2178,6 +2188,7 @@ void btrfs_free_io_failure_record(struct inode *inode, u64 start, u64 end)
 int btrfs_get_io_failure_record(struct inode *inode, u64 start, u64 end,
 		struct io_failure_record **failrec_ret)
 {
+	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	struct io_failure_record *failrec;
 	struct extent_map *em;
 	struct extent_io_tree *failure_tree = &BTRFS_I(inode)->io_failure_tree;
@@ -2225,8 +2236,9 @@ int btrfs_get_io_failure_record(struct inode *inode, u64 start, u64 end,
 						 em->compress_type);
 		}
 
-		pr_debug("Get IO Failure Record: (new) logical=%llu, start=%llu, len=%llu\n",
-			 logical, start, failrec->len);
+		btrfs_debug(fs_info,
+			"Get IO Failure Record: (new) logical=%llu, start=%llu, len=%llu",
+			logical, start, failrec->len);
 
 		failrec->logical = logical;
 		free_extent_map(em);
@@ -2244,9 +2256,10 @@ int btrfs_get_io_failure_record(struct inode *inode, u64 start, u64 end,
 			return ret;
 		}
 	} else {
-		pr_debug("Get IO Failure Record: (found) logical=%llu, start=%llu, len=%llu, validation=%d\n",
-			 failrec->logical, failrec->start, failrec->len,
-			 failrec->in_validation);
+		btrfs_debug(fs_info,
+			"Get IO Failure Record: (found) logical=%llu, start=%llu, len=%llu, validation=%d",
+			failrec->logical, failrec->start, failrec->len,
+			failrec->in_validation);
 		/*
 		 * when data can be on disk more than twice, add to failrec here
 		 * (e.g. with a list for failed_mirror) to make
@@ -2262,18 +2275,19 @@ int btrfs_get_io_failure_record(struct inode *inode, u64 start, u64 end,
 int btrfs_check_repairable(struct inode *inode, struct bio *failed_bio,
 			   struct io_failure_record *failrec, int failed_mirror)
 {
+	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	int num_copies;
 
-	num_copies = btrfs_num_copies(BTRFS_I(inode)->root->fs_info,
-				      failrec->logical, failrec->len);
+	num_copies = btrfs_num_copies(fs_info, failrec->logical, failrec->len);
 	if (num_copies == 1) {
 		/*
 		 * we only have a single copy of the data, so don't bother with
 		 * all the retry and error correction code that follows. no
 		 * matter what the error is, it is very likely to persist.
 		 */
-		pr_debug("Check Repairable: cannot repair, num_copies=%d, next_mirror %d, failed_mirror %d\n",
-			 num_copies, failrec->this_mirror, failed_mirror);
+		btrfs_debug(fs_info,
+			"Check Repairable: cannot repair, num_copies=%d, next_mirror %d, failed_mirror %d",
+			num_copies, failrec->this_mirror, failed_mirror);
 		return 0;
 	}
 
@@ -2312,8 +2326,9 @@ int btrfs_check_repairable(struct inode *inode, struct bio *failed_bio,
 	}
 
 	if (failrec->this_mirror > num_copies) {
-		pr_debug("Check Repairable: (fail) num_copies=%d, next_mirror %d, failed_mirror %d\n",
-			 num_copies, failrec->this_mirror, failed_mirror);
+		btrfs_debug(fs_info,
+			"Check Repairable: (fail) num_copies=%d, next_mirror %d, failed_mirror %d",
+			num_copies, failrec->this_mirror, failed_mirror);
 		return 0;
 	}
 
@@ -2376,7 +2391,7 @@ static int bio_readpage_error(struct bio *failed_bio, u64 phy_offset,
 	int read_mode;
 	int ret;
 
-	BUG_ON(failed_bio->bi_rw & REQ_WRITE);
+	BUG_ON(bio_op(failed_bio) == REQ_OP_WRITE);
 
 	ret = btrfs_get_io_failure_record(inode, start, end, &failrec);
 	if (ret)
@@ -2402,12 +2417,13 @@ static int bio_readpage_error(struct bio *failed_bio, u64 phy_offset,
 		free_io_failure(inode, failrec);
 		return -EIO;
 	}
+	bio_set_op_attrs(bio, REQ_OP_READ, read_mode);
 
-	pr_debug("Repair Read Error: submitting new read[%#x] to this_mirror=%d, in_validation=%d\n",
-		 read_mode, failrec->this_mirror, failrec->in_validation);
+	btrfs_debug(btrfs_sb(inode->i_sb),
+		"Repair Read Error: submitting new read[%#x] to this_mirror=%d, in_validation=%d",
+		read_mode, failrec->this_mirror, failrec->in_validation);
 
-	ret = tree->ops->submit_bio_hook(inode, read_mode, bio,
-					 failrec->this_mirror,
+	ret = tree->ops->submit_bio_hook(inode, bio, failrec->this_mirror,
 					 failrec->bio_flags, 0);
 	if (ret) {
 		free_io_failure(inode, failrec);
@@ -2473,8 +2489,7 @@ static void end_bio_extent_writepage(struct bio *bio)
 					bvec->bv_offset, bvec->bv_len);
 			else
 				btrfs_info(BTRFS_I(page->mapping->host)->root->fs_info,
-				   "incomplete page write in btrfs with offset %u and "
-				   "length %u",
+				   "incomplete page write in btrfs with offset %u and length %u",
 					bvec->bv_offset, bvec->bv_len);
 		}
 
@@ -2530,10 +2545,12 @@ static void end_bio_extent_readpage(struct bio *bio)
 	bio_for_each_segment_all(bvec, bio, i) {
 		struct page *page = bvec->bv_page;
 		struct inode *inode = page->mapping->host;
+		struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 
-		pr_debug("end_bio_extent_readpage: bi_sector=%llu, err=%d, "
-			 "mirror=%u\n", (u64)bio->bi_iter.bi_sector,
-			 bio->bi_error, io_bio->mirror_num);
+		btrfs_debug(fs_info,
+			"end_bio_extent_readpage: bi_sector=%llu, err=%d, mirror=%u",
+			(u64)bio->bi_iter.bi_sector, bio->bi_error,
+			io_bio->mirror_num);
 		tree = &BTRFS_I(inode)->io_tree;
 
 		/* We always issue full-page reads, but if some block
@@ -2543,13 +2560,12 @@ static void end_bio_extent_readpage(struct bio *bio)
 		 * if they don't add up to a full page.  */
 		if (bvec->bv_offset || bvec->bv_len != PAGE_SIZE) {
 			if (bvec->bv_offset + bvec->bv_len != PAGE_SIZE)
-				btrfs_err(BTRFS_I(page->mapping->host)->root->fs_info,
-				   "partial page read in btrfs with offset %u and length %u",
+				btrfs_err(fs_info,
+					"partial page read in btrfs with offset %u and length %u",
 					bvec->bv_offset, bvec->bv_len);
 			else
-				btrfs_info(BTRFS_I(page->mapping->host)->root->fs_info,
-				   "incomplete page read in btrfs with offset %u and "
-				   "length %u",
+				btrfs_info(fs_info,
+					"incomplete page read in btrfs with offset %u and length %u",
 					bvec->bv_offset, bvec->bv_len);
 		}
 
@@ -2686,12 +2702,6 @@ struct bio *btrfs_bio_clone(struct bio *bio, gfp_t gfp_mask)
 		btrfs_bio->csum = NULL;
 		btrfs_bio->csum_allocated = NULL;
 		btrfs_bio->end_io = NULL;
-
-#ifdef CONFIG_BLK_CGROUP
-		/* FIXME, put this into bio_clone_bioset */
-		if (bio->bi_css)
-			bio_associate_blkcg(new, bio->bi_css);
-#endif
 	}
 	return new;
 }
@@ -2713,8 +2723,8 @@ struct bio *btrfs_io_bio_alloc(gfp_t gfp_mask, unsigned int nr_iovecs)
 }
 
 
-static int __must_check submit_one_bio(int rw, struct bio *bio,
-				       int mirror_num, unsigned long bio_flags)
+static int __must_check submit_one_bio(struct bio *bio, int mirror_num,
+				       unsigned long bio_flags)
 {
 	int ret = 0;
 	struct bio_vec *bvec = bio->bi_io_vec + bio->bi_vcnt - 1;
@@ -2725,33 +2735,31 @@ static int __must_check submit_one_bio(int rw, struct bio *bio,
 	start = page_offset(page) + bvec->bv_offset;
 
 	bio->bi_private = NULL;
-
 	bio_get(bio);
 
 	if (tree->ops && tree->ops->submit_bio_hook)
-		ret = tree->ops->submit_bio_hook(page->mapping->host, rw, bio,
+		ret = tree->ops->submit_bio_hook(page->mapping->host, bio,
 					   mirror_num, bio_flags, start);
 	else
-		btrfsic_submit_bio(rw, bio);
+		btrfsic_submit_bio(bio);
 
 	bio_put(bio);
 	return ret;
 }
 
-static int merge_bio(int rw, struct extent_io_tree *tree, struct page *page,
+static int merge_bio(struct extent_io_tree *tree, struct page *page,
 		     unsigned long offset, size_t size, struct bio *bio,
 		     unsigned long bio_flags)
 {
 	int ret = 0;
 	if (tree->ops && tree->ops->merge_bio_hook)
-		ret = tree->ops->merge_bio_hook(rw, page, offset, size, bio,
+		ret = tree->ops->merge_bio_hook(page, offset, size, bio,
 						bio_flags);
-	BUG_ON(ret < 0);
 	return ret;
 
 }
 
-static int submit_extent_page(int rw, struct extent_io_tree *tree,
+static int submit_extent_page(int op, int op_flags, struct extent_io_tree *tree,
 			      struct writeback_control *wbc,
 			      struct page *page, sector_t sector,
 			      size_t size, unsigned long offset,
@@ -2779,10 +2787,9 @@ static int submit_extent_page(int rw, struct extent_io_tree *tree,
 
 		if (prev_bio_flags != bio_flags || !contig ||
 		    force_bio_submit ||
-		    merge_bio(rw, tree, page, offset, page_size, bio, bio_flags) ||
+		    merge_bio(tree, page, offset, page_size, bio, bio_flags) ||
 		    bio_add_page(bio, page, page_size, offset) < page_size) {
-			ret = submit_one_bio(rw, bio, mirror_num,
-					     prev_bio_flags);
+			ret = submit_one_bio(bio, mirror_num, prev_bio_flags);
 			if (ret < 0) {
 				*bio_ret = NULL;
 				return ret;
@@ -2803,6 +2810,7 @@ static int submit_extent_page(int rw, struct extent_io_tree *tree,
 	bio_add_page(bio, page, page_size, offset);
 	bio->bi_end_io = end_io_func;
 	bio->bi_private = tree;
+	bio_set_op_attrs(bio, op, op_flags);
 	if (wbc) {
 		wbc_init_bio(wbc, bio);
 		wbc_account_io(wbc, page, page_size);
@@ -2811,7 +2819,7 @@ static int submit_extent_page(int rw, struct extent_io_tree *tree,
 	if (bio_ret)
 		*bio_ret = bio;
 	else
-		ret = submit_one_bio(rw, bio, mirror_num, bio_flags);
+		ret = submit_one_bio(bio, mirror_num, bio_flags);
 
 	return ret;
 }
@@ -2869,13 +2877,14 @@ __get_extent_map(struct inode *inode, struct page *page, size_t pg_offset,
  * into the tree that are removed when the IO is done (by the end_io
  * handlers)
  * XXX JDM: This needs looking at to ensure proper page locking
+ * return 0 on success, otherwise return error
  */
 static int __do_readpage(struct extent_io_tree *tree,
 			 struct page *page,
 			 get_extent_t *get_extent,
 			 struct extent_map **em_cached,
 			 struct bio **bio, int mirror_num,
-			 unsigned long *bio_flags, int rw,
+			 unsigned long *bio_flags, int read_flags,
 			 u64 *prev_em_start)
 {
 	struct inode *inode = page->mapping->host;
@@ -2890,7 +2899,7 @@ static int __do_readpage(struct extent_io_tree *tree,
 	sector_t sector;
 	struct extent_map *em;
 	struct block_device *bdev;
-	int ret;
+	int ret = 0;
 	int nr = 0;
 	size_t pg_offset = 0;
 	size_t iosize;
@@ -3058,8 +3067,8 @@ static int __do_readpage(struct extent_io_tree *tree,
 		}
 
 		pnr -= page->index;
-		ret = submit_extent_page(rw, tree, NULL, page,
-					 sector, disk_io_size, pg_offset,
+		ret = submit_extent_page(REQ_OP_READ, read_flags, tree, NULL,
+					 page, sector, disk_io_size, pg_offset,
 					 bdev, bio, pnr,
 					 end_bio_extent_readpage, mirror_num,
 					 *bio_flags,
@@ -3071,6 +3080,7 @@ static int __do_readpage(struct extent_io_tree *tree,
 		} else {
 			SetPageError(page);
 			unlock_extent(tree, cur, cur + iosize - 1);
+			goto out;
 		}
 		cur = cur + iosize;
 		pg_offset += iosize;
@@ -3081,7 +3091,7 @@ out:
 			SetPageUptodate(page);
 		unlock_page(page);
 	}
-	return 0;
+	return ret;
 }
 
 static inline void __do_contiguous_readpages(struct extent_io_tree *tree,
@@ -3090,7 +3100,7 @@ static inline void __do_contiguous_readpages(struct extent_io_tree *tree,
 					     get_extent_t *get_extent,
 					     struct extent_map **em_cached,
 					     struct bio **bio, int mirror_num,
-					     unsigned long *bio_flags, int rw,
+					     unsigned long *bio_flags,
 					     u64 *prev_em_start)
 {
 	struct inode *inode;
@@ -3111,7 +3121,7 @@ static inline void __do_contiguous_readpages(struct extent_io_tree *tree,
 
 	for (index = 0; index < nr_pages; index++) {
 		__do_readpage(tree, pages[index], get_extent, em_cached, bio,
-			      mirror_num, bio_flags, rw, prev_em_start);
+			      mirror_num, bio_flags, 0, prev_em_start);
 		put_page(pages[index]);
 	}
 }
@@ -3121,7 +3131,7 @@ static void __extent_readpages(struct extent_io_tree *tree,
 			       int nr_pages, get_extent_t *get_extent,
 			       struct extent_map **em_cached,
 			       struct bio **bio, int mirror_num,
-			       unsigned long *bio_flags, int rw,
+			       unsigned long *bio_flags,
 			       u64 *prev_em_start)
 {
 	u64 start = 0;
@@ -3143,7 +3153,7 @@ static void __extent_readpages(struct extent_io_tree *tree,
 						  index - first_index, start,
 						  end, get_extent, em_cached,
 						  bio, mirror_num, bio_flags,
-						  rw, prev_em_start);
+						  prev_em_start);
 			start = page_start;
 			end = start + PAGE_SIZE - 1;
 			first_index = index;
@@ -3154,7 +3164,7 @@ static void __extent_readpages(struct extent_io_tree *tree,
 		__do_contiguous_readpages(tree, &pages[first_index],
 					  index - first_index, start,
 					  end, get_extent, em_cached, bio,
-					  mirror_num, bio_flags, rw,
+					  mirror_num, bio_flags,
 					  prev_em_start);
 }
 
@@ -3162,7 +3172,7 @@ static int __extent_read_full_page(struct extent_io_tree *tree,
 				   struct page *page,
 				   get_extent_t *get_extent,
 				   struct bio **bio, int mirror_num,
-				   unsigned long *bio_flags, int rw)
+				   unsigned long *bio_flags, int read_flags)
 {
 	struct inode *inode = page->mapping->host;
 	struct btrfs_ordered_extent *ordered;
@@ -3182,7 +3192,7 @@ static int __extent_read_full_page(struct extent_io_tree *tree,
 	}
 
 	ret = __do_readpage(tree, page, get_extent, NULL, bio, mirror_num,
-			    bio_flags, rw, NULL);
+			    bio_flags, read_flags, NULL);
 	return ret;
 }
 
@@ -3194,9 +3204,9 @@ int extent_read_full_page(struct extent_io_tree *tree, struct page *page,
 	int ret;
 
 	ret = __extent_read_full_page(tree, page, get_extent, &bio, mirror_num,
-				      &bio_flags, READ);
+				      &bio_flags, 0);
 	if (bio)
-		ret = submit_one_bio(READ, bio, mirror_num, bio_flags);
+		ret = submit_one_bio(bio, mirror_num, bio_flags);
 	return ret;
 }
 
@@ -3430,8 +3440,8 @@ static noinline_for_stack int __extent_writepage_io(struct inode *inode,
 			       page->index, cur, end);
 		}
 
-		ret = submit_extent_page(write_flags, tree, wbc, page,
-					 sector, iosize, pg_offset,
+		ret = submit_extent_page(REQ_OP_WRITE, write_flags, tree, wbc,
+					 page, sector, iosize, pg_offset,
 					 bdev, &epd->bio, max_nr,
 					 end_bio_extent_writepage,
 					 0, 0, 0, false);
@@ -3470,13 +3480,11 @@ static int __extent_writepage(struct page *page, struct writeback_control *wbc,
 	size_t pg_offset = 0;
 	loff_t i_size = i_size_read(inode);
 	unsigned long end_index = i_size >> PAGE_SHIFT;
-	int write_flags;
+	int write_flags = 0;
 	unsigned long nr_written = 0;
 
 	if (wbc->sync_mode == WB_SYNC_ALL)
 		write_flags = WRITE_SYNC;
-	else
-		write_flags = WRITE;
 
 	trace___extent_writepage(page, inode, wbc);
 
@@ -3621,7 +3629,6 @@ static void end_extent_buffer_writeback(struct extent_buffer *eb)
 static void set_btree_ioerr(struct page *page)
 {
 	struct extent_buffer *eb = (struct extent_buffer *)page->private;
-	struct btrfs_inode *btree_ino = BTRFS_I(eb->fs_info->btree_inode);
 
 	SetPageError(page);
 	if (test_and_set_bit(EXTENT_BUFFER_WRITE_ERR, &eb->bflags))
@@ -3667,13 +3674,13 @@ static void set_btree_ioerr(struct page *page)
 	 */
 	switch (eb->log_index) {
 	case -1:
-		set_bit(BTRFS_INODE_BTREE_ERR, &btree_ino->runtime_flags);
+		set_bit(BTRFS_FS_BTREE_ERR, &eb->fs_info->flags);
 		break;
 	case 0:
-		set_bit(BTRFS_INODE_BTREE_LOG1_ERR, &btree_ino->runtime_flags);
+		set_bit(BTRFS_FS_LOG1_ERR, &eb->fs_info->flags);
 		break;
 	case 1:
-		set_bit(BTRFS_INODE_BTREE_LOG2_ERR, &btree_ino->runtime_flags);
+		set_bit(BTRFS_FS_LOG2_ERR, &eb->fs_info->flags);
 		break;
 	default:
 		BUG(); /* unexpected, logic error */
@@ -3718,9 +3725,11 @@ static noinline_for_stack int write_one_eb(struct extent_buffer *eb,
 	struct block_device *bdev = fs_info->fs_devices->latest_bdev;
 	struct extent_io_tree *tree = &BTRFS_I(fs_info->btree_inode)->io_tree;
 	u64 offset = eb->start;
+	u32 nritems;
 	unsigned long i, num_pages;
 	unsigned long bio_flags = 0;
-	int rw = (epd->sync_io ? WRITE_SYNC : WRITE) | REQ_META;
+	unsigned long start, end;
+	int write_flags = (epd->sync_io ? WRITE_SYNC : 0) | REQ_META;
 	int ret = 0;
 
 	clear_bit(EXTENT_BUFFER_WRITE_ERR, &eb->bflags);
@@ -3729,14 +3738,32 @@ static noinline_for_stack int write_one_eb(struct extent_buffer *eb,
 	if (btrfs_header_owner(eb) == BTRFS_TREE_LOG_OBJECTID)
 		bio_flags = EXTENT_BIO_TREE_LOG;
 
+	/* set btree blocks beyond nritems with 0 to avoid stale content. */
+	nritems = btrfs_header_nritems(eb);
+	if (btrfs_header_level(eb) > 0) {
+		end = btrfs_node_key_ptr_offset(nritems);
+
+		memset_extent_buffer(eb, 0, end, eb->len - end);
+	} else {
+		/*
+		 * leaf:
+		 * header 0 1 2 .. N ... data_N .. data_2 data_1 data_0
+		 */
+		start = btrfs_item_nr_offset(nritems);
+		end = btrfs_leaf_data(eb) +
+		      leaf_data_end(fs_info->tree_root, eb);
+		memset_extent_buffer(eb, 0, start, end - start);
+	}
+
 	for (i = 0; i < num_pages; i++) {
 		struct page *p = eb->pages[i];
 
 		clear_page_dirty_for_io(p);
 		set_page_writeback(p);
-		ret = submit_extent_page(rw, tree, wbc, p, offset >> 9,
-					 PAGE_SIZE, 0, bdev, &epd->bio,
-					 -1, end_bio_extent_buffer_writepage,
+		ret = submit_extent_page(REQ_OP_WRITE, write_flags, tree, wbc,
+					 p, offset >> 9, PAGE_SIZE, 0, bdev,
+					 &epd->bio, -1,
+					 end_bio_extent_buffer_writepage,
 					 0, epd->bio_flags, bio_flags, false);
 		epd->bio_flags = bio_flags;
 		if (ret) {
@@ -4046,13 +4073,12 @@ retry:
 static void flush_epd_write_bio(struct extent_page_data *epd)
 {
 	if (epd->bio) {
-		int rw = WRITE;
 		int ret;
 
-		if (epd->sync_io)
-			rw = WRITE_SYNC;
+		bio_set_op_attrs(epd->bio, REQ_OP_WRITE,
+				 epd->sync_io ? WRITE_SYNC : 0);
 
-		ret = submit_one_bio(rw, epd->bio, 0, epd->bio_flags);
+		ret = submit_one_bio(epd->bio, 0, epd->bio_flags);
 		BUG_ON(ret < 0); /* -ENOMEM */
 		epd->bio = NULL;
 	}
@@ -4170,7 +4196,8 @@ int extent_readpages(struct extent_io_tree *tree,
 		prefetchw(&page->flags);
 		list_del(&page->lru);
 		if (add_to_page_cache_lru(page, mapping,
-					page->index, GFP_NOFS)) {
+					page->index,
+					readahead_gfp_mask(mapping))) {
 			put_page(page);
 			continue;
 		}
@@ -4179,19 +4206,19 @@ int extent_readpages(struct extent_io_tree *tree,
 		if (nr < ARRAY_SIZE(pagepool))
 			continue;
 		__extent_readpages(tree, pagepool, nr, get_extent, &em_cached,
-				   &bio, 0, &bio_flags, READ, &prev_em_start);
+				   &bio, 0, &bio_flags, &prev_em_start);
 		nr = 0;
 	}
 	if (nr)
 		__extent_readpages(tree, pagepool, nr, get_extent, &em_cached,
-				   &bio, 0, &bio_flags, READ, &prev_em_start);
+				   &bio, 0, &bio_flags, &prev_em_start);
 
 	if (em_cached)
 		free_extent_map(em_cached);
 
 	BUG_ON(!list_empty(pages));
 	if (bio)
-		return submit_one_bio(READ, bio, 0, bio_flags);
+		return submit_one_bio(bio, 0, bio_flags);
 	return 0;
 }
 
@@ -4483,21 +4510,36 @@ int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 			flags |= (FIEMAP_EXTENT_DELALLOC |
 				  FIEMAP_EXTENT_UNKNOWN);
 		} else if (fieinfo->fi_extents_max) {
+			struct btrfs_trans_handle *trans;
+
 			u64 bytenr = em->block_start -
 				(em->start - em->orig_start);
 
 			disko = em->block_start + offset_in_extent;
 
 			/*
+			 * We need a trans handle to get delayed refs
+			 */
+			trans = btrfs_join_transaction(root);
+			/*
+			 * It's OK if we can't start a trans we can still check
+			 * from commit_root
+			 */
+			if (IS_ERR(trans))
+				trans = NULL;
+
+			/*
 			 * As btrfs supports shared space, this information
 			 * can be exported to userspace tools via
 			 * flag FIEMAP_EXTENT_SHARED.  If fi_extents_max == 0
 			 * then we're just getting a count and we can skip the
 			 * lookup stuff.
 			 */
-			ret = btrfs_check_shared(NULL, root->fs_info,
+			ret = btrfs_check_shared(trans, root->fs_info,
 						 root->objectid,
 						 btrfs_ino(inode), bytenr);
+			if (trans)
+				btrfs_end_transaction(trans, root);
 			if (ret < 0)
 				goto out_free;
 			if (ret)
@@ -4718,16 +4760,16 @@ err:
 }
 
 struct extent_buffer *alloc_dummy_extent_buffer(struct btrfs_fs_info *fs_info,
-						u64 start)
+						u64 start, u32 nodesize)
 {
 	unsigned long len;
 
 	if (!fs_info) {
 		/*
 		 * Called only from tests that don't always have a fs_info
-		 * available, but we know that nodesize is 4096
+		 * available
 		 */
-		len = 4096;
+		len = nodesize;
 	} else {
 		len = fs_info->tree_root->nodesize;
 	}
@@ -4823,7 +4865,7 @@ struct extent_buffer *find_extent_buffer(struct btrfs_fs_info *fs_info,
 
 #ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
 struct extent_buffer *alloc_test_extent_buffer(struct btrfs_fs_info *fs_info,
-					       u64 start)
+					u64 start, u32 nodesize)
 {
 	struct extent_buffer *eb, *exists = NULL;
 	int ret;
@@ -4831,7 +4873,7 @@ struct extent_buffer *alloc_test_extent_buffer(struct btrfs_fs_info *fs_info,
 	eb = find_extent_buffer(fs_info, start);
 	if (eb)
 		return eb;
-	eb = alloc_dummy_extent_buffer(fs_info, start);
+	eb = alloc_dummy_extent_buffer(fs_info, start, nodesize);
 	if (!eb)
 		return NULL;
 	eb->fs_info = fs_info;
@@ -4882,18 +4924,25 @@ struct extent_buffer *alloc_extent_buffer(struct btrfs_fs_info *fs_info,
 	int uptodate = 1;
 	int ret;
 
+	if (!IS_ALIGNED(start, fs_info->tree_root->sectorsize)) {
+		btrfs_err(fs_info, "bad tree block start %llu", start);
+		return ERR_PTR(-EINVAL);
+	}
+
 	eb = find_extent_buffer(fs_info, start);
 	if (eb)
 		return eb;
 
 	eb = __alloc_extent_buffer(fs_info, start, len);
 	if (!eb)
-		return NULL;
+		return ERR_PTR(-ENOMEM);
 
 	for (i = 0; i < num_pages; i++, index++) {
 		p = find_or_create_page(mapping, index, GFP_NOFS|__GFP_NOFAIL);
-		if (!p)
+		if (!p) {
+			exists = ERR_PTR(-ENOMEM);
 			goto free_eb;
+		}
 
 		spin_lock(&mapping->private_lock);
 		if (PagePrivate(p)) {
@@ -4938,8 +4987,10 @@ struct extent_buffer *alloc_extent_buffer(struct btrfs_fs_info *fs_info,
 		set_bit(EXTENT_BUFFER_UPTODATE, &eb->bflags);
 again:
 	ret = radix_tree_preload(GFP_NOFS);
-	if (ret)
+	if (ret) {
+		exists = ERR_PTR(ret);
 		goto free_eb;
+	}
 
 	spin_lock(&fs_info->buffer_lock);
 	ret = radix_tree_insert(&fs_info->buffer_radix,
@@ -5160,11 +5211,10 @@ int extent_buffer_uptodate(struct extent_buffer *eb)
 }
 
 int read_extent_buffer_pages(struct extent_io_tree *tree,
-			     struct extent_buffer *eb, u64 start, int wait,
+			     struct extent_buffer *eb, int wait,
 			     get_extent_t *get_extent, int mirror_num)
 {
 	unsigned long i;
-	unsigned long start_i;
 	struct page *page;
 	int err;
 	int ret = 0;
@@ -5178,16 +5228,8 @@ int read_extent_buffer_pages(struct extent_io_tree *tree,
 	if (test_bit(EXTENT_BUFFER_UPTODATE, &eb->bflags))
 		return 0;
 
-	if (start) {
-		WARN_ON(start < eb->start);
-		start_i = (start >> PAGE_SHIFT) -
-			(eb->start >> PAGE_SHIFT);
-	} else {
-		start_i = 0;
-	}
-
 	num_pages = num_extent_pages(eb->start, eb->len);
-	for (i = start_i; i < num_pages; i++) {
+	for (i = 0; i < num_pages; i++) {
 		page = eb->pages[i];
 		if (wait == WAIT_NONE) {
 			if (!trylock_page(page))
@@ -5196,38 +5238,62 @@ int read_extent_buffer_pages(struct extent_io_tree *tree,
 			lock_page(page);
 		}
 		locked_pages++;
+	}
+	/*
+	 * We need to firstly lock all pages to make sure that
+	 * the uptodate bit of our pages won't be affected by
+	 * clear_extent_buffer_uptodate().
+	 */
+	for (i = 0; i < num_pages; i++) {
+		page = eb->pages[i];
 		if (!PageUptodate(page)) {
 			num_reads++;
 			all_uptodate = 0;
 		}
 	}
+
 	if (all_uptodate) {
-		if (start_i == 0)
-			set_bit(EXTENT_BUFFER_UPTODATE, &eb->bflags);
+		set_bit(EXTENT_BUFFER_UPTODATE, &eb->bflags);
 		goto unlock_exit;
 	}
 
 	clear_bit(EXTENT_BUFFER_READ_ERR, &eb->bflags);
 	eb->read_mirror = 0;
 	atomic_set(&eb->io_pages, num_reads);
-	for (i = start_i; i < num_pages; i++) {
+	for (i = 0; i < num_pages; i++) {
 		page = eb->pages[i];
+
 		if (!PageUptodate(page)) {
+			if (ret) {
+				atomic_dec(&eb->io_pages);
+				unlock_page(page);
+				continue;
+			}
+
 			ClearPageError(page);
 			err = __extent_read_full_page(tree, page,
 						      get_extent, &bio,
 						      mirror_num, &bio_flags,
-						      READ | REQ_META);
-			if (err)
+						      REQ_META);
+			if (err) {
 				ret = err;
+				/*
+				 * We use &bio in above __extent_read_full_page,
+				 * so we ensure that if it returns error, the
+				 * current page fails to add itself to bio and
+				 * it's been unlocked.
+				 *
+				 * We must dec io_pages by ourselves.
+				 */
+				atomic_dec(&eb->io_pages);
+			}
 		} else {
 			unlock_page(page);
 		}
 	}
 
 	if (bio) {
-		err = submit_one_bio(READ | REQ_META, bio, mirror_num,
-				     bio_flags);
+		err = submit_one_bio(bio, mirror_num, bio_flags);
 		if (err)
 			return err;
 	}
@@ -5235,7 +5301,7 @@ int read_extent_buffer_pages(struct extent_io_tree *tree,
 	if (ret || wait != WAIT_COMPLETE)
 		return ret;
 
-	for (i = start_i; i < num_pages; i++) {
+	for (i = 0; i < num_pages; i++) {
 		page = eb->pages[i];
 		wait_on_page_locked(page);
 		if (!PageUptodate(page))
@@ -5245,12 +5311,10 @@ int read_extent_buffer_pages(struct extent_io_tree *tree,
 	return ret;
 
 unlock_exit:
-	i = start_i;
 	while (locked_pages > 0) {
-		page = eb->pages[i];
-		i++;
-		unlock_page(page);
 		locked_pages--;
+		page = eb->pages[locked_pages];
+		unlock_page(page);
 	}
 	return ret;
 }
@@ -5323,6 +5387,11 @@ int read_extent_buffer_to_user(struct extent_buffer *eb, void __user *dstv,
 	return ret;
 }
 
+/*
+ * return 0 if the item is found within a page.
+ * return 1 if the item spans two pages.
+ * return -EINVAL otherwise.
+ */
 int map_private_extent_buffer(struct extent_buffer *eb, unsigned long start,
 			       unsigned long min_len, char **map,
 			       unsigned long *map_start,
@@ -5337,7 +5406,7 @@ int map_private_extent_buffer(struct extent_buffer *eb, unsigned long start,
 		PAGE_SHIFT;
 
 	if (i != end_i)
-		return -EINVAL;
+		return 1;
 
 	if (i == 0) {
 		offset = start_offset;
@@ -5348,8 +5417,7 @@ int map_private_extent_buffer(struct extent_buffer *eb, unsigned long start,
 	}
 
 	if (start + min_len > eb->len) {
-		WARN(1, KERN_ERR "btrfs bad mapping eb start %llu len %lu, "
-		       "wanted %lu %lu\n",
+		WARN(1, KERN_ERR "btrfs bad mapping eb start %llu len %lu, wanted %lu %lu\n",
 		       eb->start, eb->len, start, min_len);
 		return -EINVAL;
 	}
@@ -5490,17 +5558,45 @@ void copy_extent_buffer(struct extent_buffer *dst, struct extent_buffer *src,
 	}
 }
 
-/*
- * The extent buffer bitmap operations are done with byte granularity because
- * bitmap items are not guaranteed to be aligned to a word and therefore a
- * single word in a bitmap may straddle two pages in the extent buffer.
- */
-#define BIT_BYTE(nr) ((nr) / BITS_PER_BYTE)
-#define BYTE_MASK ((1 << BITS_PER_BYTE) - 1)
-#define BITMAP_FIRST_BYTE_MASK(start) \
-	((BYTE_MASK << ((start) & (BITS_PER_BYTE - 1))) & BYTE_MASK)
-#define BITMAP_LAST_BYTE_MASK(nbits) \
-	(BYTE_MASK >> (-(nbits) & (BITS_PER_BYTE - 1)))
+void le_bitmap_set(u8 *map, unsigned int start, int len)
+{
+	u8 *p = map + BIT_BYTE(start);
+	const unsigned int size = start + len;
+	int bits_to_set = BITS_PER_BYTE - (start % BITS_PER_BYTE);
+	u8 mask_to_set = BITMAP_FIRST_BYTE_MASK(start);
+
+	while (len - bits_to_set >= 0) {
+		*p |= mask_to_set;
+		len -= bits_to_set;
+		bits_to_set = BITS_PER_BYTE;
+		mask_to_set = ~(u8)0;
+		p++;
+	}
+	if (len) {
+		mask_to_set &= BITMAP_LAST_BYTE_MASK(size);
+		*p |= mask_to_set;
+	}
+}
+
+void le_bitmap_clear(u8 *map, unsigned int start, int len)
+{
+	u8 *p = map + BIT_BYTE(start);
+	const unsigned int size = start + len;
+	int bits_to_clear = BITS_PER_BYTE - (start % BITS_PER_BYTE);
+	u8 mask_to_clear = BITMAP_FIRST_BYTE_MASK(start);
+
+	while (len - bits_to_clear >= 0) {
+		*p &= ~mask_to_clear;
+		len -= bits_to_clear;
+		bits_to_clear = BITS_PER_BYTE;
+		mask_to_clear = ~(u8)0;
+		p++;
+	}
+	if (len) {
+		mask_to_clear &= BITMAP_LAST_BYTE_MASK(size);
+		*p &= ~mask_to_clear;
+	}
+}
 
 /*
  * eb_bitmap_offset() - calculate the page and offset of the byte containing the
@@ -5544,7 +5640,7 @@ static inline void eb_bitmap_offset(struct extent_buffer *eb,
 int extent_buffer_test_bit(struct extent_buffer *eb, unsigned long start,
 			   unsigned long nr)
 {
-	char *kaddr;
+	u8 *kaddr;
 	struct page *page;
 	unsigned long i;
 	size_t offset;
@@ -5566,13 +5662,13 @@ int extent_buffer_test_bit(struct extent_buffer *eb, unsigned long start,
 void extent_buffer_bitmap_set(struct extent_buffer *eb, unsigned long start,
 			      unsigned long pos, unsigned long len)
 {
-	char *kaddr;
+	u8 *kaddr;
 	struct page *page;
 	unsigned long i;
 	size_t offset;
 	const unsigned int size = pos + len;
 	int bits_to_set = BITS_PER_BYTE - (pos % BITS_PER_BYTE);
-	unsigned int mask_to_set = BITMAP_FIRST_BYTE_MASK(pos);
+	u8 mask_to_set = BITMAP_FIRST_BYTE_MASK(pos);
 
 	eb_bitmap_offset(eb, start, pos, &i, &offset);
 	page = eb->pages[i];
@@ -5583,7 +5679,7 @@ void extent_buffer_bitmap_set(struct extent_buffer *eb, unsigned long start,
 		kaddr[offset] |= mask_to_set;
 		len -= bits_to_set;
 		bits_to_set = BITS_PER_BYTE;
-		mask_to_set = ~0U;
+		mask_to_set = ~(u8)0;
 		if (++offset >= PAGE_SIZE && len > 0) {
 			offset = 0;
 			page = eb->pages[++i];
@@ -5608,13 +5704,13 @@ void extent_buffer_bitmap_set(struct extent_buffer *eb, unsigned long start,
 void extent_buffer_bitmap_clear(struct extent_buffer *eb, unsigned long start,
 				unsigned long pos, unsigned long len)
 {
-	char *kaddr;
+	u8 *kaddr;
 	struct page *page;
 	unsigned long i;
 	size_t offset;
 	const unsigned int size = pos + len;
 	int bits_to_clear = BITS_PER_BYTE - (pos % BITS_PER_BYTE);
-	unsigned int mask_to_clear = BITMAP_FIRST_BYTE_MASK(pos);
+	u8 mask_to_clear = BITMAP_FIRST_BYTE_MASK(pos);
 
 	eb_bitmap_offset(eb, start, pos, &i, &offset);
 	page = eb->pages[i];
@@ -5625,7 +5721,7 @@ void extent_buffer_bitmap_clear(struct extent_buffer *eb, unsigned long start,
 		kaddr[offset] &= ~mask_to_clear;
 		len -= bits_to_clear;
 		bits_to_clear = BITS_PER_BYTE;
-		mask_to_clear = ~0U;
+		mask_to_clear = ~(u8)0;
 		if (++offset >= PAGE_SIZE && len > 0) {
 			offset = 0;
 			page = eb->pages[++i];
@@ -5679,14 +5775,14 @@ void memcpy_extent_buffer(struct extent_buffer *dst, unsigned long dst_offset,
 
 	if (src_offset + len > dst->len) {
 		btrfs_err(dst->fs_info,
-			"memmove bogus src_offset %lu move "
-		       "len %lu dst len %lu", src_offset, len, dst->len);
+			"memmove bogus src_offset %lu move len %lu dst len %lu",
+			 src_offset, len, dst->len);
 		BUG_ON(1);
 	}
 	if (dst_offset + len > dst->len) {
 		btrfs_err(dst->fs_info,
-			"memmove bogus dst_offset %lu move "
-		       "len %lu dst len %lu", dst_offset, len, dst->len);
+			"memmove bogus dst_offset %lu move len %lu dst len %lu",
+			 dst_offset, len, dst->len);
 		BUG_ON(1);
 	}
 
@@ -5726,13 +5822,15 @@ void memmove_extent_buffer(struct extent_buffer *dst, unsigned long dst_offset,
 	unsigned long src_i;
 
 	if (src_offset + len > dst->len) {
-		btrfs_err(dst->fs_info, "memmove bogus src_offset %lu move "
-		       "len %lu len %lu", src_offset, len, dst->len);
+		btrfs_err(dst->fs_info,
+			  "memmove bogus src_offset %lu move len %lu len %lu",
+			  src_offset, len, dst->len);
 		BUG_ON(1);
 	}
 	if (dst_offset + len > dst->len) {
-		btrfs_err(dst->fs_info, "memmove bogus dst_offset %lu move "
-		       "len %lu len %lu", dst_offset, len, dst->len);
+		btrfs_err(dst->fs_info,
+			  "memmove bogus dst_offset %lu move len %lu len %lu",
+			  dst_offset, len, dst->len);
 		BUG_ON(1);
 	}
 	if (dst_offset < src_offset) {
diff --git a/fs/btrfs/extent_io.h b/fs/btrfs/extent_io.h
index 1baf19c9b79d..ab31d145227e 100644
--- a/fs/btrfs/extent_io.h
+++ b/fs/btrfs/extent_io.h
@@ -20,6 +20,7 @@
 #define EXTENT_DAMAGED		(1U << 14)
 #define EXTENT_NORESERVE	(1U << 15)
 #define EXTENT_QGROUP_RESERVED	(1U << 16)
+#define EXTENT_CLEAR_DATA_RESV	(1U << 17)
 #define EXTENT_IOBITS		(EXTENT_LOCKED | EXTENT_WRITEBACK)
 #define EXTENT_CTLBITS		(EXTENT_DO_ACCOUNTING | EXTENT_FIRST_DELALLOC)
 
@@ -58,21 +59,43 @@
  */
 #define EXTENT_PAGE_PRIVATE 1
 
+/*
+ * The extent buffer bitmap operations are done with byte granularity instead of
+ * word granularity for two reasons:
+ * 1. The bitmaps must be little-endian on disk.
+ * 2. Bitmap items are not guaranteed to be aligned to a word and therefore a
+ *    single word in a bitmap may straddle two pages in the extent buffer.
+ */
+#define BIT_BYTE(nr) ((nr) / BITS_PER_BYTE)
+#define BYTE_MASK ((1 << BITS_PER_BYTE) - 1)
+#define BITMAP_FIRST_BYTE_MASK(start) \
+	((BYTE_MASK << ((start) & (BITS_PER_BYTE - 1))) & BYTE_MASK)
+#define BITMAP_LAST_BYTE_MASK(nbits) \
+	(BYTE_MASK >> (-(nbits) & (BITS_PER_BYTE - 1)))
+
+static inline int le_test_bit(int nr, const u8 *addr)
+{
+	return 1U & (addr[BIT_BYTE(nr)] >> (nr & (BITS_PER_BYTE-1)));
+}
+
+extern void le_bitmap_set(u8 *map, unsigned int start, int len);
+extern void le_bitmap_clear(u8 *map, unsigned int start, int len);
+
 struct extent_state;
 struct btrfs_root;
 struct btrfs_io_bio;
 struct io_failure_record;
 
-typedef	int (extent_submit_bio_hook_t)(struct inode *inode, int rw,
-				       struct bio *bio, int mirror_num,
-				       unsigned long bio_flags, u64 bio_offset);
+typedef	int (extent_submit_bio_hook_t)(struct inode *inode, struct bio *bio,
+				       int mirror_num, unsigned long bio_flags,
+				       u64 bio_offset);
 struct extent_io_ops {
 	int (*fill_delalloc)(struct inode *inode, struct page *locked_page,
 			     u64 start, u64 end, int *page_started,
 			     unsigned long *nr_written);
 	int (*writepage_start_hook)(struct page *page, u64 start, u64 end);
 	extent_submit_bio_hook_t *submit_bio_hook;
-	int (*merge_bio_hook)(int rw, struct page *page, unsigned long offset,
+	int (*merge_bio_hook)(struct page *page, unsigned long offset,
 			      size_t size, struct bio *bio,
 			      unsigned long bio_flags);
 	int (*readpage_io_failed_hook)(struct page *page, int failed_mirror);
@@ -348,7 +371,7 @@ struct extent_buffer *alloc_extent_buffer(struct btrfs_fs_info *fs_info,
 struct extent_buffer *__alloc_dummy_extent_buffer(struct btrfs_fs_info *fs_info,
 						  u64 start, unsigned long len);
 struct extent_buffer *alloc_dummy_extent_buffer(struct btrfs_fs_info *fs_info,
-						u64 start);
+						u64 start, u32 nodesize);
 struct extent_buffer *btrfs_clone_extent_buffer(struct extent_buffer *src);
 struct extent_buffer *find_extent_buffer(struct btrfs_fs_info *fs_info,
 					 u64 start);
@@ -358,7 +381,7 @@ void free_extent_buffer_stale(struct extent_buffer *eb);
 #define WAIT_COMPLETE	1
 #define WAIT_PAGE_LOCK	2
 int read_extent_buffer_pages(struct extent_io_tree *tree,
-			     struct extent_buffer *eb, u64 start, int wait,
+			     struct extent_buffer *eb, int wait,
 			     get_extent_t *get_extent, int mirror_num);
 void wait_on_extent_buffer_writeback(struct extent_buffer *eb);
 
@@ -412,7 +435,7 @@ int map_private_extent_buffer(struct extent_buffer *eb, unsigned long offset,
 void extent_range_clear_dirty_for_io(struct inode *inode, u64 start, u64 end);
 void extent_range_redirty_for_io(struct inode *inode, u64 start, u64 end);
 void extent_clear_unlock_delalloc(struct inode *inode, u64 start, u64 end,
-				 struct page *locked_page,
+				 u64 delalloc_end, struct page *locked_page,
 				 unsigned bits_to_clear,
 				 unsigned long page_ops);
 struct bio *
@@ -468,5 +491,5 @@ noinline u64 find_lock_delalloc_range(struct inode *inode,
 				      u64 *end, u64 max_bytes);
 #endif
 struct extent_buffer *alloc_test_extent_buffer(struct btrfs_fs_info *fs_info,
-					       u64 start);
+					       u64 start, u32 nodesize);
 #endif
diff --git a/fs/btrfs/extent_map.c b/fs/btrfs/extent_map.c
index e0715fcfb11e..26f9ac719d20 100644
--- a/fs/btrfs/extent_map.c
+++ b/fs/btrfs/extent_map.c
@@ -13,7 +13,7 @@ int __init extent_map_init(void)
 {
 	extent_map_cache = kmem_cache_create("btrfs_extent_map",
 			sizeof(struct extent_map), 0,
-			SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD, NULL);
+			SLAB_MEM_SPREAD, NULL);
 	if (!extent_map_cache)
 		return -ENOMEM;
 	return 0;
diff --git a/fs/btrfs/file-item.c b/fs/btrfs/file-item.c
index 62a81ee13a5f..d0d571c47d33 100644
--- a/fs/btrfs/file-item.c
+++ b/fs/btrfs/file-item.c
@@ -250,7 +250,7 @@ static int __btrfs_lookup_bio_sums(struct btrfs_root *root,
 						offset + root->sectorsize - 1,
 						EXTENT_NODATASUM);
 				} else {
-					btrfs_info(BTRFS_I(inode)->root->fs_info,
+					btrfs_info_rl(BTRFS_I(inode)->root->fs_info,
 						   "no csum found for inode %llu start %llu",
 					       btrfs_ino(inode), offset);
 				}
@@ -699,7 +699,7 @@ int btrfs_del_csums(struct btrfs_trans_handle *trans,
 			 */
 			ret = btrfs_split_item(trans, root, path, &key, offset);
 			if (ret && ret != -EAGAIN) {
-				btrfs_abort_transaction(trans, root, ret);
+				btrfs_abort_transaction(trans, ret);
 				goto out;
 			}
 
diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index e0c9bd3fb02d..3a14c87d9c92 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -132,7 +132,7 @@ static int __btrfs_add_inode_defrag(struct inode *inode,
 
 static inline int __need_auto_defrag(struct btrfs_root *root)
 {
-	if (!btrfs_test_opt(root, AUTO_DEFRAG))
+	if (!btrfs_test_opt(root->fs_info, AUTO_DEFRAG))
 		return 0;
 
 	if (btrfs_fs_closing(root->fs_info))
@@ -503,7 +503,7 @@ int btrfs_dirty_pages(struct btrfs_root *root, struct inode *inode,
 
 	end_of_last_block = start_pos + num_bytes - 1;
 	err = btrfs_set_extent_delalloc(inode, start_pos, end_of_last_block,
-					cached);
+					cached, 0);
 	if (err)
 		return err;
 
@@ -950,7 +950,7 @@ delete_extent_item:
 			ret = btrfs_del_items(trans, root, path, del_slot,
 					      del_nr);
 			if (ret) {
-				btrfs_abort_transaction(trans, root, ret);
+				btrfs_abort_transaction(trans, ret);
 				break;
 			}
 
@@ -974,7 +974,7 @@ delete_extent_item:
 		path->slots[0] = del_slot;
 		ret = btrfs_del_items(trans, root, path, del_slot, del_nr);
 		if (ret)
-			btrfs_abort_transaction(trans, root, ret);
+			btrfs_abort_transaction(trans, ret);
 	}
 
 	leaf = path->nodes[0];
@@ -1110,13 +1110,25 @@ again:
 
 	leaf = path->nodes[0];
 	btrfs_item_key_to_cpu(leaf, &key, path->slots[0]);
-	BUG_ON(key.objectid != ino || key.type != BTRFS_EXTENT_DATA_KEY);
+	if (key.objectid != ino ||
+	    key.type != BTRFS_EXTENT_DATA_KEY) {
+		ret = -EINVAL;
+		btrfs_abort_transaction(trans, ret);
+		goto out;
+	}
 	fi = btrfs_item_ptr(leaf, path->slots[0],
 			    struct btrfs_file_extent_item);
-	BUG_ON(btrfs_file_extent_type(leaf, fi) !=
-	       BTRFS_FILE_EXTENT_PREALLOC);
+	if (btrfs_file_extent_type(leaf, fi) != BTRFS_FILE_EXTENT_PREALLOC) {
+		ret = -EINVAL;
+		btrfs_abort_transaction(trans, ret);
+		goto out;
+	}
 	extent_end = key.offset + btrfs_file_extent_num_bytes(leaf, fi);
-	BUG_ON(key.offset > start || extent_end < end);
+	if (key.offset > start || extent_end < end) {
+		ret = -EINVAL;
+		btrfs_abort_transaction(trans, ret);
+		goto out;
+	}
 
 	bytenr = btrfs_file_extent_disk_bytenr(leaf, fi);
 	num_bytes = btrfs_file_extent_disk_num_bytes(leaf, fi);
@@ -1190,7 +1202,7 @@ again:
 			goto again;
 		}
 		if (ret < 0) {
-			btrfs_abort_transaction(trans, root, ret);
+			btrfs_abort_transaction(trans, ret);
 			goto out;
 		}
 
@@ -1213,12 +1225,19 @@ again:
 		ret = btrfs_inc_extent_ref(trans, root, bytenr, num_bytes, 0,
 					   root->root_key.objectid,
 					   ino, orig_offset);
-		BUG_ON(ret); /* -ENOMEM */
+		if (ret) {
+			btrfs_abort_transaction(trans, ret);
+			goto out;
+		}
 
 		if (split == start) {
 			key.offset = start;
 		} else {
-			BUG_ON(start != key.offset);
+			if (start != key.offset) {
+				ret = -EINVAL;
+				btrfs_abort_transaction(trans, ret);
+				goto out;
+			}
 			path->slots[0]--;
 			extent_end = end;
 		}
@@ -1240,7 +1259,10 @@ again:
 		ret = btrfs_free_extent(trans, root, bytenr, num_bytes,
 					0, root->root_key.objectid,
 					ino, orig_offset);
-		BUG_ON(ret); /* -ENOMEM */
+		if (ret) {
+			btrfs_abort_transaction(trans, ret);
+			goto out;
+		}
 	}
 	other_start = 0;
 	other_end = start;
@@ -1257,7 +1279,10 @@ again:
 		ret = btrfs_free_extent(trans, root, bytenr, num_bytes,
 					0, root->root_key.objectid,
 					ino, orig_offset);
-		BUG_ON(ret); /* -ENOMEM */
+		if (ret) {
+			btrfs_abort_transaction(trans, ret);
+			goto out;
+		}
 	}
 	if (del_nr == 0) {
 		fi = btrfs_item_ptr(leaf, path->slots[0],
@@ -1278,7 +1303,7 @@ again:
 
 		ret = btrfs_del_items(trans, root, path, del_slot, del_nr);
 		if (ret < 0) {
-			btrfs_abort_transaction(trans, root, ret);
+			btrfs_abort_transaction(trans, ret);
 			goto out;
 		}
 	}
@@ -1534,30 +1559,30 @@ static noinline ssize_t __btrfs_buffered_write(struct file *file,
 		reserve_bytes = round_up(write_bytes + sector_offset,
 				root->sectorsize);
 
-		if ((BTRFS_I(inode)->flags & (BTRFS_INODE_NODATACOW |
-					      BTRFS_INODE_PREALLOC)) &&
-		    check_can_nocow(inode, pos, &write_bytes) > 0) {
-			/*
-			 * For nodata cow case, no need to reserve
-			 * data space.
-			 */
-			only_release_metadata = true;
-			/*
-			 * our prealloc extent may be smaller than
-			 * write_bytes, so scale down.
-			 */
-			num_pages = DIV_ROUND_UP(write_bytes + offset,
-						 PAGE_SIZE);
-			reserve_bytes = round_up(write_bytes + sector_offset,
-					root->sectorsize);
-			goto reserve_metadata;
-		}
-
 		ret = btrfs_check_data_free_space(inode, pos, write_bytes);
-		if (ret < 0)
-			break;
+		if (ret < 0) {
+			if ((BTRFS_I(inode)->flags & (BTRFS_INODE_NODATACOW |
+						      BTRFS_INODE_PREALLOC)) &&
+			    check_can_nocow(inode, pos, &write_bytes) > 0) {
+				/*
+				 * For nodata cow case, no need to reserve
+				 * data space.
+				 */
+				only_release_metadata = true;
+				/*
+				 * our prealloc extent may be smaller than
+				 * write_bytes, so scale down.
+				 */
+				num_pages = DIV_ROUND_UP(write_bytes + offset,
+							 PAGE_SIZE);
+				reserve_bytes = round_up(write_bytes +
+							 sector_offset,
+							 root->sectorsize);
+			} else {
+				break;
+			}
+		}
 
-reserve_metadata:
 		ret = btrfs_delalloc_reserve_metadata(inode, reserve_bytes);
 		if (ret) {
 			if (!only_release_metadata)
@@ -1629,13 +1654,11 @@ again:
 		 * managed to copy.
 		 */
 		if (num_sectors > dirty_sectors) {
-			/*
-			 * we round down because we don't want to count
-			 * any partial blocks actually sent through the
-			 * IO machines
-			 */
-			release_bytes = round_down(release_bytes - copied,
-				      root->sectorsize);
+
+			/* release everything except the sectors we dirtied */
+			release_bytes -= dirty_sectors <<
+				root->fs_info->sb->s_blocksize_bits;
+
 			if (copied > 0) {
 				spin_lock(&BTRFS_I(inode)->lock);
 				BTRFS_I(inode)->outstanding_extents++;
@@ -1759,7 +1782,7 @@ static void update_time_for_write(struct inode *inode)
 	if (IS_NOCMTIME(inode))
 		return;
 
-	now = current_fs_time(inode->i_sb);
+	now = current_time(inode);
 	if (!timespec_equal(&inode->i_mtime, &now))
 		inode->i_mtime = now;
 
@@ -2035,6 +2058,14 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
 		 */
 		clear_bit(BTRFS_INODE_NEEDS_FULL_SYNC,
 			  &BTRFS_I(inode)->runtime_flags);
+		/*
+		 * An ordered extent might have started before and completed
+		 * already with io errors, in which case the inode was not
+		 * updated and we end up here. So check the inode's mapping
+		 * flags for any errors that might have happened while doing
+		 * writeback of file data.
+		 */
+		ret = filemap_check_errors(inode->i_mapping);
 		inode_unlock(inode);
 		goto out;
 	}
@@ -2064,7 +2095,7 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
 	}
 	trans->sync = true;
 
-	btrfs_init_log_ctx(&ctx);
+	btrfs_init_log_ctx(&ctx, inode);
 
 	ret = btrfs_log_dentry_safe(trans, root, dentry, start, end, &ctx);
 	if (ret < 0) {
@@ -2479,7 +2510,7 @@ static int btrfs_punch_hole(struct inode *inode, loff_t offset, loff_t len)
 	}
 
 	ret = btrfs_block_rsv_migrate(&root->fs_info->trans_block_rsv, rsv,
-				      min_size);
+				      min_size, 0);
 	BUG_ON(ret);
 	trans->block_rsv = rsv;
 
@@ -2522,7 +2553,7 @@ static int btrfs_punch_hole(struct inode *inode, loff_t offset, loff_t len)
 		}
 
 		ret = btrfs_block_rsv_migrate(&root->fs_info->trans_block_rsv,
-					      rsv, min_size);
+					      rsv, min_size, 0);
 		BUG_ON(ret);	/* shouldn't happen */
 		trans->block_rsv = rsv;
 
@@ -2572,7 +2603,7 @@ out_trans:
 		goto out_free;
 
 	inode_inc_iversion(inode);
-	inode->i_mtime = inode->i_ctime = current_fs_time(inode->i_sb);
+	inode->i_mtime = inode->i_ctime = current_time(inode);
 
 	trans->block_rsv = &root->fs_info->trans_block_rsv;
 	ret = btrfs_update_inode(trans, root, inode);
@@ -2669,6 +2700,7 @@ static long btrfs_fallocate(struct file *file, int mode,
 
 	alloc_start = round_down(offset, blocksize);
 	alloc_end = round_up(offset + len, blocksize);
+	cur_offset = alloc_start;
 
 	/* Make sure we aren't being give some crap mode */
 	if (mode & ~(FALLOC_FL_KEEP_SIZE | FALLOC_FL_PUNCH_HOLE))
@@ -2761,7 +2793,6 @@ static long btrfs_fallocate(struct file *file, int mode,
 
 	/* First, check if we exceed the qgroup limit */
 	INIT_LIST_HEAD(&reserve_list);
-	cur_offset = alloc_start;
 	while (1) {
 		em = btrfs_get_extent(inode, NULL, 0, cur_offset,
 				      alloc_end - cur_offset, 0);
@@ -2788,6 +2819,14 @@ static long btrfs_fallocate(struct file *file, int mode,
 					last_byte - cur_offset);
 			if (ret < 0)
 				break;
+		} else {
+			/*
+			 * Do not need to reserve unwritten extent for this
+			 * range, free reserved data space first, otherwise
+			 * it'll result in false ENOSPC error.
+			 */
+			btrfs_free_reserved_data_space(inode, cur_offset,
+				last_byte - cur_offset);
 		}
 		free_extent_map(em);
 		cur_offset = last_byte;
@@ -2805,6 +2844,9 @@ static long btrfs_fallocate(struct file *file, int mode,
 					range->start,
 					range->len, 1 << inode->i_blkbits,
 					offset + len, &alloc_hint);
+		else
+			btrfs_free_reserved_data_space(inode, range->start,
+						       range->len);
 		list_del(&range->list);
 		kfree(range);
 	}
@@ -2825,7 +2867,7 @@ static long btrfs_fallocate(struct file *file, int mode,
 		if (IS_ERR(trans)) {
 			ret = PTR_ERR(trans);
 		} else {
-			inode->i_ctime = current_fs_time(inode->i_sb);
+			inode->i_ctime = current_time(inode);
 			i_size_write(inode, actual_end);
 			btrfs_ordered_update_i_size(inode, actual_end, NULL);
 			ret = btrfs_update_inode(trans, root, inode);
@@ -2839,18 +2881,11 @@ out_unlock:
 	unlock_extent_cached(&BTRFS_I(inode)->io_tree, alloc_start, locked_end,
 			     &cached_state, GFP_KERNEL);
 out:
-	/*
-	 * As we waited the extent range, the data_rsv_map must be empty
-	 * in the range, as written data range will be released from it.
-	 * And for prealloacted extent, it will also be released when
-	 * its metadata is written.
-	 * So this is completely used as cleanup.
-	 */
-	btrfs_qgroup_free_data(inode, alloc_start, alloc_end - alloc_start);
 	inode_unlock(inode);
 	/* Let go of our reservation. */
-	btrfs_free_reserved_data_space(inode, alloc_start,
-				       alloc_end - alloc_start);
+	if (ret != 0)
+		btrfs_free_reserved_data_space(inode, alloc_start,
+				       alloc_end - cur_offset);
 	return ret;
 }
 
@@ -2977,7 +3012,7 @@ int btrfs_auto_defrag_init(void)
 {
 	btrfs_inode_defrag_cachep = kmem_cache_create("btrfs_inode_defrag",
 					sizeof(struct inode_defrag), 0,
-					SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD,
+					SLAB_MEM_SPREAD,
 					NULL);
 	if (!btrfs_inode_defrag_cachep)
 		return -ENOMEM;
diff --git a/fs/btrfs/free-space-cache.c b/fs/btrfs/free-space-cache.c
index c6dc1183f542..e4b48f377d3a 100644
--- a/fs/btrfs/free-space-cache.c
+++ b/fs/btrfs/free-space-cache.c
@@ -29,7 +29,7 @@
 #include "inode-map.h"
 #include "volumes.h"
 
-#define BITS_PER_BITMAP		(PAGE_SIZE * 8)
+#define BITS_PER_BITMAP		(PAGE_SIZE * 8UL)
 #define MAX_CACHE_BYTES_PER_GIG	SZ_32K
 
 struct btrfs_trim_range {
@@ -280,7 +280,7 @@ fail:
 	if (locked)
 		mutex_unlock(&trans->transaction->cache_write_mutex);
 	if (ret)
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 
 	return ret;
 }
@@ -716,8 +716,7 @@ static int __load_free_space_cache(struct btrfs_root *root, struct inode *inode,
 
 	if (BTRFS_I(inode)->generation != generation) {
 		btrfs_err(root->fs_info,
-			"free space inode generation (%llu) "
-			"did not match free space cache generation (%llu)",
+			"free space inode generation (%llu) did not match free space cache generation (%llu)",
 			BTRFS_I(inode)->generation, generation);
 		return 0;
 	}
@@ -879,8 +878,9 @@ int load_free_space_cache(struct btrfs_fs_info *fs_info,
 
 	if (!matched) {
 		__btrfs_remove_free_space_cache(ctl);
-		btrfs_warn(fs_info, "block group %llu has wrong amount of free space",
-			block_group->key.objectid);
+		btrfs_warn(fs_info,
+			   "block group %llu has wrong amount of free space",
+			   block_group->key.objectid);
 		ret = -1;
 	}
 out:
@@ -891,8 +891,9 @@ out:
 		spin_unlock(&block_group->lock);
 		ret = 0;
 
-		btrfs_warn(fs_info, "failed to load free space cache for block group %llu, rebuilding it now",
-			block_group->key.objectid);
+		btrfs_warn(fs_info,
+			   "failed to load free space cache for block group %llu, rebuilding it now",
+			   block_group->key.objectid);
 	}
 
 	iput(inode);
@@ -1415,11 +1416,11 @@ static inline u64 offset_to_bitmap(struct btrfs_free_space_ctl *ctl,
 				   u64 offset)
 {
 	u64 bitmap_start;
-	u32 bytes_per_bitmap;
+	u64 bytes_per_bitmap;
 
 	bytes_per_bitmap = BITS_PER_BITMAP * ctl->unit;
 	bitmap_start = offset - ctl->start;
-	bitmap_start = div_u64(bitmap_start, bytes_per_bitmap);
+	bitmap_start = div64_u64(bitmap_start, bytes_per_bitmap);
 	bitmap_start *= bytes_per_bitmap;
 	bitmap_start += ctl->start;
 
@@ -1638,10 +1639,10 @@ static void recalculate_thresholds(struct btrfs_free_space_ctl *ctl)
 	u64 bitmap_bytes;
 	u64 extent_bytes;
 	u64 size = block_group->key.offset;
-	u32 bytes_per_bg = BITS_PER_BITMAP * ctl->unit;
-	u32 max_bitmaps = div_u64(size + bytes_per_bg - 1, bytes_per_bg);
+	u64 bytes_per_bg = BITS_PER_BITMAP * ctl->unit;
+	u64 max_bitmaps = div64_u64(size + bytes_per_bg - 1, bytes_per_bg);
 
-	max_bitmaps = max_t(u32, max_bitmaps, 1);
+	max_bitmaps = max_t(u64, max_bitmaps, 1);
 
 	ASSERT(ctl->total_bitmaps <= max_bitmaps);
 
@@ -1660,7 +1661,7 @@ static void recalculate_thresholds(struct btrfs_free_space_ctl *ctl)
 	 * sure we don't go over our overall goal of MAX_CACHE_BYTES_PER_GIG as
 	 * we add more bitmaps.
 	 */
-	bitmap_bytes = (ctl->total_bitmaps + 1) * PAGE_SIZE;
+	bitmap_bytes = (ctl->total_bitmaps + 1) * ctl->unit;
 
 	if (bitmap_bytes >= max_bytes) {
 		ctl->extents_thresh = 0;
@@ -2298,7 +2299,8 @@ static void steal_from_bitmap(struct btrfs_free_space_ctl *ctl,
 	}
 }
 
-int __btrfs_add_free_space(struct btrfs_free_space_ctl *ctl,
+int __btrfs_add_free_space(struct btrfs_fs_info *fs_info,
+			   struct btrfs_free_space_ctl *ctl,
 			   u64 offset, u64 bytes)
 {
 	struct btrfs_free_space *info;
@@ -2345,7 +2347,7 @@ out:
 	spin_unlock(&ctl->tree_lock);
 
 	if (ret) {
-		printk(KERN_CRIT "BTRFS: unable to add free space :%d\n", ret);
+		btrfs_crit(fs_info, "unable to add free space :%d", ret);
 		ASSERT(ret != -EEXIST);
 	}
 
@@ -2621,7 +2623,8 @@ out:
 	spin_unlock(&ctl->tree_lock);
 
 	if (align_gap_len)
-		__btrfs_add_free_space(ctl, align_gap, align_gap_len);
+		__btrfs_add_free_space(block_group->fs_info, ctl,
+				       align_gap, align_gap_len);
 	return ret;
 }
 
@@ -3026,7 +3029,7 @@ int btrfs_find_space_cluster(struct btrfs_root *root,
 	 * For metadata, allow allocates with smaller extents.  For
 	 * data, keep it dense.
 	 */
-	if (btrfs_test_opt(root, SSD_SPREAD)) {
+	if (btrfs_test_opt(root->fs_info, SSD_SPREAD)) {
 		cont1_bytes = min_bytes = bytes + empty_size;
 	} else if (block_group->flags & BTRFS_BLOCK_GROUP_METADATA) {
 		cont1_bytes = bytes;
@@ -3470,7 +3473,7 @@ int load_free_ino_cache(struct btrfs_fs_info *fs_info, struct btrfs_root *root)
 	int ret = 0;
 	u64 root_gen = btrfs_root_generation(&root->root_item);
 
-	if (!btrfs_test_opt(root, INODE_MAP_CACHE))
+	if (!btrfs_test_opt(root->fs_info, INODE_MAP_CACHE))
 		return 0;
 
 	/*
@@ -3514,7 +3517,7 @@ int btrfs_write_out_ino_cache(struct btrfs_root *root,
 	struct btrfs_io_ctl io_ctl;
 	bool release_metadata = true;
 
-	if (!btrfs_test_opt(root, INODE_MAP_CACHE))
+	if (!btrfs_test_opt(root->fs_info, INODE_MAP_CACHE))
 		return 0;
 
 	memset(&io_ctl, 0, sizeof(io_ctl));
@@ -3662,7 +3665,7 @@ have_info:
 			if (tmp->offset + tmp->bytes < offset)
 				break;
 			if (offset + bytes < tmp->offset) {
-				n = rb_prev(&info->offset_index);
+				n = rb_prev(&tmp->offset_index);
 				continue;
 			}
 			info = tmp;
@@ -3676,7 +3679,7 @@ have_info:
 			if (offset + bytes < tmp->offset)
 				break;
 			if (tmp->offset + tmp->bytes < offset) {
-				n = rb_next(&info->offset_index);
+				n = rb_next(&tmp->offset_index);
 				continue;
 			}
 			info = tmp;
diff --git a/fs/btrfs/free-space-cache.h b/fs/btrfs/free-space-cache.h
index 3af651c2bbc7..363fdd955e5d 100644
--- a/fs/btrfs/free-space-cache.h
+++ b/fs/btrfs/free-space-cache.h
@@ -89,13 +89,15 @@ int btrfs_write_out_ino_cache(struct btrfs_root *root,
 			      struct inode *inode);
 
 void btrfs_init_free_space_ctl(struct btrfs_block_group_cache *block_group);
-int __btrfs_add_free_space(struct btrfs_free_space_ctl *ctl,
+int __btrfs_add_free_space(struct btrfs_fs_info *fs_info,
+			   struct btrfs_free_space_ctl *ctl,
 			   u64 bytenr, u64 size);
 static inline int
 btrfs_add_free_space(struct btrfs_block_group_cache *block_group,
 		     u64 bytenr, u64 size)
 {
-	return __btrfs_add_free_space(block_group->free_space_ctl,
+	return __btrfs_add_free_space(block_group->fs_info,
+				      block_group->free_space_ctl,
 				      bytenr, size);
 }
 int btrfs_remove_free_space(struct btrfs_block_group_cache *block_group,
diff --git a/fs/btrfs/free-space-tree.c b/fs/btrfs/free-space-tree.c
index 53dbeaf6ce94..57401b474ec6 100644
--- a/fs/btrfs/free-space-tree.c
+++ b/fs/btrfs/free-space-tree.c
@@ -107,7 +107,7 @@ search_free_space_info(struct btrfs_trans_handle *trans,
 	if (ret < 0)
 		return ERR_PTR(ret);
 	if (ret != 0) {
-		btrfs_warn(fs_info, "missing free space info for %llu\n",
+		btrfs_warn(fs_info, "missing free space info for %llu",
 			   block_group->key.objectid);
 		ASSERT(0);
 		return ERR_PTR(-ENOENT);
@@ -151,7 +151,7 @@ static inline u32 free_space_bitmap_size(u64 size, u32 sectorsize)
 	return DIV_ROUND_UP((u32)div_u64(size, sectorsize), BITS_PER_BYTE);
 }
 
-static unsigned long *alloc_bitmap(u32 bitmap_size)
+static u8 *alloc_bitmap(u32 bitmap_size)
 {
 	void *mem;
 
@@ -180,8 +180,7 @@ int convert_free_space_to_bitmaps(struct btrfs_trans_handle *trans,
 	struct btrfs_free_space_info *info;
 	struct btrfs_key key, found_key;
 	struct extent_buffer *leaf;
-	unsigned long *bitmap;
-	char *bitmap_cursor;
+	u8 *bitmap, *bitmap_cursor;
 	u64 start, end;
 	u64 bitmap_range, i;
 	u32 bitmap_size, flags, expected_extent_count;
@@ -231,7 +230,7 @@ int convert_free_space_to_bitmaps(struct btrfs_trans_handle *trans,
 						block_group->sectorsize);
 				last = div_u64(found_key.objectid + found_key.offset - start,
 					       block_group->sectorsize);
-				bitmap_set(bitmap, first, last - first);
+				le_bitmap_set(bitmap, first, last - first);
 
 				extent_count++;
 				nr++;
@@ -261,7 +260,8 @@ int convert_free_space_to_bitmaps(struct btrfs_trans_handle *trans,
 	btrfs_release_path(path);
 
 	if (extent_count != expected_extent_count) {
-		btrfs_err(fs_info, "incorrect extent count for %llu; counted %u, expected %u",
+		btrfs_err(fs_info,
+			  "incorrect extent count for %llu; counted %u, expected %u",
 			  block_group->key.objectid, extent_count,
 			  expected_extent_count);
 		ASSERT(0);
@@ -269,7 +269,7 @@ int convert_free_space_to_bitmaps(struct btrfs_trans_handle *trans,
 		goto out;
 	}
 
-	bitmap_cursor = (char *)bitmap;
+	bitmap_cursor = bitmap;
 	bitmap_range = block_group->sectorsize * BTRFS_FREE_SPACE_BITMAP_BITS;
 	i = start;
 	while (i < end) {
@@ -305,7 +305,7 @@ int convert_free_space_to_bitmaps(struct btrfs_trans_handle *trans,
 out:
 	kvfree(bitmap);
 	if (ret)
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 	return ret;
 }
 
@@ -318,7 +318,7 @@ int convert_free_space_to_extents(struct btrfs_trans_handle *trans,
 	struct btrfs_free_space_info *info;
 	struct btrfs_key key, found_key;
 	struct extent_buffer *leaf;
-	unsigned long *bitmap;
+	u8 *bitmap;
 	u64 start, end;
 	/* Initialize to silence GCC. */
 	u64 extent_start = 0;
@@ -362,7 +362,7 @@ int convert_free_space_to_extents(struct btrfs_trans_handle *trans,
 				break;
 			} else if (found_key.type == BTRFS_FREE_SPACE_BITMAP_KEY) {
 				unsigned long ptr;
-				char *bitmap_cursor;
+				u8 *bitmap_cursor;
 				u32 bitmap_pos, data_size;
 
 				ASSERT(found_key.objectid >= start);
@@ -372,7 +372,7 @@ int convert_free_space_to_extents(struct btrfs_trans_handle *trans,
 				bitmap_pos = div_u64(found_key.objectid - start,
 						     block_group->sectorsize *
 						     BITS_PER_BYTE);
-				bitmap_cursor = ((char *)bitmap) + bitmap_pos;
+				bitmap_cursor = bitmap + bitmap_pos;
 				data_size = free_space_bitmap_size(found_key.offset,
 								   block_group->sectorsize);
 
@@ -409,7 +409,7 @@ int convert_free_space_to_extents(struct btrfs_trans_handle *trans,
 	offset = start;
 	bitnr = 0;
 	while (offset < end) {
-		bit = !!test_bit(bitnr, bitmap);
+		bit = !!le_test_bit(bitnr, bitmap);
 		if (prev_bit == 0 && bit == 1) {
 			extent_start = offset;
 		} else if (prev_bit == 1 && bit == 0) {
@@ -442,7 +442,8 @@ int convert_free_space_to_extents(struct btrfs_trans_handle *trans,
 	}
 
 	if (extent_count != expected_extent_count) {
-		btrfs_err(fs_info, "incorrect extent count for %llu; counted %u, expected %u",
+		btrfs_err(fs_info,
+			  "incorrect extent count for %llu; counted %u, expected %u",
 			  block_group->key.objectid, extent_count,
 			  expected_extent_count);
 		ASSERT(0);
@@ -454,7 +455,7 @@ int convert_free_space_to_extents(struct btrfs_trans_handle *trans,
 out:
 	kvfree(bitmap);
 	if (ret)
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 	return ret;
 }
 
@@ -851,7 +852,7 @@ int remove_from_free_space_tree(struct btrfs_trans_handle *trans,
 out:
 	btrfs_free_path(path);
 	if (ret)
-		btrfs_abort_transaction(trans, fs_info->free_space_root, ret);
+		btrfs_abort_transaction(trans, ret);
 	return ret;
 }
 
@@ -1047,7 +1048,7 @@ int add_to_free_space_tree(struct btrfs_trans_handle *trans,
 out:
 	btrfs_free_path(path);
 	if (ret)
-		btrfs_abort_transaction(trans, fs_info->free_space_root, ret);
+		btrfs_abort_transaction(trans, ret);
 	return ret;
 }
 
@@ -1163,7 +1164,7 @@ int btrfs_create_free_space_tree(struct btrfs_fs_info *fs_info)
 	if (IS_ERR(trans))
 		return PTR_ERR(trans);
 
-	fs_info->creating_free_space_tree = 1;
+	set_bit(BTRFS_FS_CREATING_FREE_SPACE_TREE, &fs_info->flags);
 	free_space_root = btrfs_create_tree(trans, fs_info,
 					    BTRFS_FREE_SPACE_TREE_OBJECTID);
 	if (IS_ERR(free_space_root)) {
@@ -1183,7 +1184,8 @@ int btrfs_create_free_space_tree(struct btrfs_fs_info *fs_info)
 	}
 
 	btrfs_set_fs_compat_ro(fs_info, FREE_SPACE_TREE);
-	fs_info->creating_free_space_tree = 0;
+	btrfs_set_fs_compat_ro(fs_info, FREE_SPACE_TREE_VALID);
+	clear_bit(BTRFS_FS_CREATING_FREE_SPACE_TREE, &fs_info->flags);
 
 	ret = btrfs_commit_transaction(trans, tree_root);
 	if (ret)
@@ -1192,8 +1194,8 @@ int btrfs_create_free_space_tree(struct btrfs_fs_info *fs_info)
 	return 0;
 
 abort:
-	fs_info->creating_free_space_tree = 0;
-	btrfs_abort_transaction(trans, tree_root, ret);
+	clear_bit(BTRFS_FS_CREATING_FREE_SPACE_TREE, &fs_info->flags);
+	btrfs_abort_transaction(trans, ret);
 	btrfs_end_transaction(trans, tree_root);
 	return ret;
 }
@@ -1251,6 +1253,7 @@ int btrfs_clear_free_space_tree(struct btrfs_fs_info *fs_info)
 		return PTR_ERR(trans);
 
 	btrfs_clear_fs_compat_ro(fs_info, FREE_SPACE_TREE);
+	btrfs_clear_fs_compat_ro(fs_info, FREE_SPACE_TREE_VALID);
 	fs_info->free_space_root = NULL;
 
 	ret = clear_free_space_tree(trans, free_space_root);
@@ -1280,7 +1283,7 @@ int btrfs_clear_free_space_tree(struct btrfs_fs_info *fs_info)
 	return 0;
 
 abort:
-	btrfs_abort_transaction(trans, tree_root, ret);
+	btrfs_abort_transaction(trans, ret);
 	btrfs_end_transaction(trans, tree_root);
 	return ret;
 }
@@ -1333,7 +1336,7 @@ out:
 	btrfs_free_path(path);
 	mutex_unlock(&block_group->free_space_lock);
 	if (ret)
-		btrfs_abort_transaction(trans, fs_info->free_space_root, ret);
+		btrfs_abort_transaction(trans, ret);
 	return ret;
 }
 
@@ -1410,7 +1413,7 @@ int remove_block_group_free_space(struct btrfs_trans_handle *trans,
 out:
 	btrfs_free_path(path);
 	if (ret)
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 	return ret;
 }
 
@@ -1480,7 +1483,8 @@ static int load_free_space_bitmaps(struct btrfs_caching_control *caching_ctl,
 	}
 
 	if (extent_count != expected_extent_count) {
-		btrfs_err(fs_info, "incorrect extent count for %llu; counted %u, expected %u",
+		btrfs_err(fs_info,
+			  "incorrect extent count for %llu; counted %u, expected %u",
 			  block_group->key.objectid, extent_count,
 			  expected_extent_count);
 		ASSERT(0);
@@ -1542,7 +1546,8 @@ static int load_free_space_extents(struct btrfs_caching_control *caching_ctl,
 	}
 
 	if (extent_count != expected_extent_count) {
-		btrfs_err(fs_info, "incorrect extent count for %llu; counted %u, expected %u",
+		btrfs_err(fs_info,
+			  "incorrect extent count for %llu; counted %u, expected %u",
 			  block_group->key.objectid, extent_count,
 			  expected_extent_count);
 		ASSERT(0);
diff --git a/fs/btrfs/hash.c b/fs/btrfs/hash.c
index aae520b2aee5..a97fdc156a03 100644
--- a/fs/btrfs/hash.c
+++ b/fs/btrfs/hash.c
@@ -24,6 +24,11 @@ int __init btrfs_hash_init(void)
 	return PTR_ERR_OR_ZERO(tfm);
 }
 
+const char* btrfs_crc32c_impl(void)
+{
+	return crypto_tfm_alg_driver_name(crypto_shash_tfm(tfm));
+}
+
 void btrfs_hash_exit(void)
 {
 	crypto_free_shash(tfm);
diff --git a/fs/btrfs/hash.h b/fs/btrfs/hash.h
index 118a2316e5d3..c3a2ec554361 100644
--- a/fs/btrfs/hash.h
+++ b/fs/btrfs/hash.h
@@ -22,6 +22,7 @@
 int __init btrfs_hash_init(void);
 
 void btrfs_hash_exit(void);
+const char* btrfs_crc32c_impl(void);
 
 u32 btrfs_crc32c(u32 crc, const void *address, unsigned int length);
 
diff --git a/fs/btrfs/inode-map.c b/fs/btrfs/inode-map.c
index 70107f7c9307..d27014b8bf72 100644
--- a/fs/btrfs/inode-map.c
+++ b/fs/btrfs/inode-map.c
@@ -38,7 +38,7 @@ static int caching_kthread(void *data)
 	int slot;
 	int ret;
 
-	if (!btrfs_test_opt(root, INODE_MAP_CACHE))
+	if (!btrfs_test_opt(root->fs_info, INODE_MAP_CACHE))
 		return 0;
 
 	path = btrfs_alloc_path();
@@ -104,7 +104,7 @@ again:
 			break;
 
 		if (last != (u64)-1 && last + 1 != key.objectid) {
-			__btrfs_add_free_space(ctl, last + 1,
+			__btrfs_add_free_space(fs_info, ctl, last + 1,
 					       key.objectid - last - 1);
 			wake_up(&root->ino_cache_wait);
 		}
@@ -115,7 +115,7 @@ next:
 	}
 
 	if (last < root->highest_objectid - 1) {
-		__btrfs_add_free_space(ctl, last + 1,
+		__btrfs_add_free_space(fs_info, ctl, last + 1,
 				       root->highest_objectid - last - 1);
 	}
 
@@ -136,12 +136,13 @@ out:
 
 static void start_caching(struct btrfs_root *root)
 {
+	struct btrfs_fs_info *fs_info = root->fs_info;
 	struct btrfs_free_space_ctl *ctl = root->free_ino_ctl;
 	struct task_struct *tsk;
 	int ret;
 	u64 objectid;
 
-	if (!btrfs_test_opt(root, INODE_MAP_CACHE))
+	if (!btrfs_test_opt(fs_info, INODE_MAP_CACHE))
 		return;
 
 	spin_lock(&root->ino_cache_lock);
@@ -153,7 +154,7 @@ static void start_caching(struct btrfs_root *root)
 	root->ino_cache_state = BTRFS_CACHE_STARTED;
 	spin_unlock(&root->ino_cache_lock);
 
-	ret = load_free_ino_cache(root->fs_info, root);
+	ret = load_free_ino_cache(fs_info, root);
 	if (ret == 1) {
 		spin_lock(&root->ino_cache_lock);
 		root->ino_cache_state = BTRFS_CACHE_FINISHED;
@@ -170,22 +171,22 @@ static void start_caching(struct btrfs_root *root)
 	 */
 	ret = btrfs_find_free_objectid(root, &objectid);
 	if (!ret && objectid <= BTRFS_LAST_FREE_OBJECTID) {
-		__btrfs_add_free_space(ctl, objectid,
+		__btrfs_add_free_space(fs_info, ctl, objectid,
 				       BTRFS_LAST_FREE_OBJECTID - objectid + 1);
 	}
 
 	tsk = kthread_run(caching_kthread, root, "btrfs-ino-cache-%llu",
 			  root->root_key.objectid);
 	if (IS_ERR(tsk)) {
-		btrfs_warn(root->fs_info, "failed to start inode caching task");
-		btrfs_clear_pending_and_info(root->fs_info, INODE_MAP_CACHE,
+		btrfs_warn(fs_info, "failed to start inode caching task");
+		btrfs_clear_pending_and_info(fs_info, INODE_MAP_CACHE,
 				"disabling inode map caching");
 	}
 }
 
 int btrfs_find_free_ino(struct btrfs_root *root, u64 *objectid)
 {
-	if (!btrfs_test_opt(root, INODE_MAP_CACHE))
+	if (!btrfs_test_opt(root->fs_info, INODE_MAP_CACHE))
 		return btrfs_find_free_objectid(root, objectid);
 
 again:
@@ -209,28 +210,29 @@ again:
 
 void btrfs_return_ino(struct btrfs_root *root, u64 objectid)
 {
+	struct btrfs_fs_info *fs_info = root->fs_info;
 	struct btrfs_free_space_ctl *pinned = root->free_ino_pinned;
 
-	if (!btrfs_test_opt(root, INODE_MAP_CACHE))
+	if (!btrfs_test_opt(fs_info, INODE_MAP_CACHE))
 		return;
 again:
 	if (root->ino_cache_state == BTRFS_CACHE_FINISHED) {
-		__btrfs_add_free_space(pinned, objectid, 1);
+		__btrfs_add_free_space(fs_info, pinned, objectid, 1);
 	} else {
-		down_write(&root->fs_info->commit_root_sem);
+		down_write(&fs_info->commit_root_sem);
 		spin_lock(&root->ino_cache_lock);
 		if (root->ino_cache_state == BTRFS_CACHE_FINISHED) {
 			spin_unlock(&root->ino_cache_lock);
-			up_write(&root->fs_info->commit_root_sem);
+			up_write(&fs_info->commit_root_sem);
 			goto again;
 		}
 		spin_unlock(&root->ino_cache_lock);
 
 		start_caching(root);
 
-		__btrfs_add_free_space(pinned, objectid, 1);
+		__btrfs_add_free_space(fs_info, pinned, objectid, 1);
 
-		up_write(&root->fs_info->commit_root_sem);
+		up_write(&fs_info->commit_root_sem);
 	}
 }
 
@@ -251,7 +253,7 @@ void btrfs_unpin_free_ino(struct btrfs_root *root)
 	struct rb_node *n;
 	u64 count;
 
-	if (!btrfs_test_opt(root, INODE_MAP_CACHE))
+	if (!btrfs_test_opt(root->fs_info, INODE_MAP_CACHE))
 		return;
 
 	while (1) {
@@ -277,7 +279,8 @@ void btrfs_unpin_free_ino(struct btrfs_root *root)
 		rb_erase(&info->offset_index, rbroot);
 		spin_unlock(rbroot_lock);
 		if (add_to_ctl)
-			__btrfs_add_free_space(ctl, info->offset, count);
+			__btrfs_add_free_space(root->fs_info, ctl,
+					       info->offset, count);
 		kmem_cache_free(btrfs_free_space_cachep, info);
 	}
 }
@@ -412,7 +415,7 @@ int btrfs_save_ino_cache(struct btrfs_root *root,
 	if (btrfs_root_refs(&root->root_item) == 0)
 		return 0;
 
-	if (!btrfs_test_opt(root, INODE_MAP_CACHE))
+	if (!btrfs_test_opt(root->fs_info, INODE_MAP_CACHE))
 		return 0;
 
 	path = btrfs_alloc_path();
@@ -458,7 +461,7 @@ again:
 	BTRFS_I(inode)->generation = 0;
 	ret = btrfs_update_inode(trans, root, inode);
 	if (ret) {
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 		goto out_put;
 	}
 
@@ -466,7 +469,7 @@ again:
 		ret = btrfs_truncate_free_space_cache(root, trans, NULL, inode);
 		if (ret) {
 			if (ret != -ENOSPC)
-				btrfs_abort_transaction(trans, root, ret);
+				btrfs_abort_transaction(trans, ret);
 			goto out_put;
 		}
 	}
@@ -495,10 +498,9 @@ again:
 	ret = btrfs_prealloc_file_range_trans(inode, trans, 0, 0, prealloc,
 					      prealloc, prealloc, &alloc_hint);
 	if (ret) {
-		btrfs_delalloc_release_space(inode, 0, prealloc);
+		btrfs_delalloc_release_metadata(inode, prealloc);
 		goto out_put;
 	}
-	btrfs_free_reserved_data_space(inode, 0, prealloc);
 
 	ret = btrfs_write_out_ino_cache(root, trans, path, inode);
 out_put:
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 270499598ed4..2b790bda7998 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -60,6 +60,7 @@
 #include "hash.h"
 #include "props.h"
 #include "qgroup.h"
+#include "dedupe.h"
 
 struct btrfs_iget_args {
 	struct btrfs_key *location;
@@ -105,8 +106,9 @@ static int btrfs_truncate(struct inode *inode);
 static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent);
 static noinline int cow_file_range(struct inode *inode,
 				   struct page *locked_page,
-				   u64 start, u64 end, int *page_started,
-				   unsigned long *nr_written, int unlock);
+				   u64 start, u64 end, u64 delalloc_end,
+				   int *page_started, unsigned long *nr_written,
+				   int unlock, struct btrfs_dedupe_hash *hash);
 static struct extent_map *create_pinned_em(struct inode *inode, u64 start,
 					   u64 len, u64 orig_start,
 					   u64 block_start, u64 block_len,
@@ -294,7 +296,7 @@ static noinline int cow_file_range_inline(struct btrfs_root *root,
 				   start, aligned_end, NULL,
 				   1, 1, extent_item_size, &extent_inserted);
 	if (ret) {
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 		goto out;
 	}
 
@@ -305,7 +307,7 @@ static noinline int cow_file_range_inline(struct btrfs_root *root,
 				   inline_len, compressed_size,
 				   compress_type, compressed_pages);
 	if (ret && ret != -ENOSPC) {
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 		goto out;
 	} else if (ret == -ENOSPC) {
 		ret = 1;
@@ -374,12 +376,12 @@ static inline int inode_need_compress(struct inode *inode)
 	struct btrfs_root *root = BTRFS_I(inode)->root;
 
 	/* force compress */
-	if (btrfs_test_opt(root, FORCE_COMPRESS))
+	if (btrfs_test_opt(root->fs_info, FORCE_COMPRESS))
 		return 1;
 	/* bad compression ratios */
 	if (BTRFS_I(inode)->flags & BTRFS_INODE_NOCOMPRESS)
 		return 0;
-	if (btrfs_test_opt(root, COMPRESS) ||
+	if (btrfs_test_opt(root->fs_info, COMPRESS) ||
 	    BTRFS_I(inode)->flags & BTRFS_INODE_COMPRESS ||
 	    BTRFS_I(inode)->force_compress)
 		return 1;
@@ -558,12 +560,15 @@ cont:
 			 * we don't need to create any more async work items.
 			 * Unlock and free up our temp pages.
 			 */
-			extent_clear_unlock_delalloc(inode, start, end, NULL,
-						     clear_flags, PAGE_UNLOCK |
+			extent_clear_unlock_delalloc(inode, start, end, end,
+						     NULL, clear_flags,
+						     PAGE_UNLOCK |
 						     PAGE_CLEAR_DIRTY |
 						     PAGE_SET_WRITEBACK |
 						     page_error_op |
 						     PAGE_END_WRITEBACK);
+			btrfs_free_reserved_data_space_noquota(inode, start,
+						end - start + 1);
 			goto free_pages_out;
 		}
 	}
@@ -585,9 +590,27 @@ cont:
 			will_compress = 0;
 		} else {
 			num_bytes = total_in;
+			*num_added += 1;
+
+			/*
+			 * The async work queues will take care of doing actual
+			 * allocation on disk for these compressed pages, and
+			 * will submit them to the elevator.
+			 */
+			add_async_extent(async_cow, start, num_bytes,
+					total_compressed, pages, nr_pages_ret,
+					compress_type);
+
+			if (start + num_bytes < end) {
+				start += num_bytes;
+				pages = NULL;
+				cond_resched();
+				goto again;
+			}
+			return;
 		}
 	}
-	if (!will_compress && pages) {
+	if (pages) {
 		/*
 		 * the compression code ran but failed to make things smaller,
 		 * free any pages it allocated and our page pointer array
@@ -602,48 +625,28 @@ cont:
 		nr_pages_ret = 0;
 
 		/* flag the file so we don't compress in the future */
-		if (!btrfs_test_opt(root, FORCE_COMPRESS) &&
+		if (!btrfs_test_opt(root->fs_info, FORCE_COMPRESS) &&
 		    !(BTRFS_I(inode)->force_compress)) {
 			BTRFS_I(inode)->flags |= BTRFS_INODE_NOCOMPRESS;
 		}
 	}
-	if (will_compress) {
-		*num_added += 1;
-
-		/* the async work queues will take care of doing actual
-		 * allocation on disk for these compressed pages,
-		 * and will submit them to the elevator.
-		 */
-		add_async_extent(async_cow, start, num_bytes,
-				 total_compressed, pages, nr_pages_ret,
-				 compress_type);
-
-		if (start + num_bytes < end) {
-			start += num_bytes;
-			pages = NULL;
-			cond_resched();
-			goto again;
-		}
-	} else {
 cleanup_and_bail_uncompressed:
-		/*
-		 * No compression, but we still need to write the pages in
-		 * the file we've been given so far.  redirty the locked
-		 * page if it corresponds to our extent and set things up
-		 * for the async work queue to run cow_file_range to do
-		 * the normal delalloc dance
-		 */
-		if (page_offset(locked_page) >= start &&
-		    page_offset(locked_page) <= end) {
-			__set_page_dirty_nobuffers(locked_page);
-			/* unlocked later on in the async handlers */
-		}
-		if (redirty)
-			extent_range_redirty_for_io(inode, start, end);
-		add_async_extent(async_cow, start, end - start + 1,
-				 0, NULL, 0, BTRFS_COMPRESS_NONE);
-		*num_added += 1;
-	}
+	/*
+	 * No compression, but we still need to write the pages in the file
+	 * we've been given so far.  redirty the locked page if it corresponds
+	 * to our extent and set things up for the async work queue to run
+	 * cow_file_range to do the normal delalloc dance.
+	 */
+	if (page_offset(locked_page) >= start &&
+	    page_offset(locked_page) <= end)
+		__set_page_dirty_nobuffers(locked_page);
+		/* unlocked later on in the async handlers */
+
+	if (redirty)
+		extent_range_redirty_for_io(inode, start, end);
+	add_async_extent(async_cow, start, end - start + 1, 0, NULL, 0,
+			 BTRFS_COMPRESS_NONE);
+	*num_added += 1;
 
 	return;
 
@@ -712,7 +715,10 @@ retry:
 					     async_extent->start,
 					     async_extent->start +
 					     async_extent->ram_size - 1,
-					     &page_started, &nr_written, 0);
+					     async_extent->start +
+					     async_extent->ram_size - 1,
+					     &page_started, &nr_written, 0,
+					     NULL);
 
 			/* JDM XXX */
 
@@ -739,7 +745,7 @@ retry:
 		lock_extent(io_tree, async_extent->start,
 			    async_extent->start + async_extent->ram_size - 1);
 
-		ret = btrfs_reserve_extent(root,
+		ret = btrfs_reserve_extent(root, async_extent->ram_size,
 					   async_extent->compressed_size,
 					   async_extent->compressed_size,
 					   0, alloc_hint, &ins, 1, 1);
@@ -832,6 +838,8 @@ retry:
 		extent_clear_unlock_delalloc(inode, async_extent->start,
 				async_extent->start +
 				async_extent->ram_size - 1,
+				async_extent->start +
+				async_extent->ram_size - 1,
 				NULL, EXTENT_LOCKED | EXTENT_DELALLOC,
 				PAGE_UNLOCK | PAGE_CLEAR_DIRTY |
 				PAGE_SET_WRITEBACK);
@@ -851,7 +859,8 @@ retry:
 			tree->ops->writepage_end_io_hook(p, start, end,
 							 NULL, 0);
 			p->mapping = NULL;
-			extent_clear_unlock_delalloc(inode, start, end, NULL, 0,
+			extent_clear_unlock_delalloc(inode, start, end, end,
+						     NULL, 0,
 						     PAGE_END_WRITEBACK |
 						     PAGE_SET_ERROR);
 			free_async_extent_pages(async_extent);
@@ -868,6 +877,8 @@ out_free:
 	extent_clear_unlock_delalloc(inode, async_extent->start,
 				     async_extent->start +
 				     async_extent->ram_size - 1,
+				     async_extent->start +
+				     async_extent->ram_size - 1,
 				     NULL, EXTENT_LOCKED | EXTENT_DELALLOC |
 				     EXTENT_DEFRAG | EXTENT_DO_ACCOUNTING,
 				     PAGE_UNLOCK | PAGE_CLEAR_DIRTY |
@@ -925,9 +936,9 @@ static u64 get_extent_allocation_hint(struct inode *inode, u64 start,
  */
 static noinline int cow_file_range(struct inode *inode,
 				   struct page *locked_page,
-				   u64 start, u64 end, int *page_started,
-				   unsigned long *nr_written,
-				   int unlock)
+				   u64 start, u64 end, u64 delalloc_end,
+				   int *page_started, unsigned long *nr_written,
+				   int unlock, struct btrfs_dedupe_hash *hash)
 {
 	struct btrfs_root *root = BTRFS_I(inode)->root;
 	u64 alloc_hint = 0;
@@ -961,12 +972,14 @@ static noinline int cow_file_range(struct inode *inode,
 		ret = cow_file_range_inline(root, inode, start, end, 0, 0,
 					    NULL);
 		if (ret == 0) {
-			extent_clear_unlock_delalloc(inode, start, end, NULL,
+			extent_clear_unlock_delalloc(inode, start, end,
+				     delalloc_end, NULL,
 				     EXTENT_LOCKED | EXTENT_DELALLOC |
 				     EXTENT_DEFRAG, PAGE_UNLOCK |
 				     PAGE_CLEAR_DIRTY | PAGE_SET_WRITEBACK |
 				     PAGE_END_WRITEBACK);
-
+			btrfs_free_reserved_data_space_noquota(inode, start,
+						end - start + 1);
 			*nr_written = *nr_written +
 			     (end - start + PAGE_SIZE) / PAGE_SIZE;
 			*page_started = 1;
@@ -986,7 +999,7 @@ static noinline int cow_file_range(struct inode *inode,
 		unsigned long op;
 
 		cur_alloc_size = disk_num_bytes;
-		ret = btrfs_reserve_extent(root, cur_alloc_size,
+		ret = btrfs_reserve_extent(root, cur_alloc_size, cur_alloc_size,
 					   root->sectorsize, 0, alloc_hint,
 					   &ins, 1, 1);
 		if (ret < 0)
@@ -1056,7 +1069,8 @@ static noinline int cow_file_range(struct inode *inode,
 		op |= PAGE_SET_PRIVATE2;
 
 		extent_clear_unlock_delalloc(inode, start,
-					     start + ram_size - 1, locked_page,
+					     start + ram_size - 1,
+					     delalloc_end, locked_page,
 					     EXTENT_LOCKED | EXTENT_DELALLOC,
 					     op);
 		disk_num_bytes -= cur_alloc_size;
@@ -1073,7 +1087,8 @@ out_reserve:
 	btrfs_dec_block_group_reservations(root->fs_info, ins.objectid);
 	btrfs_free_reserved_extent(root, ins.objectid, ins.offset, 1);
 out_unlock:
-	extent_clear_unlock_delalloc(inode, start, end, locked_page,
+	extent_clear_unlock_delalloc(inode, start, end, delalloc_end,
+				     locked_page,
 				     EXTENT_LOCKED | EXTENT_DO_ACCOUNTING |
 				     EXTENT_DELALLOC | EXTENT_DEFRAG,
 				     PAGE_UNLOCK | PAGE_CLEAR_DIRTY |
@@ -1156,7 +1171,7 @@ static int cow_file_range_async(struct inode *inode, struct page *locked_page,
 		async_cow->start = start;
 
 		if (BTRFS_I(inode)->flags & BTRFS_INODE_NOCOMPRESS &&
-		    !btrfs_test_opt(root, FORCE_COMPRESS))
+		    !btrfs_test_opt(root->fs_info, FORCE_COMPRESS))
 			cur_end = end;
 		else
 			cur_end = min(end, start + SZ_512K - 1);
@@ -1252,7 +1267,8 @@ static noinline int run_delalloc_nocow(struct inode *inode,
 
 	path = btrfs_alloc_path();
 	if (!path) {
-		extent_clear_unlock_delalloc(inode, start, end, locked_page,
+		extent_clear_unlock_delalloc(inode, start, end, end,
+					     locked_page,
 					     EXTENT_LOCKED | EXTENT_DELALLOC |
 					     EXTENT_DO_ACCOUNTING |
 					     EXTENT_DEFRAG, PAGE_UNLOCK |
@@ -1270,7 +1286,8 @@ static noinline int run_delalloc_nocow(struct inode *inode,
 		trans = btrfs_join_transaction(root);
 
 	if (IS_ERR(trans)) {
-		extent_clear_unlock_delalloc(inode, start, end, locked_page,
+		extent_clear_unlock_delalloc(inode, start, end, end,
+					     locked_page,
 					     EXTENT_LOCKED | EXTENT_DELALLOC |
 					     EXTENT_DO_ACCOUNTING |
 					     EXTENT_DEFRAG, PAGE_UNLOCK |
@@ -1418,7 +1435,8 @@ out_check:
 		if (cow_start != (u64)-1) {
 			ret = cow_file_range(inode, locked_page,
 					     cow_start, found_key.offset - 1,
-					     page_started, nr_written, 1);
+					     end, page_started, nr_written, 1,
+					     NULL);
 			if (ret) {
 				if (!nolock && nocow)
 					btrfs_end_write_no_snapshoting(root);
@@ -1483,10 +1501,12 @@ out_check:
 		}
 
 		extent_clear_unlock_delalloc(inode, cur_offset,
-					     cur_offset + num_bytes - 1,
+					     cur_offset + num_bytes - 1, end,
 					     locked_page, EXTENT_LOCKED |
-					     EXTENT_DELALLOC, PAGE_UNLOCK |
-					     PAGE_SET_PRIVATE2);
+					     EXTENT_DELALLOC |
+					     EXTENT_CLEAR_DATA_RESV,
+					     PAGE_UNLOCK | PAGE_SET_PRIVATE2);
+
 		if (!nolock && nocow)
 			btrfs_end_write_no_snapshoting(root);
 		cur_offset = extent_end;
@@ -1501,8 +1521,8 @@ out_check:
 	}
 
 	if (cow_start != (u64)-1) {
-		ret = cow_file_range(inode, locked_page, cow_start, end,
-				     page_started, nr_written, 1);
+		ret = cow_file_range(inode, locked_page, cow_start, end, end,
+				     page_started, nr_written, 1, NULL);
 		if (ret)
 			goto error;
 	}
@@ -1513,7 +1533,7 @@ error:
 		ret = err;
 
 	if (ret && cur_offset < end)
-		extent_clear_unlock_delalloc(inode, cur_offset, end,
+		extent_clear_unlock_delalloc(inode, cur_offset, end, end,
 					     locked_page, EXTENT_LOCKED |
 					     EXTENT_DELALLOC | EXTENT_DEFRAG |
 					     EXTENT_DO_ACCOUNTING, PAGE_UNLOCK |
@@ -1561,8 +1581,8 @@ static int run_delalloc_range(struct inode *inode, struct page *locked_page,
 		ret = run_delalloc_nocow(inode, locked_page, start, end,
 					 page_started, 0, nr_written);
 	} else if (!inode_need_compress(inode)) {
-		ret = cow_file_range(inode, locked_page, start, end,
-				      page_started, nr_written, 1);
+		ret = cow_file_range(inode, locked_page, start, end, end,
+				      page_started, nr_written, 1, NULL);
 	} else {
 		set_bit(BTRFS_INODE_HAS_ASYNC_EXTENT,
 			&BTRFS_I(inode)->runtime_flags);
@@ -1740,7 +1760,7 @@ static void btrfs_set_bit_hook(struct inode *inode,
 		}
 
 		/* For sanity tests */
-		if (btrfs_test_is_dummy_root(root))
+		if (btrfs_is_testing(root->fs_info))
 			return;
 
 		__percpu_counter_add(&root->fs_info->delalloc_bytes, len,
@@ -1799,11 +1819,13 @@ static void btrfs_clear_bit_hook(struct inode *inode,
 			btrfs_delalloc_release_metadata(inode, len);
 
 		/* For sanity tests. */
-		if (btrfs_test_is_dummy_root(root))
+		if (btrfs_is_testing(root->fs_info))
 			return;
 
 		if (root->root_key.objectid != BTRFS_DATA_RELOC_TREE_OBJECTID
-		    && do_list && !(state->state & EXTENT_NORESERVE))
+		    && do_list && !(state->state & EXTENT_NORESERVE)
+		    && (*bits & (EXTENT_DO_ACCOUNTING |
+		    EXTENT_CLEAR_DATA_RESV)))
 			btrfs_free_reserved_data_space_noquota(inode,
 					state->start, len);
 
@@ -1822,8 +1844,12 @@ static void btrfs_clear_bit_hook(struct inode *inode,
 /*
  * extent_io.c merge_bio_hook, this must check the chunk tree to make sure
  * we don't create bios that span stripes or chunks
+ *
+ * return 1 if page cannot be merged to bio
+ * return 0 if page can be merged to bio
+ * return error otherwise
  */
-int btrfs_merge_bio_hook(int rw, struct page *page, unsigned long offset,
+int btrfs_merge_bio_hook(struct page *page, unsigned long offset,
 			 size_t size, struct bio *bio,
 			 unsigned long bio_flags)
 {
@@ -1838,10 +1864,10 @@ int btrfs_merge_bio_hook(int rw, struct page *page, unsigned long offset,
 
 	length = bio->bi_iter.bi_size;
 	map_length = length;
-	ret = btrfs_map_block(root->fs_info, rw, logical,
+	ret = btrfs_map_block(root->fs_info, bio_op(bio), logical,
 			      &map_length, NULL, 0);
-	/* Will always return 0 with map_multi == NULL */
-	BUG_ON(ret < 0);
+	if (ret < 0)
+		return ret;
 	if (map_length < length + size)
 		return 1;
 	return 0;
@@ -1855,9 +1881,8 @@ int btrfs_merge_bio_hook(int rw, struct page *page, unsigned long offset,
  * At IO completion time the cums attached on the ordered extent record
  * are inserted into the btree
  */
-static int __btrfs_submit_bio_start(struct inode *inode, int rw,
-				    struct bio *bio, int mirror_num,
-				    unsigned long bio_flags,
+static int __btrfs_submit_bio_start(struct inode *inode, struct bio *bio,
+				    int mirror_num, unsigned long bio_flags,
 				    u64 bio_offset)
 {
 	struct btrfs_root *root = BTRFS_I(inode)->root;
@@ -1876,14 +1901,14 @@ static int __btrfs_submit_bio_start(struct inode *inode, int rw,
  * At IO completion time the cums attached on the ordered extent record
  * are inserted into the btree
  */
-static int __btrfs_submit_bio_done(struct inode *inode, int rw, struct bio *bio,
+static int __btrfs_submit_bio_done(struct inode *inode, struct bio *bio,
 			  int mirror_num, unsigned long bio_flags,
 			  u64 bio_offset)
 {
 	struct btrfs_root *root = BTRFS_I(inode)->root;
 	int ret;
 
-	ret = btrfs_map_bio(root, rw, bio, mirror_num, 1);
+	ret = btrfs_map_bio(root, bio, mirror_num, 1);
 	if (ret) {
 		bio->bi_error = ret;
 		bio_endio(bio);
@@ -1895,7 +1920,7 @@ static int __btrfs_submit_bio_done(struct inode *inode, int rw, struct bio *bio,
  * extent_io.c submission hook. This does the right thing for csum calculation
  * on write, or reading the csums from the tree before a read
  */
-static int btrfs_submit_bio_hook(struct inode *inode, int rw, struct bio *bio,
+static int btrfs_submit_bio_hook(struct inode *inode, struct bio *bio,
 			  int mirror_num, unsigned long bio_flags,
 			  u64 bio_offset)
 {
@@ -1910,7 +1935,7 @@ static int btrfs_submit_bio_hook(struct inode *inode, int rw, struct bio *bio,
 	if (btrfs_is_free_space_inode(inode))
 		metadata = BTRFS_WQ_ENDIO_FREE_SPACE;
 
-	if (!(rw & REQ_WRITE)) {
+	if (bio_op(bio) != REQ_OP_WRITE) {
 		ret = btrfs_bio_wq_end_io(root->fs_info, bio, metadata);
 		if (ret)
 			goto out;
@@ -1932,7 +1957,7 @@ static int btrfs_submit_bio_hook(struct inode *inode, int rw, struct bio *bio,
 			goto mapit;
 		/* we're doing a write, do the async checksumming */
 		ret = btrfs_wq_submit_bio(BTRFS_I(inode)->root->fs_info,
-				   inode, rw, bio, mirror_num,
+				   inode, bio, mirror_num,
 				   bio_flags, bio_offset,
 				   __btrfs_submit_bio_start,
 				   __btrfs_submit_bio_done);
@@ -1944,7 +1969,7 @@ static int btrfs_submit_bio_hook(struct inode *inode, int rw, struct bio *bio,
 	}
 
 mapit:
-	ret = btrfs_map_bio(root, rw, bio, mirror_num, 0);
+	ret = btrfs_map_bio(root, bio, mirror_num, 0);
 
 out:
 	if (ret < 0) {
@@ -1974,7 +1999,7 @@ static noinline int add_pending_csums(struct btrfs_trans_handle *trans,
 }
 
 int btrfs_set_extent_delalloc(struct inode *inode, u64 start, u64 end,
-			      struct extent_state **cached_state)
+			      struct extent_state **cached_state, int dedupe)
 {
 	WARN_ON((end & (PAGE_SIZE - 1)) == 0);
 	return set_extent_delalloc(&BTRFS_I(inode)->io_tree, start, end,
@@ -2038,7 +2063,8 @@ again:
 		goto out;
 	 }
 
-	btrfs_set_extent_delalloc(inode, page_start, page_end, &cached_state);
+	btrfs_set_extent_delalloc(inode, page_start, page_end, &cached_state,
+				  0);
 	ClearPageChecked(page);
 	set_page_dirty(page);
 out:
@@ -2295,7 +2321,7 @@ static noinline int record_one_backref(u64 inum, u64 offset, u64 root_id,
 		if (PTR_ERR(root) == -ENOENT)
 			return 0;
 		WARN_ON(1);
-		pr_debug("inum=%llu, offset=%llu, root_id=%llu\n",
+		btrfs_debug(fs_info, "inum=%llu, offset=%llu, root_id=%llu",
 			 inum, offset, root_id);
 		return PTR_ERR(root);
 	}
@@ -2595,7 +2621,7 @@ again:
 	ret = btrfs_insert_empty_item(trans, root, path, &key,
 					sizeof(*extent));
 	if (ret) {
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 		goto out_free_path;
 	}
 
@@ -2622,7 +2648,7 @@ again:
 			backref->root_id, backref->inum,
 			new->file_pos);	/* start - extent_offset */
 	if (ret) {
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 		goto out_free_path;
 	}
 
@@ -2891,7 +2917,7 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent)
 		trans->block_rsv = &root->fs_info->delalloc_block_rsv;
 		ret = btrfs_update_inode_fallback(trans, root, inode);
 		if (ret) /* -ENOMEM or corruption */
-			btrfs_abort_transaction(trans, root, ret);
+			btrfs_abort_transaction(trans, ret);
 		goto out;
 	}
 
@@ -2951,7 +2977,7 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent)
 			   ordered_extent->file_offset, ordered_extent->len,
 			   trans->transid);
 	if (ret < 0) {
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 		goto out_unlock;
 	}
 
@@ -2961,7 +2987,7 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent)
 	btrfs_ordered_update_i_size(inode, 0, ordered_extent);
 	ret = btrfs_update_inode_fallback(trans, root, inode);
 	if (ret) { /* -ENOMEM or corruption */
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 		goto out_unlock;
 	}
 	ret = 0;
@@ -3205,7 +3231,7 @@ void btrfs_orphan_commit_root(struct btrfs_trans_handle *trans,
 		ret = btrfs_del_orphan_item(trans, root->fs_info->tree_root,
 					    root->root_key.objectid);
 		if (ret)
-			btrfs_abort_transaction(trans, root, ret);
+			btrfs_abort_transaction(trans, ret);
 		else
 			clear_bit(BTRFS_ROOT_ORPHAN_ITEM_INSERTED,
 				  &root->state);
@@ -3271,7 +3297,16 @@ int btrfs_orphan_add(struct btrfs_trans_handle *trans, struct inode *inode)
 	/* grab metadata reservation from transaction handle */
 	if (reserve) {
 		ret = btrfs_orphan_reserve_metadata(trans, inode);
-		BUG_ON(ret); /* -ENOSPC in reservation; Logic error? JDM */
+		ASSERT(!ret);
+		if (ret) {
+			atomic_dec(&root->orphan_inodes);
+			clear_bit(BTRFS_INODE_ORPHAN_META_RESERVED,
+				  &BTRFS_I(inode)->runtime_flags);
+			if (insert)
+				clear_bit(BTRFS_INODE_HAS_ORPHAN_ITEM,
+					  &BTRFS_I(inode)->runtime_flags);
+			return ret;
+		}
 	}
 
 	/* insert an orphan item to track this unlinked/truncated file */
@@ -3287,7 +3322,7 @@ int btrfs_orphan_add(struct btrfs_trans_handle *trans, struct inode *inode)
 			if (ret != -EEXIST) {
 				clear_bit(BTRFS_INODE_HAS_ORPHAN_ITEM,
 					  &BTRFS_I(inode)->runtime_flags);
-				btrfs_abort_transaction(trans, root, ret);
+				btrfs_abort_transaction(trans, ret);
 				return ret;
 			}
 		}
@@ -3299,7 +3334,7 @@ int btrfs_orphan_add(struct btrfs_trans_handle *trans, struct inode *inode)
 		ret = btrfs_insert_orphan_item(trans, root->fs_info->tree_root,
 					       root->root_key.objectid);
 		if (ret && ret != -EEXIST) {
-			btrfs_abort_transaction(trans, root, ret);
+			btrfs_abort_transaction(trans, ret);
 			return ret;
 		}
 	}
@@ -3419,10 +3454,10 @@ int btrfs_orphan_cleanup(struct btrfs_root *root)
 		found_key.offset = 0;
 		inode = btrfs_iget(root->fs_info->sb, &found_key, root, NULL);
 		ret = PTR_ERR_OR_ZERO(inode);
-		if (ret && ret != -ESTALE)
+		if (ret && ret != -ENOENT)
 			goto out;
 
-		if (ret == -ESTALE && root == root->fs_info->tree_root) {
+		if (ret == -ENOENT && root == root->fs_info->tree_root) {
 			struct btrfs_root *dead_root;
 			struct btrfs_fs_info *fs_info = root->fs_info;
 			int is_dead_root = 0;
@@ -3458,7 +3493,7 @@ int btrfs_orphan_cleanup(struct btrfs_root *root)
 		 * Inode is already gone but the orphan item is still there,
 		 * kill the orphan item.
 		 */
-		if (ret == -ESTALE) {
+		if (ret == -ENOENT) {
 			trans = btrfs_start_transaction(root, 1);
 			if (IS_ERR(trans)) {
 				ret = PTR_ERR(trans);
@@ -3617,7 +3652,7 @@ static noinline int acls_after_inode_item(struct extent_buffer *leaf,
 /*
  * read an inode from the btree into the in-memory inode
  */
-static void btrfs_read_locked_inode(struct inode *inode)
+static int btrfs_read_locked_inode(struct inode *inode)
 {
 	struct btrfs_path *path;
 	struct extent_buffer *leaf;
@@ -3636,14 +3671,19 @@ static void btrfs_read_locked_inode(struct inode *inode)
 		filled = true;
 
 	path = btrfs_alloc_path();
-	if (!path)
+	if (!path) {
+		ret = -ENOMEM;
 		goto make_bad;
+	}
 
 	memcpy(&location, &BTRFS_I(inode)->location, sizeof(location));
 
 	ret = btrfs_lookup_inode(NULL, root, path, &location, 0);
-	if (ret)
+	if (ret) {
+		if (ret > 0)
+			ret = -ENOENT;
 		goto make_bad;
+	}
 
 	leaf = path->nodes[0];
 
@@ -3796,11 +3836,12 @@ cache_acl:
 	}
 
 	btrfs_update_iflags(inode);
-	return;
+	return 0;
 
 make_bad:
 	btrfs_free_path(path);
 	make_bad_inode(inode);
+	return ret;
 }
 
 /*
@@ -3907,7 +3948,7 @@ noinline int btrfs_update_inode(struct btrfs_trans_handle *trans,
 	 */
 	if (!btrfs_is_free_space_inode(inode)
 	    && root->root_key.objectid != BTRFS_DATA_RELOC_TREE_OBJECTID
-	    && !root->fs_info->log_root_recovering) {
+	    && !test_bit(BTRFS_FS_LOG_RECOVERING, &root->fs_info->flags)) {
 		btrfs_update_root_times(trans, root);
 
 		ret = btrfs_delayed_update_inode(trans, root, inode);
@@ -3998,20 +4039,20 @@ static int __btrfs_unlink_inode(struct btrfs_trans_handle *trans,
 		btrfs_info(root->fs_info,
 			"failed to delete reference to %.*s, inode %llu parent %llu",
 			name_len, name, ino, dir_ino);
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 		goto err;
 	}
 skip_backref:
 	ret = btrfs_delete_delayed_dir_index(trans, root, dir, index);
 	if (ret) {
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 		goto err;
 	}
 
 	ret = btrfs_del_inode_ref_in_log(trans, root, name, name_len,
 					 inode, dir_ino);
 	if (ret != 0 && ret != -ENOENT) {
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 		goto err;
 	}
 
@@ -4020,7 +4061,7 @@ skip_backref:
 	if (ret == -ENOENT)
 		ret = 0;
 	else if (ret)
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 err:
 	btrfs_free_path(path);
 	if (ret)
@@ -4030,7 +4071,7 @@ err:
 	inode_inc_iversion(inode);
 	inode_inc_iversion(dir);
 	inode->i_ctime = dir->i_mtime =
-		dir->i_ctime = current_fs_time(inode->i_sb);
+		dir->i_ctime = current_time(inode);
 	ret = btrfs_update_inode(trans, root, dir);
 out:
 	return ret;
@@ -4134,7 +4175,7 @@ int btrfs_unlink_subvol(struct btrfs_trans_handle *trans,
 	WARN_ON(key.type != BTRFS_ROOT_ITEM_KEY || key.objectid != objectid);
 	ret = btrfs_delete_one_dir_name(trans, root, path, di);
 	if (ret) {
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 		goto out;
 	}
 	btrfs_release_path(path);
@@ -4144,7 +4185,7 @@ int btrfs_unlink_subvol(struct btrfs_trans_handle *trans,
 				 dir_ino, &index, name, name_len);
 	if (ret < 0) {
 		if (ret != -ENOENT) {
-			btrfs_abort_transaction(trans, root, ret);
+			btrfs_abort_transaction(trans, ret);
 			goto out;
 		}
 		di = btrfs_search_dir_index_item(root, path, dir_ino,
@@ -4154,7 +4195,7 @@ int btrfs_unlink_subvol(struct btrfs_trans_handle *trans,
 				ret = -ENOENT;
 			else
 				ret = PTR_ERR(di);
-			btrfs_abort_transaction(trans, root, ret);
+			btrfs_abort_transaction(trans, ret);
 			goto out;
 		}
 
@@ -4167,16 +4208,16 @@ int btrfs_unlink_subvol(struct btrfs_trans_handle *trans,
 
 	ret = btrfs_delete_delayed_dir_index(trans, root, dir, index);
 	if (ret) {
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 		goto out;
 	}
 
 	btrfs_i_size_write(dir, dir->i_size - name_len * 2);
 	inode_inc_iversion(dir);
-	dir->i_mtime = dir->i_ctime = current_fs_time(dir->i_sb);
+	dir->i_mtime = dir->i_ctime = current_time(dir);
 	ret = btrfs_update_inode_fallback(trans, root, dir);
 	if (ret)
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 out:
 	btrfs_free_path(path);
 	return ret;
@@ -4188,6 +4229,7 @@ static int btrfs_rmdir(struct inode *dir, struct dentry *dentry)
 	int err = 0;
 	struct btrfs_root *root = BTRFS_I(dir)->root;
 	struct btrfs_trans_handle *trans;
+	u64 last_unlink_trans;
 
 	if (inode->i_size > BTRFS_EMPTY_DIR_SIZE)
 		return -ENOTEMPTY;
@@ -4210,11 +4252,27 @@ static int btrfs_rmdir(struct inode *dir, struct dentry *dentry)
 	if (err)
 		goto out;
 
+	last_unlink_trans = BTRFS_I(inode)->last_unlink_trans;
+
 	/* now the directory is empty */
 	err = btrfs_unlink_inode(trans, root, dir, d_inode(dentry),
 				 dentry->d_name.name, dentry->d_name.len);
-	if (!err)
+	if (!err) {
 		btrfs_i_size_write(inode, 0);
+		/*
+		 * Propagate the last_unlink_trans value of the deleted dir to
+		 * its parent directory. This is to prevent an unrecoverable
+		 * log tree in the case we do something like this:
+		 * 1) create dir foo
+		 * 2) create snapshot under dir foo
+		 * 3) delete the snapshot
+		 * 4) rmdir foo
+		 * 5) mkdir foo
+		 * 6) fsync foo or some file inside foo
+		 */
+		if (last_unlink_trans >= trans->transid)
+			BTRFS_I(dir)->last_unlink_trans = last_unlink_trans;
+	}
 out:
 	btrfs_end_transaction(trans, root);
 	btrfs_btree_balance_dirty(root);
@@ -4497,7 +4555,6 @@ search_again:
 							      pending_del_nr);
 					if (err) {
 						btrfs_abort_transaction(trans,
-									root,
 									err);
 						goto error;
 					}
@@ -4509,8 +4566,7 @@ search_again:
 							     item_end,
 							     new_size);
 				if (err) {
-					btrfs_abort_transaction(trans,
-								root, err);
+					btrfs_abort_transaction(trans, err);
 					goto error;
 				}
 			} else if (test_bit(BTRFS_ROOT_REF_COWS,
@@ -4549,6 +4605,7 @@ delete:
 			BUG_ON(ret);
 			if (btrfs_should_throttle_delayed_refs(trans, root))
 				btrfs_async_run_delayed_refs(root,
+							     trans->transid,
 					trans->delayed_ref_updates * 2, 0);
 			if (be_nice) {
 				if (truncate_space_check(trans, root,
@@ -4573,8 +4630,7 @@ delete:
 						pending_del_slot,
 						pending_del_nr);
 				if (ret) {
-					btrfs_abort_transaction(trans,
-								root, ret);
+					btrfs_abort_transaction(trans, ret);
 					goto error;
 				}
 				pending_del_nr = 0;
@@ -4607,7 +4663,7 @@ out:
 		ret = btrfs_del_items(trans, root, path, pending_del_slot,
 				      pending_del_nr);
 		if (ret)
-			btrfs_abort_transaction(trans, root, ret);
+			btrfs_abort_transaction(trans, ret);
 	}
 error:
 	if (root->root_key.objectid != BTRFS_TREE_LOG_OBJECTID)
@@ -4713,7 +4769,7 @@ again:
 			  0, 0, &cached_state, GFP_NOFS);
 
 	ret = btrfs_set_extent_delalloc(inode, block_start, block_end,
-					&cached_state);
+					&cached_state, 0);
 	if (ret) {
 		unlock_extent_cached(io_tree, block_start, block_end,
 				     &cached_state, GFP_NOFS);
@@ -4776,7 +4832,7 @@ static int maybe_insert_hole(struct btrfs_root *root, struct inode *inode,
 
 	ret = btrfs_drop_extents(trans, root, inode, offset, offset + len, 1);
 	if (ret) {
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 		btrfs_end_transaction(trans, root);
 		return ret;
 	}
@@ -4784,7 +4840,7 @@ static int maybe_insert_hole(struct btrfs_root *root, struct inode *inode,
 	ret = btrfs_insert_file_extent(trans, root, btrfs_ino(inode), offset,
 				       0, 0, len, 0, len, 0, 0, 0);
 	if (ret)
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 	else
 		btrfs_update_inode(trans, root, inode);
 	btrfs_end_transaction(trans, root);
@@ -4921,7 +4977,7 @@ static int btrfs_setsize(struct inode *inode, struct iattr *attr)
 		inode_inc_iversion(inode);
 		if (!(mask & (ATTR_CTIME | ATTR_MTIME)))
 			inode->i_ctime = inode->i_mtime =
-				current_fs_time(inode->i_sb);
+				current_time(inode);
 	}
 
 	if (newsize > oldsize) {
@@ -5011,7 +5067,7 @@ static int btrfs_setsize(struct inode *inode, struct iattr *attr)
 			i_size_write(inode, BTRFS_I(inode)->disk_i_size);
 			err = btrfs_orphan_del(trans, inode);
 			if (err)
-				btrfs_abort_transaction(trans, root, err);
+				btrfs_abort_transaction(trans, err);
 			btrfs_end_transaction(trans, root);
 		}
 	}
@@ -5028,7 +5084,7 @@ static int btrfs_setattr(struct dentry *dentry, struct iattr *attr)
 	if (btrfs_root_readonly(root))
 		return -EROFS;
 
-	err = inode_change_ok(inode, attr);
+	err = setattr_prepare(dentry, attr);
 	if (err)
 		return err;
 
@@ -5149,11 +5205,18 @@ void btrfs_evict_inode(struct inode *inode)
 	struct btrfs_root *root = BTRFS_I(inode)->root;
 	struct btrfs_block_rsv *rsv, *global_rsv;
 	int steal_from_global = 0;
-	u64 min_size = btrfs_calc_trunc_metadata_size(root, 1);
+	u64 min_size;
 	int ret;
 
 	trace_btrfs_inode_evict(inode);
 
+	if (!root) {
+		kmem_cache_free(btrfs_inode_cachep, BTRFS_I(inode));
+		return;
+	}
+
+	min_size = btrfs_calc_trunc_metadata_size(root, 1);
+
 	evict_inode_truncate_pages(inode);
 
 	if (inode->i_nlink &&
@@ -5172,7 +5235,7 @@ void btrfs_evict_inode(struct inode *inode)
 
 	btrfs_free_io_failure_record(inode, 0, (u64)-1);
 
-	if (root->fs_info->log_root_recovering) {
+	if (test_bit(BTRFS_FS_LOG_RECOVERING, &root->fs_info->flags)) {
 		BUG_ON(test_bit(BTRFS_INODE_HAS_ORPHAN_ITEM,
 				 &BTRFS_I(inode)->runtime_flags));
 		goto no_delete;
@@ -5254,7 +5317,7 @@ void btrfs_evict_inode(struct inode *inode)
 		if (steal_from_global) {
 			if (!btrfs_check_space_for_delayed_refs(trans, root))
 				ret = btrfs_block_rsv_migrate(global_rsv, rsv,
-							      min_size);
+							      min_size, 0);
 			else
 				ret = -ENOSPC;
 		}
@@ -5585,7 +5648,9 @@ struct inode *btrfs_iget(struct super_block *s, struct btrfs_key *location,
 		return ERR_PTR(-ENOMEM);
 
 	if (inode->i_state & I_NEW) {
-		btrfs_read_locked_inode(inode);
+		int ret;
+
+		ret = btrfs_read_locked_inode(inode);
 		if (!is_bad_inode(inode)) {
 			inode_tree_add(inode);
 			unlock_new_inode(inode);
@@ -5594,7 +5659,8 @@ struct inode *btrfs_iget(struct super_block *s, struct btrfs_key *location,
 		} else {
 			unlock_new_inode(inode);
 			iput(inode);
-			inode = ERR_PTR(-ESTALE);
+			ASSERT(ret < 0);
+			inode = ERR_PTR(ret < 0 ? ret : -ESTALE);
 		}
 	}
 
@@ -5618,7 +5684,7 @@ static struct inode *new_simple_dir(struct super_block *s,
 	inode->i_op = &btrfs_dir_ro_inode_operations;
 	inode->i_fop = &simple_dir_operations;
 	inode->i_mode = S_IFDIR | S_IRUGO | S_IWUSR | S_IXUGO;
-	inode->i_mtime = current_fs_time(inode->i_sb);
+	inode->i_mtime = current_time(inode);
 	inode->i_atime = inode->i_mtime;
 	inode->i_ctime = inode->i_mtime;
 	BTRFS_I(inode)->i_otime = inode->i_mtime;
@@ -5748,6 +5814,7 @@ static int btrfs_real_readdir(struct file *file, struct dir_context *ctx)
 	int name_len;
 	int is_curr = 0;	/* ctx->pos points to the current index? */
 	bool emitted;
+	bool put = false;
 
 	/* FIXME, use a real flag for deciding about the key type */
 	if (root->fs_info->tree_root == root)
@@ -5765,7 +5832,8 @@ static int btrfs_real_readdir(struct file *file, struct dir_context *ctx)
 	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);
+		put = btrfs_readdir_get_delayed_items(inode, &ins_list,
+						      &del_list);
 	}
 
 	key.type = key_type;
@@ -5912,8 +5980,8 @@ next:
 nopos:
 	ret = 0;
 err:
-	if (key_type == BTRFS_DIR_INDEX_KEY)
-		btrfs_put_delayed_items(&ins_list, &del_list);
+	if (put)
+		btrfs_readdir_put_delayed_items(inode, &ins_list, &del_list);
 	btrfs_free_path(path);
 	return ret;
 }
@@ -6202,7 +6270,7 @@ static struct inode *btrfs_new_inode(struct btrfs_trans_handle *trans,
 	inode_init_owner(inode, dir, mode);
 	inode_set_bytes(inode, 0);
 
-	inode->i_mtime = current_fs_time(inode->i_sb);
+	inode->i_mtime = current_time(inode);
 	inode->i_atime = inode->i_mtime;
 	inode->i_ctime = inode->i_mtime;
 	BTRFS_I(inode)->i_otime = inode->i_mtime;
@@ -6228,9 +6296,9 @@ static struct inode *btrfs_new_inode(struct btrfs_trans_handle *trans,
 	btrfs_inherit_iflags(inode, dir);
 
 	if (S_ISREG(mode)) {
-		if (btrfs_test_opt(root, NODATASUM))
+		if (btrfs_test_opt(root->fs_info, NODATASUM))
 			BTRFS_I(inode)->flags |= BTRFS_INODE_NODATASUM;
-		if (btrfs_test_opt(root, NODATACOW))
+		if (btrfs_test_opt(root->fs_info, NODATACOW))
 			BTRFS_I(inode)->flags |= BTRFS_INODE_NODATACOW |
 				BTRFS_INODE_NODATASUM;
 	}
@@ -6308,7 +6376,7 @@ int btrfs_add_link(struct btrfs_trans_handle *trans,
 	if (ret == -EEXIST || ret == -EOVERFLOW)
 		goto fail_dir_item;
 	else if (ret) {
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 		return ret;
 	}
 
@@ -6316,10 +6384,10 @@ int btrfs_add_link(struct btrfs_trans_handle *trans,
 			   name_len * 2);
 	inode_inc_iversion(parent_inode);
 	parent_inode->i_mtime = parent_inode->i_ctime =
-		current_fs_time(parent_inode->i_sb);
+		current_time(parent_inode);
 	ret = btrfs_update_inode(trans, root, parent_inode);
 	if (ret)
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 	return ret;
 
 fail_dir_item:
@@ -6534,7 +6602,7 @@ static int btrfs_link(struct dentry *old_dentry, struct inode *dir,
 	BTRFS_I(inode)->dir_index = 0ULL;
 	inc_nlink(inode);
 	inode_inc_iversion(inode);
-	inode->i_ctime = current_fs_time(inode->i_sb);
+	inode->i_ctime = current_time(inode);
 	ihold(inode);
 	set_bit(BTRFS_INODE_COPY_EVERYTHING, &BTRFS_I(inode)->runtime_flags);
 
@@ -6956,8 +7024,9 @@ not_found_em:
 insert:
 	btrfs_release_path(path);
 	if (em->start > start || extent_map_end(em) <= start) {
-		btrfs_err(root->fs_info, "bad extent! em: [%llu %llu] passed [%llu %llu]",
-			em->start, em->len, start, len);
+		btrfs_err(root->fs_info,
+			  "bad extent! em: [%llu %llu] passed [%llu %llu]",
+			  em->start, em->len, start, len);
 		err = -EIO;
 		goto out;
 	}
@@ -6979,7 +7048,18 @@ insert:
 		 * existing will always be non-NULL, since there must be
 		 * extent causing the -EEXIST.
 		 */
-		if (start >= extent_map_end(existing) ||
+		if (existing->start == em->start &&
+		    extent_map_end(existing) == extent_map_end(em) &&
+		    em->block_start == existing->block_start) {
+			/*
+			 * these two extents are the same, it happens
+			 * with inlines especially
+			 */
+			free_extent_map(em);
+			em = existing;
+			err = 0;
+
+		} else if (start >= extent_map_end(existing) ||
 		    start <= existing->start) {
 			/*
 			 * The existing extent map is the one nearest to
@@ -7191,7 +7271,7 @@ static struct extent_map *btrfs_new_extent_direct(struct inode *inode,
 	int ret;
 
 	alloc_hint = get_extent_allocation_hint(inode, start, len);
-	ret = btrfs_reserve_extent(root, len, root->sectorsize, 0,
+	ret = btrfs_reserve_extent(root, len, len, root->sectorsize, 0,
 				   alloc_hint, &ins, 1, 1);
 	if (ret)
 		return ERR_PTR(ret);
@@ -7691,6 +7771,13 @@ static int btrfs_get_blocks_direct(struct inode *inode, sector_t iblock,
 				ret = PTR_ERR(em2);
 				goto unlock_err;
 			}
+			/*
+			 * For inode marked NODATACOW or extent marked PREALLOC,
+			 * use the existing or preallocated extent, so does not
+			 * need to adjust btrfs_space_info's bytes_may_use.
+			 */
+			btrfs_free_reserved_data_space_noquota(inode,
+					start, len);
 			goto unlock;
 		}
 	}
@@ -7725,7 +7812,6 @@ unlock:
 			i_size_write(inode, start + len);
 
 		adjust_dio_outstanding_extents(inode, dio_data, len);
-		btrfs_free_reserved_data_space(inode, start, len);
 		WARN_ON(dio_data->reserve < len);
 		dio_data->reserve -= len;
 		dio_data->unsubmitted_oe_range_end = start + len;
@@ -7767,12 +7853,12 @@ err:
 }
 
 static inline int submit_dio_repair_bio(struct inode *inode, struct bio *bio,
-					int rw, int mirror_num)
+					int mirror_num)
 {
 	struct btrfs_root *root = BTRFS_I(inode)->root;
 	int ret;
 
-	BUG_ON(rw & REQ_WRITE);
+	BUG_ON(bio_op(bio) == REQ_OP_WRITE);
 
 	bio_get(bio);
 
@@ -7781,7 +7867,7 @@ static inline int submit_dio_repair_bio(struct inode *inode, struct bio *bio,
 	if (ret)
 		goto err;
 
-	ret = btrfs_map_bio(root, rw, bio, mirror_num, 0);
+	ret = btrfs_map_bio(root, bio, mirror_num, 0);
 err:
 	bio_put(bio);
 	return ret;
@@ -7792,18 +7878,19 @@ static int btrfs_check_dio_repairable(struct inode *inode,
 				      struct io_failure_record *failrec,
 				      int failed_mirror)
 {
+	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	int num_copies;
 
-	num_copies = btrfs_num_copies(BTRFS_I(inode)->root->fs_info,
-				      failrec->logical, failrec->len);
+	num_copies = btrfs_num_copies(fs_info, failrec->logical, failrec->len);
 	if (num_copies == 1) {
 		/*
 		 * we only have a single copy of the data, so don't bother with
 		 * all the retry and error correction code that follows. no
 		 * matter what the error is, it is very likely to persist.
 		 */
-		pr_debug("Check DIO Repairable: cannot repair, num_copies=%d, next_mirror %d, failed_mirror %d\n",
-			 num_copies, failrec->this_mirror, failed_mirror);
+		btrfs_debug(fs_info,
+			"Check DIO Repairable: cannot repair, num_copies=%d, next_mirror %d, failed_mirror %d",
+			num_copies, failrec->this_mirror, failed_mirror);
 		return 0;
 	}
 
@@ -7813,8 +7900,9 @@ static int btrfs_check_dio_repairable(struct inode *inode,
 		failrec->this_mirror++;
 
 	if (failrec->this_mirror > num_copies) {
-		pr_debug("Check DIO Repairable: (fail) num_copies=%d, next_mirror %d, failed_mirror %d\n",
-			 num_copies, failrec->this_mirror, failed_mirror);
+		btrfs_debug(fs_info,
+			"Check DIO Repairable: (fail) num_copies=%d, next_mirror %d, failed_mirror %d",
+			num_copies, failrec->this_mirror, failed_mirror);
 		return 0;
 	}
 
@@ -7832,7 +7920,7 @@ static int dio_read_error(struct inode *inode, struct bio *failed_bio,
 	int read_mode;
 	int ret;
 
-	BUG_ON(failed_bio->bi_rw & REQ_WRITE);
+	BUG_ON(bio_op(failed_bio) == REQ_OP_WRITE);
 
 	ret = btrfs_get_io_failure_record(inode, start, end, &failrec);
 	if (ret)
@@ -7860,13 +7948,13 @@ static int dio_read_error(struct inode *inode, struct bio *failed_bio,
 		free_io_failure(inode, failrec);
 		return -EIO;
 	}
+	bio_set_op_attrs(bio, REQ_OP_READ, read_mode);
 
 	btrfs_debug(BTRFS_I(inode)->root->fs_info,
 		    "Repair DIO Read Error: submitting new dio read[%#x] to this_mirror=%d, in_validation=%d\n",
 		    read_mode, failrec->this_mirror, failrec->in_validation);
 
-	ret = submit_dio_repair_bio(inode, bio, read_mode,
-				    failrec->this_mirror);
+	ret = submit_dio_repair_bio(inode, bio, failrec->this_mirror);
 	if (ret) {
 		free_io_failure(inode, failrec);
 		bio_put(bio);
@@ -8156,7 +8244,7 @@ static void btrfs_endio_direct_write(struct bio *bio)
 	bio_put(bio);
 }
 
-static int __btrfs_submit_bio_start_direct_io(struct inode *inode, int rw,
+static int __btrfs_submit_bio_start_direct_io(struct inode *inode,
 				    struct bio *bio, int mirror_num,
 				    unsigned long bio_flags, u64 offset)
 {
@@ -8174,8 +8262,8 @@ static void btrfs_end_dio_bio(struct bio *bio)
 
 	if (err)
 		btrfs_warn(BTRFS_I(dip->inode)->root->fs_info,
-			   "direct IO failed ino %llu rw %lu sector %#Lx len %u err no %d",
-			   btrfs_ino(dip->inode), bio->bi_rw,
+			   "direct IO failed ino %llu rw %d,%u sector %#Lx len %u err no %d",
+			   btrfs_ino(dip->inode), bio_op(bio), bio->bi_opf,
 			   (unsigned long long)bio->bi_iter.bi_sector,
 			   bio->bi_iter.bi_size, err);
 
@@ -8249,11 +8337,11 @@ static inline int btrfs_lookup_and_bind_dio_csum(struct btrfs_root *root,
 }
 
 static inline int __btrfs_submit_dio_bio(struct bio *bio, struct inode *inode,
-					 int rw, u64 file_offset, int skip_sum,
+					 u64 file_offset, int skip_sum,
 					 int async_submit)
 {
 	struct btrfs_dio_private *dip = bio->bi_private;
-	int write = rw & REQ_WRITE;
+	bool write = bio_op(bio) == REQ_OP_WRITE;
 	struct btrfs_root *root = BTRFS_I(inode)->root;
 	int ret;
 
@@ -8274,8 +8362,7 @@ static inline int __btrfs_submit_dio_bio(struct bio *bio, struct inode *inode,
 
 	if (write && async_submit) {
 		ret = btrfs_wq_submit_bio(root->fs_info,
-				   inode, rw, bio, 0, 0,
-				   file_offset,
+				   inode, bio, 0, 0, file_offset,
 				   __btrfs_submit_bio_start_direct_io,
 				   __btrfs_submit_bio_done);
 		goto err;
@@ -8294,13 +8381,13 @@ static inline int __btrfs_submit_dio_bio(struct bio *bio, struct inode *inode,
 			goto err;
 	}
 map:
-	ret = btrfs_map_bio(root, rw, bio, 0, async_submit);
+	ret = btrfs_map_bio(root, bio, 0, async_submit);
 err:
 	bio_put(bio);
 	return ret;
 }
 
-static int btrfs_submit_direct_hook(int rw, struct btrfs_dio_private *dip,
+static int btrfs_submit_direct_hook(struct btrfs_dio_private *dip,
 				    int skip_sum)
 {
 	struct inode *inode = dip->inode;
@@ -8319,8 +8406,8 @@ static int btrfs_submit_direct_hook(int rw, struct btrfs_dio_private *dip,
 	int i;
 
 	map_length = orig_bio->bi_iter.bi_size;
-	ret = btrfs_map_block(root->fs_info, rw, start_sector << 9,
-			      &map_length, NULL, 0);
+	ret = btrfs_map_block(root->fs_info, bio_op(orig_bio),
+			      start_sector << 9, &map_length, NULL, 0);
 	if (ret)
 		return -EIO;
 
@@ -8340,6 +8427,7 @@ static int btrfs_submit_direct_hook(int rw, struct btrfs_dio_private *dip,
 	if (!bio)
 		return -ENOMEM;
 
+	bio_set_op_attrs(bio, bio_op(orig_bio), bio_flags(orig_bio));
 	bio->bi_private = dip;
 	bio->bi_end_io = btrfs_end_dio_bio;
 	btrfs_io_bio(bio)->logical = file_offset;
@@ -8359,7 +8447,7 @@ next_block:
 			 * before we're done setting it up
 			 */
 			atomic_inc(&dip->pending_bios);
-			ret = __btrfs_submit_dio_bio(bio, inode, rw,
+			ret = __btrfs_submit_dio_bio(bio, inode,
 						     file_offset, skip_sum,
 						     async_submit);
 			if (ret) {
@@ -8377,12 +8465,14 @@ next_block:
 						  start_sector, GFP_NOFS);
 			if (!bio)
 				goto out_err;
+			bio_set_op_attrs(bio, bio_op(orig_bio),
+					 bio_flags(orig_bio));
 			bio->bi_private = dip;
 			bio->bi_end_io = btrfs_end_dio_bio;
 			btrfs_io_bio(bio)->logical = file_offset;
 
 			map_length = orig_bio->bi_iter.bi_size;
-			ret = btrfs_map_block(root->fs_info, rw,
+			ret = btrfs_map_block(root->fs_info, bio_op(orig_bio),
 					      start_sector << 9,
 					      &map_length, NULL, 0);
 			if (ret) {
@@ -8402,7 +8492,7 @@ next_block:
 	}
 
 submit:
-	ret = __btrfs_submit_dio_bio(bio, inode, rw, file_offset, skip_sum,
+	ret = __btrfs_submit_dio_bio(bio, inode, file_offset, skip_sum,
 				     async_submit);
 	if (!ret)
 		return 0;
@@ -8422,14 +8512,14 @@ out_err:
 	return 0;
 }
 
-static void btrfs_submit_direct(int rw, struct bio *dio_bio,
-				struct inode *inode, loff_t file_offset)
+static void btrfs_submit_direct(struct bio *dio_bio, struct inode *inode,
+				loff_t file_offset)
 {
 	struct btrfs_dio_private *dip = NULL;
 	struct bio *io_bio = NULL;
 	struct btrfs_io_bio *btrfs_bio;
 	int skip_sum;
-	int write = rw & REQ_WRITE;
+	bool write = (bio_op(dio_bio) == REQ_OP_WRITE);
 	int ret = 0;
 
 	skip_sum = BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM;
@@ -8480,7 +8570,7 @@ static void btrfs_submit_direct(int rw, struct bio *dio_bio,
 			dio_data->unsubmitted_oe_range_end;
 	}
 
-	ret = btrfs_submit_direct_hook(rw, dip, skip_sum);
+	ret = btrfs_submit_direct_hook(dip, skip_sum);
 	if (!ret)
 		return;
 
@@ -8544,7 +8634,7 @@ static ssize_t check_direct_IO(struct btrfs_root *root, struct kiocb *iocb,
 		goto out;
 
 	/* If this is a write we don't need to check anymore */
-	if (iov_iter_rw(iter) == WRITE)
+	if (iov_iter_rw(iter) != READ || !iter_is_iovec(iter))
 		return 0;
 	/*
 	 * Check to make sure we don't have duplicate iov_base's in this
@@ -8980,7 +9070,7 @@ again:
 			  0, 0, &cached_state, GFP_NOFS);
 
 	ret = btrfs_set_extent_delalloc(inode, page_start, end,
-					&cached_state);
+					&cached_state, 0);
 	if (ret) {
 		unlock_extent_cached(io_tree, page_start, page_end,
 				     &cached_state, GFP_NOFS);
@@ -9093,7 +9183,7 @@ static int btrfs_truncate(struct inode *inode)
 
 	/* Migrate the slack space for the truncate to our reserve */
 	ret = btrfs_block_rsv_migrate(&root->fs_info->trans_block_rsv, rsv,
-				      min_size);
+				      min_size, 0);
 	BUG_ON(ret);
 
 	/*
@@ -9133,7 +9223,7 @@ static int btrfs_truncate(struct inode *inode)
 		}
 
 		ret = btrfs_block_rsv_migrate(&root->fs_info->trans_block_rsv,
-					      rsv, min_size);
+					      rsv, min_size, 0);
 		BUG_ON(ret);	/* shouldn't happen */
 		trans->block_rsv = rsv;
 	}
@@ -9154,7 +9244,6 @@ static int btrfs_truncate(struct inode *inode)
 		ret = btrfs_end_transaction(trans, root);
 		btrfs_btree_balance_dirty(root);
 	}
-
 out:
 	btrfs_free_block_rsv(root, rsv);
 
@@ -9303,8 +9392,9 @@ void btrfs_destroy_inode(struct inode *inode)
 		if (!ordered)
 			break;
 		else {
-			btrfs_err(root->fs_info, "found ordered extent %llu %llu on inode cleanup",
-				ordered->file_offset, ordered->len);
+			btrfs_err(root->fs_info,
+				  "found ordered extent %llu %llu on inode cleanup",
+				  ordered->file_offset, ordered->len);
 			btrfs_remove_ordered_extent(inode, ordered);
 			btrfs_put_ordered_extent(ordered);
 			btrfs_put_ordered_extent(ordered);
@@ -9363,25 +9453,25 @@ int btrfs_init_cachep(void)
 
 	btrfs_trans_handle_cachep = kmem_cache_create("btrfs_trans_handle",
 			sizeof(struct btrfs_trans_handle), 0,
-			SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD, NULL);
+			SLAB_TEMPORARY | SLAB_MEM_SPREAD, NULL);
 	if (!btrfs_trans_handle_cachep)
 		goto fail;
 
 	btrfs_transaction_cachep = kmem_cache_create("btrfs_transaction",
 			sizeof(struct btrfs_transaction), 0,
-			SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD, NULL);
+			SLAB_TEMPORARY | SLAB_MEM_SPREAD, NULL);
 	if (!btrfs_transaction_cachep)
 		goto fail;
 
 	btrfs_path_cachep = kmem_cache_create("btrfs_path",
 			sizeof(struct btrfs_path), 0,
-			SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD, NULL);
+			SLAB_MEM_SPREAD, NULL);
 	if (!btrfs_path_cachep)
 		goto fail;
 
 	btrfs_free_space_cachep = kmem_cache_create("btrfs_free_space",
 			sizeof(struct btrfs_free_space), 0,
-			SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD, NULL);
+			SLAB_MEM_SPREAD, NULL);
 	if (!btrfs_free_space_cachep)
 		goto fail;
 
@@ -9419,7 +9509,7 @@ static int btrfs_rename_exchange(struct inode *old_dir,
 	struct btrfs_root *dest = BTRFS_I(new_dir)->root;
 	struct inode *new_inode = new_dentry->d_inode;
 	struct inode *old_inode = old_dentry->d_inode;
-	struct timespec ctime = CURRENT_TIME;
+	struct timespec ctime = current_time(old_inode);
 	struct dentry *parent;
 	u64 old_ino = btrfs_ino(old_inode);
 	u64 new_ino = btrfs_ino(new_inode);
@@ -9531,7 +9621,7 @@ static int btrfs_rename_exchange(struct inode *old_dir,
 			ret = btrfs_update_inode(trans, root, old_inode);
 	}
 	if (ret) {
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 		goto out_fail;
 	}
 
@@ -9551,7 +9641,7 @@ static int btrfs_rename_exchange(struct inode *old_dir,
 			ret = btrfs_update_inode(trans, dest, new_inode);
 	}
 	if (ret) {
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 		goto out_fail;
 	}
 
@@ -9559,7 +9649,7 @@ static int btrfs_rename_exchange(struct inode *old_dir,
 			     new_dentry->d_name.name,
 			     new_dentry->d_name.len, 0, old_idx);
 	if (ret) {
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 		goto out_fail;
 	}
 
@@ -9567,7 +9657,7 @@ static int btrfs_rename_exchange(struct inode *old_dir,
 			     old_dentry->d_name.name,
 			     old_dentry->d_name.len, 0, new_idx);
 	if (ret) {
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 		goto out_fail;
 	}
 
@@ -9787,7 +9877,7 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 	inode_inc_iversion(old_inode);
 	old_dir->i_ctime = old_dir->i_mtime =
 	new_dir->i_ctime = new_dir->i_mtime =
-	old_inode->i_ctime = current_fs_time(old_dir->i_sb);
+	old_inode->i_ctime = current_time(old_dir);
 
 	if (old_dentry->d_parent != new_dentry->d_parent)
 		btrfs_record_unlink_dir(trans, old_dir, old_inode, 1);
@@ -9806,13 +9896,13 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 			ret = btrfs_update_inode(trans, root, old_inode);
 	}
 	if (ret) {
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 		goto out_fail;
 	}
 
 	if (new_inode) {
 		inode_inc_iversion(new_inode);
-		new_inode->i_ctime = current_fs_time(new_inode->i_sb);
+		new_inode->i_ctime = current_time(new_inode);
 		if (unlikely(btrfs_ino(new_inode) ==
 			     BTRFS_EMPTY_SUBVOL_DIR_OBJECTID)) {
 			root_objectid = BTRFS_I(new_inode)->location.objectid;
@@ -9830,7 +9920,7 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 		if (!ret && new_inode->i_nlink == 0)
 			ret = btrfs_orphan_add(trans, d_inode(new_dentry));
 		if (ret) {
-			btrfs_abort_transaction(trans, root, ret);
+			btrfs_abort_transaction(trans, ret);
 			goto out_fail;
 		}
 	}
@@ -9839,7 +9929,7 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 			     new_dentry->d_name.name,
 			     new_dentry->d_name.len, 0, index);
 	if (ret) {
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 		goto out_fail;
 	}
 
@@ -9859,7 +9949,7 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 						old_dentry);
 
 		if (ret) {
-			btrfs_abort_transaction(trans, root, ret);
+			btrfs_abort_transaction(trans, ret);
 			goto out_fail;
 		}
 	}
@@ -10246,6 +10336,7 @@ static int __btrfs_prealloc_file_range(struct inode *inode, int mode,
 	u64 last_alloc = (u64)-1;
 	int ret = 0;
 	bool own_trans = true;
+	u64 end = start + num_bytes - 1;
 
 	if (trans)
 		own_trans = false;
@@ -10267,8 +10358,8 @@ static int __btrfs_prealloc_file_range(struct inode *inode, int mode,
 		 * sized chunks.
 		 */
 		cur_bytes = min(cur_bytes, last_alloc);
-		ret = btrfs_reserve_extent(root, cur_bytes, min_size, 0,
-					   *alloc_hint, &ins, 1, 0);
+		ret = btrfs_reserve_extent(root, cur_bytes, cur_bytes,
+				min_size, 0, *alloc_hint, &ins, 1, 0);
 		if (ret) {
 			if (own_trans)
 				btrfs_end_transaction(trans, root);
@@ -10285,7 +10376,7 @@ static int __btrfs_prealloc_file_range(struct inode *inode, int mode,
 		if (ret) {
 			btrfs_free_reserved_extent(root, ins.objectid,
 						   ins.offset, 0);
-			btrfs_abort_transaction(trans, root, ret);
+			btrfs_abort_transaction(trans, ret);
 			if (own_trans)
 				btrfs_end_transaction(trans, root);
 			break;
@@ -10329,7 +10420,7 @@ next:
 		*alloc_hint = ins.objectid + ins.offset;
 
 		inode_inc_iversion(inode);
-		inode->i_ctime = current_fs_time(inode->i_sb);
+		inode->i_ctime = current_time(inode);
 		BTRFS_I(inode)->flags |= BTRFS_INODE_PREALLOC;
 		if (!(mode & FALLOC_FL_KEEP_SIZE) &&
 		    (actual_len > inode->i_size) &&
@@ -10345,7 +10436,7 @@ next:
 		ret = btrfs_update_inode(trans, root, inode);
 
 		if (ret) {
-			btrfs_abort_transaction(trans, root, ret);
+			btrfs_abort_transaction(trans, ret);
 			if (own_trans)
 				btrfs_end_transaction(trans, root);
 			break;
@@ -10354,6 +10445,9 @@ next:
 		if (own_trans)
 			btrfs_end_transaction(trans, root);
 	}
+	if (cur_offset < end)
+		btrfs_free_reserved_data_space(inode, cur_offset,
+			end - cur_offset + 1);
 	return ret;
 }
 
@@ -10466,21 +10560,6 @@ out_inode:
 
 }
 
-/* Inspired by filemap_check_errors() */
-int btrfs_inode_check_errors(struct inode *inode)
-{
-	int ret = 0;
-
-	if (test_bit(AS_ENOSPC, &inode->i_mapping->flags) &&
-	    test_and_clear_bit(AS_ENOSPC, &inode->i_mapping->flags))
-		ret = -ENOSPC;
-	if (test_bit(AS_EIO, &inode->i_mapping->flags) &&
-	    test_and_clear_bit(AS_EIO, &inode->i_mapping->flags))
-		ret = -EIO;
-
-	return ret;
-}
-
 static const struct inode_operations btrfs_dir_inode_operations = {
 	.getattr	= btrfs_getattr,
 	.lookup		= btrfs_lookup,
@@ -10489,14 +10568,11 @@ static const struct inode_operations btrfs_dir_inode_operations = {
 	.link		= btrfs_link,
 	.mkdir		= btrfs_mkdir,
 	.rmdir		= btrfs_rmdir,
-	.rename2	= btrfs_rename2,
+	.rename		= btrfs_rename2,
 	.symlink	= btrfs_symlink,
 	.setattr	= btrfs_setattr,
 	.mknod		= btrfs_mknod,
-	.setxattr	= generic_setxattr,
-	.getxattr	= generic_getxattr,
 	.listxattr	= btrfs_listxattr,
-	.removexattr	= generic_removexattr,
 	.permission	= btrfs_permission,
 	.get_acl	= btrfs_get_acl,
 	.set_acl	= btrfs_set_acl,
@@ -10514,7 +10590,7 @@ static const struct inode_operations btrfs_dir_ro_inode_operations = {
 static const struct file_operations btrfs_dir_file_operations = {
 	.llseek		= generic_file_llseek,
 	.read		= generic_read_dir,
-	.iterate	= btrfs_real_readdir,
+	.iterate_shared	= btrfs_real_readdir,
 	.unlocked_ioctl	= btrfs_ioctl,
 #ifdef CONFIG_COMPAT
 	.compat_ioctl	= btrfs_compat_ioctl,
@@ -10570,10 +10646,7 @@ static const struct address_space_operations btrfs_symlink_aops = {
 static const struct inode_operations btrfs_file_inode_operations = {
 	.getattr	= btrfs_getattr,
 	.setattr	= btrfs_setattr,
-	.setxattr	= generic_setxattr,
-	.getxattr	= generic_getxattr,
 	.listxattr      = btrfs_listxattr,
-	.removexattr	= generic_removexattr,
 	.permission	= btrfs_permission,
 	.fiemap		= btrfs_fiemap,
 	.get_acl	= btrfs_get_acl,
@@ -10584,10 +10657,7 @@ static const struct inode_operations btrfs_special_inode_operations = {
 	.getattr	= btrfs_getattr,
 	.setattr	= btrfs_setattr,
 	.permission	= btrfs_permission,
-	.setxattr	= generic_setxattr,
-	.getxattr	= generic_getxattr,
 	.listxattr	= btrfs_listxattr,
-	.removexattr	= generic_removexattr,
 	.get_acl	= btrfs_get_acl,
 	.set_acl	= btrfs_set_acl,
 	.update_time	= btrfs_update_time,
@@ -10598,10 +10668,7 @@ static const struct inode_operations btrfs_symlink_inode_operations = {
 	.getattr	= btrfs_getattr,
 	.setattr	= btrfs_setattr,
 	.permission	= btrfs_permission,
-	.setxattr	= generic_setxattr,
-	.getxattr	= generic_getxattr,
 	.listxattr	= btrfs_listxattr,
-	.removexattr	= generic_removexattr,
 	.update_time	= btrfs_update_time,
 };
 
diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c
index 05173563e4a6..18e1aa0f85f5 100644
--- a/fs/btrfs/ioctl.c
+++ b/fs/btrfs/ioctl.c
@@ -349,7 +349,7 @@ static int btrfs_ioctl_setflags(struct file *file, void __user *arg)
 
 	btrfs_update_iflags(inode);
 	inode_inc_iversion(inode);
-	inode->i_ctime = current_fs_time(inode->i_sb);
+	inode->i_ctime = current_time(inode);
 	ret = btrfs_update_inode(trans, root, inode);
 
 	btrfs_end_transaction(trans, root);
@@ -445,7 +445,7 @@ static noinline int create_subvol(struct inode *dir,
 	struct btrfs_root *root = BTRFS_I(dir)->root;
 	struct btrfs_root *new_root;
 	struct btrfs_block_rsv block_rsv;
-	struct timespec cur_time = current_fs_time(dir->i_sb);
+	struct timespec cur_time = current_time(dir);
 	struct inode *inode;
 	int ret;
 	int err;
@@ -561,7 +561,7 @@ static noinline int create_subvol(struct inode *dir,
 	new_root = btrfs_read_fs_root_no_name(root->fs_info, &key);
 	if (IS_ERR(new_root)) {
 		ret = PTR_ERR(new_root);
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 		goto fail;
 	}
 
@@ -570,7 +570,7 @@ static noinline int create_subvol(struct inode *dir,
 	ret = btrfs_create_subvol_root(trans, new_root, root, new_dirid);
 	if (ret) {
 		/* We potentially lose an unused inode item here */
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 		goto fail;
 	}
 
@@ -583,7 +583,7 @@ static noinline int create_subvol(struct inode *dir,
 	 */
 	ret = btrfs_set_inode_index(dir, &index);
 	if (ret) {
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 		goto fail;
 	}
 
@@ -591,7 +591,7 @@ static noinline int create_subvol(struct inode *dir,
 				    name, namelen, dir, &key,
 				    BTRFS_FT_DIR, index);
 	if (ret) {
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 		goto fail;
 	}
 
@@ -608,7 +608,7 @@ static noinline int create_subvol(struct inode *dir,
 				  root_item->uuid, BTRFS_UUID_KEY_SUBVOL,
 				  objectid);
 	if (ret)
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 
 fail:
 	kfree(root_item);
@@ -1634,6 +1634,9 @@ static noinline int btrfs_ioctl_snap_create_transid(struct file *file,
 	int namelen;
 	int ret = 0;
 
+	if (!S_ISDIR(file_inode(file)->i_mode))
+		return -ENOTDIR;
+
 	ret = mnt_want_write_file(file);
 	if (ret)
 		goto out;
@@ -1691,6 +1694,9 @@ static noinline int btrfs_ioctl_snap_create(struct file *file,
 	struct btrfs_ioctl_vol_args *vol_args;
 	int ret;
 
+	if (!S_ISDIR(file_inode(file)->i_mode))
+		return -ENOTDIR;
+
 	vol_args = memdup_user(arg, sizeof(*vol_args));
 	if (IS_ERR(vol_args))
 		return PTR_ERR(vol_args);
@@ -1714,6 +1720,9 @@ static noinline int btrfs_ioctl_snap_create_v2(struct file *file,
 	bool readonly = false;
 	struct btrfs_qgroup_inherit *inherit = NULL;
 
+	if (!S_ISDIR(file_inode(file)->i_mode))
+		return -ENOTDIR;
+
 	vol_args = memdup_user(arg, sizeof(*vol_args));
 	if (IS_ERR(vol_args))
 		return PTR_ERR(vol_args);
@@ -1894,8 +1903,9 @@ static noinline int may_destroy_subvol(struct btrfs_root *root)
 		btrfs_dir_item_key_to_cpu(path->nodes[0], di, &key);
 		if (key.objectid == root->root_key.objectid) {
 			ret = -EPERM;
-			btrfs_err(root->fs_info, "deleting default subvolume "
-				  "%llu is not allowed", key.objectid);
+			btrfs_err(root->fs_info,
+				  "deleting default subvolume %llu is not allowed",
+				  key.objectid);
 			goto out;
 		}
 		btrfs_release_path(path);
@@ -1948,8 +1958,7 @@ static noinline int key_in_sk(struct btrfs_key *key,
 	return 1;
 }
 
-static noinline int copy_to_sk(struct btrfs_root *root,
-			       struct btrfs_path *path,
+static noinline int copy_to_sk(struct btrfs_path *path,
 			       struct btrfs_key *key,
 			       struct btrfs_ioctl_search_key *sk,
 			       size_t *buf_size,
@@ -2120,7 +2129,7 @@ static noinline int search_ioctl(struct inode *inode,
 				ret = 0;
 			goto err;
 		}
-		ret = copy_to_sk(root, path, &key, sk, buf_size, ubuf,
+		ret = copy_to_sk(path, &key, sk, buf_size, ubuf,
 				 &sk_offset, &num_found);
 		btrfs_release_path(path);
 		if (ret)
@@ -2358,6 +2367,9 @@ static noinline int btrfs_ioctl_snap_destroy(struct file *file,
 	int ret;
 	int err = 0;
 
+	if (!S_ISDIR(dir->i_mode))
+		return -ENOTDIR;
+
 	vol_args = memdup_user(arg, sizeof(*vol_args));
 	if (IS_ERR(vol_args))
 		return PTR_ERR(vol_args);
@@ -2406,7 +2418,7 @@ static noinline int btrfs_ioctl_snap_destroy(struct file *file,
 		 * rmdir(2).
 		 */
 		err = -EPERM;
-		if (!btrfs_test_opt(root, USER_SUBVOL_RM_ALLOWED))
+		if (!btrfs_test_opt(root->fs_info, USER_SUBVOL_RM_ALLOWED))
 			goto out_dput;
 
 		/*
@@ -2489,7 +2501,7 @@ static noinline int btrfs_ioctl_snap_destroy(struct file *file,
 				dentry->d_name.len);
 	if (ret) {
 		err = ret;
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 		goto out_end_trans;
 	}
 
@@ -2505,7 +2517,7 @@ static noinline int btrfs_ioctl_snap_destroy(struct file *file,
 					root->fs_info->tree_root,
 					dest->root_key.objectid);
 		if (ret) {
-			btrfs_abort_transaction(trans, root, ret);
+			btrfs_abort_transaction(trans, ret);
 			err = ret;
 			goto out_end_trans;
 		}
@@ -2515,7 +2527,7 @@ static noinline int btrfs_ioctl_snap_destroy(struct file *file,
 				  dest->root_item.uuid, BTRFS_UUID_KEY_SUBVOL,
 				  dest->root_key.objectid);
 	if (ret && ret != -ENOENT) {
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 		err = ret;
 		goto out_end_trans;
 	}
@@ -2525,7 +2537,7 @@ static noinline int btrfs_ioctl_snap_destroy(struct file *file,
 					  BTRFS_UUID_KEY_RECEIVED_SUBVOL,
 					  dest->root_key.objectid);
 		if (ret && ret != -ENOENT) {
-			btrfs_abort_transaction(trans, root, ret);
+			btrfs_abort_transaction(trans, ret);
 			err = ret;
 			goto out_end_trans;
 		}
@@ -3280,7 +3292,7 @@ static int clone_finish_inode_update(struct btrfs_trans_handle *trans,
 
 	inode_inc_iversion(inode);
 	if (!no_time_update)
-		inode->i_mtime = inode->i_ctime = current_fs_time(inode->i_sb);
+		inode->i_mtime = inode->i_ctime = current_time(inode);
 	/*
 	 * We round up to the block size at eof when determining which
 	 * extents to clone above, but shouldn't round up the file size.
@@ -3292,7 +3304,7 @@ static int clone_finish_inode_update(struct btrfs_trans_handle *trans,
 
 	ret = btrfs_update_inode(trans, root, inode);
 	if (ret) {
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 		btrfs_end_transaction(trans, root);
 		goto out;
 	}
@@ -3694,7 +3706,7 @@ process_slot:
 				if (ret) {
 					if (ret != -EOPNOTSUPP)
 						btrfs_abort_transaction(trans,
-								root, ret);
+									ret);
 					btrfs_end_transaction(trans, root);
 					goto out;
 				}
@@ -3702,8 +3714,7 @@ process_slot:
 				ret = btrfs_insert_empty_item(trans, root, path,
 							      &new_key, size);
 				if (ret) {
-					btrfs_abort_transaction(trans, root,
-								ret);
+					btrfs_abort_transaction(trans, ret);
 					btrfs_end_transaction(trans, root);
 					goto out;
 				}
@@ -3735,7 +3746,6 @@ process_slot:
 							new_key.offset - datao);
 					if (ret) {
 						btrfs_abort_transaction(trans,
-									root,
 									ret);
 						btrfs_end_transaction(trans,
 								      root);
@@ -3772,7 +3782,6 @@ process_slot:
 				if (ret) {
 					if (ret != -EOPNOTSUPP)
 						btrfs_abort_transaction(trans,
-									root,
 									ret);
 					btrfs_end_transaction(trans, root);
 					goto out;
@@ -3828,7 +3837,7 @@ process_slot:
 					 last_dest_end, destoff + len, 1);
 		if (ret) {
 			if (ret != -EOPNOTSUPP)
-				btrfs_abort_transaction(trans, root, ret);
+				btrfs_abort_transaction(trans, ret);
 			btrfs_end_transaction(trans, root);
 			goto out;
 		}
@@ -4089,8 +4098,8 @@ static long btrfs_ioctl_default_subvol(struct file *file, void __user *argp)
 	if (IS_ERR_OR_NULL(di)) {
 		btrfs_free_path(path);
 		btrfs_end_transaction(trans, root);
-		btrfs_err(new_root->fs_info, "Umm, you don't have the default dir"
-			   "item, this isn't going to work");
+		btrfs_err(new_root->fs_info,
+			  "Umm, you don't have the default diritem, this isn't going to work");
 		ret = -ENOENT;
 		goto out;
 	}
@@ -5088,7 +5097,7 @@ static long btrfs_ioctl_quota_rescan_wait(struct file *file, void __user *arg)
 	if (!capable(CAP_SYS_ADMIN))
 		return -EPERM;
 
-	return btrfs_qgroup_wait_for_completion(root->fs_info);
+	return btrfs_qgroup_wait_for_completion(root->fs_info, true);
 }
 
 static long _btrfs_ioctl_set_received_subvol(struct file *file,
@@ -5098,7 +5107,7 @@ static long _btrfs_ioctl_set_received_subvol(struct file *file,
 	struct btrfs_root *root = BTRFS_I(inode)->root;
 	struct btrfs_root_item *root_item = &root->root_item;
 	struct btrfs_trans_handle *trans;
-	struct timespec ct = current_fs_time(inode->i_sb);
+	struct timespec ct = current_time(inode);
 	int ret = 0;
 	int received_uuid_changed;
 
@@ -5164,13 +5173,13 @@ static long _btrfs_ioctl_set_received_subvol(struct file *file,
 					  BTRFS_UUID_KEY_RECEIVED_SUBVOL,
 					  root->root_key.objectid);
 		if (ret < 0 && ret != -EEXIST) {
-			btrfs_abort_transaction(trans, root, ret);
+			btrfs_abort_transaction(trans, ret);
 			goto out;
 		}
 	}
 	ret = btrfs_commit_transaction(trans, root);
 	if (ret < 0) {
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 		goto out;
 	}
 
@@ -5299,8 +5308,9 @@ static int btrfs_ioctl_set_fslabel(struct file *file, void __user *arg)
 		return -EFAULT;
 
 	if (strnlen(label, BTRFS_LABEL_SIZE) == BTRFS_LABEL_SIZE) {
-		btrfs_err(root->fs_info, "unable to set label with more than %d bytes",
-		       BTRFS_LABEL_SIZE - 1);
+		btrfs_err(root->fs_info,
+			  "unable to set label with more than %d bytes",
+			  BTRFS_LABEL_SIZE - 1);
 		return -EINVAL;
 	}
 
diff --git a/fs/btrfs/lzo.c b/fs/btrfs/lzo.c
index 1adfbe7be6b8..48655da0f4ca 100644
--- a/fs/btrfs/lzo.c
+++ b/fs/btrfs/lzo.c
@@ -141,7 +141,7 @@ static int lzo_compress_pages(struct list_head *ws,
 		ret = lzo1x_1_compress(data_in, in_len, workspace->cbuf,
 				       &out_len, workspace->mem);
 		if (ret != LZO_E_OK) {
-			printk(KERN_DEBUG "BTRFS: deflate in loop returned %d\n",
+			pr_debug("BTRFS: deflate in loop returned %d\n",
 			       ret);
 			ret = -EIO;
 			goto out;
@@ -356,7 +356,7 @@ cont:
 		if (need_unmap)
 			kunmap(pages_in[page_in_index - 1]);
 		if (ret != LZO_E_OK) {
-			printk(KERN_WARNING "BTRFS: decompress failed\n");
+			pr_warn("BTRFS: decompress failed\n");
 			ret = -EIO;
 			break;
 		}
@@ -402,7 +402,7 @@ static int lzo_decompress(struct list_head *ws, unsigned char *data_in,
 	out_len = PAGE_SIZE;
 	ret = lzo1x_decompress_safe(data_in, in_len, workspace->buf, &out_len);
 	if (ret != LZO_E_OK) {
-		printk(KERN_WARNING "BTRFS: decompress failed!\n");
+		pr_warn("BTRFS: decompress failed!\n");
 		ret = -EIO;
 		goto out;
 	}
diff --git a/fs/btrfs/ordered-data.c b/fs/btrfs/ordered-data.c
index 559170464d7c..b2d1e95de7be 100644
--- a/fs/btrfs/ordered-data.c
+++ b/fs/btrfs/ordered-data.c
@@ -67,8 +67,8 @@ static void ordered_data_tree_panic(struct inode *inode, int errno,
 					       u64 offset)
 {
 	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
-	btrfs_panic(fs_info, errno, "Inconsistency in ordered tree at offset "
-		    "%llu", offset);
+	btrfs_panic(fs_info, errno,
+		    "Inconsistency in ordered tree at offset %llu", offset);
 }
 
 /*
@@ -718,12 +718,13 @@ int btrfs_wait_ordered_extents(struct btrfs_root *root, int nr,
 	return count;
 }
 
-void btrfs_wait_ordered_roots(struct btrfs_fs_info *fs_info, int nr,
+int btrfs_wait_ordered_roots(struct btrfs_fs_info *fs_info, int nr,
 			      const u64 range_start, const u64 range_len)
 {
 	struct btrfs_root *root;
 	struct list_head splice;
 	int done;
+	int total_done = 0;
 
 	INIT_LIST_HEAD(&splice);
 
@@ -742,6 +743,7 @@ void btrfs_wait_ordered_roots(struct btrfs_fs_info *fs_info, int nr,
 		done = btrfs_wait_ordered_extents(root, nr,
 						  range_start, range_len);
 		btrfs_put_fs_root(root);
+		total_done += done;
 
 		spin_lock(&fs_info->ordered_root_lock);
 		if (nr != -1) {
@@ -752,6 +754,8 @@ void btrfs_wait_ordered_roots(struct btrfs_fs_info *fs_info, int nr,
 	list_splice_tail(&splice, &fs_info->ordered_roots);
 	spin_unlock(&fs_info->ordered_root_lock);
 	mutex_unlock(&fs_info->ordered_operations_mutex);
+
+	return total_done;
 }
 
 /*
@@ -964,6 +968,7 @@ int btrfs_ordered_update_i_size(struct inode *inode, u64 offset,
 	struct rb_node *prev = NULL;
 	struct btrfs_ordered_extent *test;
 	int ret = 1;
+	u64 orig_offset = offset;
 
 	spin_lock_irq(&tree->lock);
 	if (ordered) {
@@ -979,7 +984,7 @@ int btrfs_ordered_update_i_size(struct inode *inode, u64 offset,
 
 	/* truncate file */
 	if (disk_i_size > i_size) {
-		BTRFS_I(inode)->disk_i_size = i_size;
+		BTRFS_I(inode)->disk_i_size = orig_offset;
 		ret = 0;
 		goto out;
 	}
@@ -1117,7 +1122,7 @@ int __init ordered_data_init(void)
 {
 	btrfs_ordered_extent_cache = kmem_cache_create("btrfs_ordered_extent",
 				     sizeof(struct btrfs_ordered_extent), 0,
-				     SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD,
+				     SLAB_MEM_SPREAD,
 				     NULL);
 	if (!btrfs_ordered_extent_cache)
 		return -ENOMEM;
diff --git a/fs/btrfs/ordered-data.h b/fs/btrfs/ordered-data.h
index 2049c9be85ee..451507776ff5 100644
--- a/fs/btrfs/ordered-data.h
+++ b/fs/btrfs/ordered-data.h
@@ -199,7 +199,7 @@ int btrfs_find_ordered_sum(struct inode *inode, u64 offset, u64 disk_bytenr,
 			   u32 *sum, int len);
 int btrfs_wait_ordered_extents(struct btrfs_root *root, int nr,
 			       const u64 range_start, const u64 range_len);
-void btrfs_wait_ordered_roots(struct btrfs_fs_info *fs_info, int nr,
+int btrfs_wait_ordered_roots(struct btrfs_fs_info *fs_info, int nr,
 			      const u64 range_start, const u64 range_len);
 void btrfs_get_logged_extents(struct inode *inode,
 			      struct list_head *logged_list,
diff --git a/fs/btrfs/print-tree.c b/fs/btrfs/print-tree.c
index 147dc6ca5de1..438575ea8d25 100644
--- a/fs/btrfs/print-tree.c
+++ b/fs/btrfs/print-tree.c
@@ -24,12 +24,11 @@ static void print_chunk(struct extent_buffer *eb, struct btrfs_chunk *chunk)
 {
 	int num_stripes = btrfs_chunk_num_stripes(eb, chunk);
 	int i;
-	printk(KERN_INFO "\t\tchunk length %llu owner %llu type %llu "
-	       "num_stripes %d\n",
+	pr_info("\t\tchunk length %llu owner %llu type %llu num_stripes %d\n",
 	       btrfs_chunk_length(eb, chunk), btrfs_chunk_owner(eb, chunk),
 	       btrfs_chunk_type(eb, chunk), num_stripes);
 	for (i = 0 ; i < num_stripes ; i++) {
-		printk(KERN_INFO "\t\t\tstripe %d devid %llu offset %llu\n", i,
+		pr_info("\t\t\tstripe %d devid %llu offset %llu\n", i,
 		      btrfs_stripe_devid_nr(eb, chunk, i),
 		      btrfs_stripe_offset_nr(eb, chunk, i));
 	}
@@ -37,8 +36,7 @@ static void print_chunk(struct extent_buffer *eb, struct btrfs_chunk *chunk)
 static void print_dev_item(struct extent_buffer *eb,
 			   struct btrfs_dev_item *dev_item)
 {
-	printk(KERN_INFO "\t\tdev item devid %llu "
-	       "total_bytes %llu bytes used %llu\n",
+	pr_info("\t\tdev item devid %llu total_bytes %llu bytes used %llu\n",
 	       btrfs_device_id(eb, dev_item),
 	       btrfs_device_total_bytes(eb, dev_item),
 	       btrfs_device_bytes_used(eb, dev_item));
@@ -46,8 +44,7 @@ static void print_dev_item(struct extent_buffer *eb,
 static void print_extent_data_ref(struct extent_buffer *eb,
 				  struct btrfs_extent_data_ref *ref)
 {
-	printk(KERN_INFO "\t\textent data backref root %llu "
-	       "objectid %llu offset %llu count %u\n",
+	pr_info("\t\textent data backref root %llu objectid %llu offset %llu count %u\n",
 	       btrfs_extent_data_ref_root(eb, ref),
 	       btrfs_extent_data_ref_objectid(eb, ref),
 	       btrfs_extent_data_ref_offset(eb, ref),
@@ -72,7 +69,7 @@ static void print_extent_item(struct extent_buffer *eb, int slot, int type)
 		struct btrfs_extent_item_v0 *ei0;
 		BUG_ON(item_size != sizeof(*ei0));
 		ei0 = btrfs_item_ptr(eb, slot, struct btrfs_extent_item_v0);
-		printk(KERN_INFO "\t\textent refs %u\n",
+		pr_info("\t\textent refs %u\n",
 		       btrfs_extent_refs_v0(eb, ei0));
 		return;
 #else
@@ -83,7 +80,7 @@ static void print_extent_item(struct extent_buffer *eb, int slot, int type)
 	ei = btrfs_item_ptr(eb, slot, struct btrfs_extent_item);
 	flags = btrfs_extent_flags(eb, ei);
 
-	printk(KERN_INFO "\t\textent refs %llu gen %llu flags %llu\n",
+	pr_info("\t\textent refs %llu gen %llu flags %llu\n",
 	       btrfs_extent_refs(eb, ei), btrfs_extent_generation(eb, ei),
 	       flags);
 
@@ -92,8 +89,7 @@ static void print_extent_item(struct extent_buffer *eb, int slot, int type)
 		struct btrfs_tree_block_info *info;
 		info = (struct btrfs_tree_block_info *)(ei + 1);
 		btrfs_tree_block_key(eb, info, &key);
-		printk(KERN_INFO "\t\ttree block key (%llu %u %llu) "
-		       "level %d\n",
+		pr_info("\t\ttree block key (%llu %u %llu) level %d\n",
 		       btrfs_disk_key_objectid(&key), key.type,
 		       btrfs_disk_key_offset(&key),
 		       btrfs_tree_block_level(eb, info));
@@ -110,12 +106,10 @@ static void print_extent_item(struct extent_buffer *eb, int slot, int type)
 		offset = btrfs_extent_inline_ref_offset(eb, iref);
 		switch (type) {
 		case BTRFS_TREE_BLOCK_REF_KEY:
-			printk(KERN_INFO "\t\ttree block backref "
-				"root %llu\n", offset);
+			pr_info("\t\ttree block backref root %llu\n", offset);
 			break;
 		case BTRFS_SHARED_BLOCK_REF_KEY:
-			printk(KERN_INFO "\t\tshared block backref "
-				"parent %llu\n", offset);
+			pr_info("\t\tshared block backref parent %llu\n", offset);
 			break;
 		case BTRFS_EXTENT_DATA_REF_KEY:
 			dref = (struct btrfs_extent_data_ref *)(&iref->offset);
@@ -123,8 +117,7 @@ static void print_extent_item(struct extent_buffer *eb, int slot, int type)
 			break;
 		case BTRFS_SHARED_DATA_REF_KEY:
 			sref = (struct btrfs_shared_data_ref *)(iref + 1);
-			printk(KERN_INFO "\t\tshared data backref "
-			       "parent %llu count %u\n",
+			pr_info("\t\tshared data backref parent %llu count %u\n",
 			       offset, btrfs_shared_data_ref_count(eb, sref));
 			break;
 		default:
@@ -141,8 +134,7 @@ static void print_extent_ref_v0(struct extent_buffer *eb, int slot)
 	struct btrfs_extent_ref_v0 *ref0;
 
 	ref0 = btrfs_item_ptr(eb, slot, struct btrfs_extent_ref_v0);
-	printk("\t\textent back ref root %llu gen %llu "
-		"owner %llu num_refs %lu\n",
+	printk("\t\textent back ref root %llu gen %llu owner %llu num_refs %lu\n",
 		btrfs_ref_root_v0(eb, ref0),
 		btrfs_ref_generation_v0(eb, ref0),
 		btrfs_ref_objectid_v0(eb, ref0),
@@ -162,7 +154,7 @@ static void print_uuid_item(struct extent_buffer *l, unsigned long offset,
 		__le64 subvol_id;
 
 		read_extent_buffer(l, &subvol_id, offset, sizeof(subvol_id));
-		printk(KERN_INFO "\t\tsubvol_id %llu\n",
+		pr_info("\t\tsubvol_id %llu\n",
 		       (unsigned long long)le64_to_cpu(subvol_id));
 		item_size -= sizeof(u64);
 		offset += sizeof(u64);
@@ -196,15 +188,13 @@ void btrfs_print_leaf(struct btrfs_root *root, struct extent_buffer *l)
 		item = btrfs_item_nr(i);
 		btrfs_item_key_to_cpu(l, &key, i);
 		type = key.type;
-		printk(KERN_INFO "\titem %d key (%llu %u %llu) itemoff %d "
-		       "itemsize %d\n",
+		pr_info("\titem %d key (%llu %u %llu) itemoff %d itemsize %d\n",
 			i, key.objectid, type, key.offset,
 			btrfs_item_offset(l, item), btrfs_item_size(l, item));
 		switch (type) {
 		case BTRFS_INODE_ITEM_KEY:
 			ii = btrfs_item_ptr(l, i, struct btrfs_inode_item);
-			printk(KERN_INFO "\t\tinode generation %llu size %llu "
-			       "mode %o\n",
+			pr_info("\t\tinode generation %llu size %llu mode %o\n",
 			       btrfs_inode_generation(l, ii),
 			       btrfs_inode_size(l, ii),
 			       btrfs_inode_mode(l, ii));
@@ -212,13 +202,13 @@ void btrfs_print_leaf(struct btrfs_root *root, struct extent_buffer *l)
 		case BTRFS_DIR_ITEM_KEY:
 			di = btrfs_item_ptr(l, i, struct btrfs_dir_item);
 			btrfs_dir_item_key_to_cpu(l, di, &found_key);
-			printk(KERN_INFO "\t\tdir oid %llu type %u\n",
+			pr_info("\t\tdir oid %llu type %u\n",
 				found_key.objectid,
 				btrfs_dir_type(l, di));
 			break;
 		case BTRFS_ROOT_ITEM_KEY:
 			ri = btrfs_item_ptr(l, i, struct btrfs_root_item);
-			printk(KERN_INFO "\t\troot data bytenr %llu refs %u\n",
+			pr_info("\t\troot data bytenr %llu refs %u\n",
 				btrfs_disk_root_bytenr(l, ri),
 				btrfs_disk_root_refs(l, ri));
 			break;
@@ -227,10 +217,10 @@ void btrfs_print_leaf(struct btrfs_root *root, struct extent_buffer *l)
 			print_extent_item(l, i, type);
 			break;
 		case BTRFS_TREE_BLOCK_REF_KEY:
-			printk(KERN_INFO "\t\ttree block backref\n");
+			pr_info("\t\ttree block backref\n");
 			break;
 		case BTRFS_SHARED_BLOCK_REF_KEY:
-			printk(KERN_INFO "\t\tshared block backref\n");
+			pr_info("\t\tshared block backref\n");
 			break;
 		case BTRFS_EXTENT_DATA_REF_KEY:
 			dref = btrfs_item_ptr(l, i,
@@ -240,7 +230,7 @@ void btrfs_print_leaf(struct btrfs_root *root, struct extent_buffer *l)
 		case BTRFS_SHARED_DATA_REF_KEY:
 			sref = btrfs_item_ptr(l, i,
 					      struct btrfs_shared_data_ref);
-			printk(KERN_INFO "\t\tshared data backref count %u\n",
+			pr_info("\t\tshared data backref count %u\n",
 			       btrfs_shared_data_ref_count(l, sref));
 			break;
 		case BTRFS_EXTENT_DATA_KEY:
@@ -248,17 +238,14 @@ void btrfs_print_leaf(struct btrfs_root *root, struct extent_buffer *l)
 					    struct btrfs_file_extent_item);
 			if (btrfs_file_extent_type(l, fi) ==
 			    BTRFS_FILE_EXTENT_INLINE) {
-				printk(KERN_INFO "\t\tinline extent data "
-				       "size %u\n",
+				pr_info("\t\tinline extent data size %u\n",
 				       btrfs_file_extent_inline_len(l, i, fi));
 				break;
 			}
-			printk(KERN_INFO "\t\textent data disk bytenr %llu "
-			       "nr %llu\n",
+			pr_info("\t\textent data disk bytenr %llu nr %llu\n",
 			       btrfs_file_extent_disk_bytenr(l, fi),
 			       btrfs_file_extent_disk_num_bytes(l, fi));
-			printk(KERN_INFO "\t\textent data offset %llu "
-			       "nr %llu ram %llu\n",
+			pr_info("\t\textent data offset %llu nr %llu ram %llu\n",
 			       btrfs_file_extent_offset(l, fi),
 			       btrfs_file_extent_num_bytes(l, fi),
 			       btrfs_file_extent_ram_bytes(l, fi));
@@ -273,7 +260,7 @@ void btrfs_print_leaf(struct btrfs_root *root, struct extent_buffer *l)
 		case BTRFS_BLOCK_GROUP_ITEM_KEY:
 			bi = btrfs_item_ptr(l, i,
 					    struct btrfs_block_group_item);
-			printk(KERN_INFO "\t\tblock group used %llu\n",
+			pr_info("\t\tblock group used %llu\n",
 			       btrfs_disk_block_group_used(l, bi));
 			break;
 		case BTRFS_CHUNK_ITEM_KEY:
@@ -287,38 +274,36 @@ void btrfs_print_leaf(struct btrfs_root *root, struct extent_buffer *l)
 		case BTRFS_DEV_EXTENT_KEY:
 			dev_extent = btrfs_item_ptr(l, i,
 						    struct btrfs_dev_extent);
-			printk(KERN_INFO "\t\tdev extent chunk_tree %llu\n"
-			       "\t\tchunk objectid %llu chunk offset %llu "
-			       "length %llu\n",
+			pr_info("\t\tdev extent chunk_tree %llu\n\t\tchunk objectid %llu chunk offset %llu length %llu\n",
 			       btrfs_dev_extent_chunk_tree(l, dev_extent),
 			       btrfs_dev_extent_chunk_objectid(l, dev_extent),
 			       btrfs_dev_extent_chunk_offset(l, dev_extent),
 			       btrfs_dev_extent_length(l, dev_extent));
 			break;
 		case BTRFS_PERSISTENT_ITEM_KEY:
-			printk(KERN_INFO "\t\tpersistent item objectid %llu offset %llu\n",
+			pr_info("\t\tpersistent item objectid %llu offset %llu\n",
 					key.objectid, key.offset);
 			switch (key.objectid) {
 			case BTRFS_DEV_STATS_OBJECTID:
-				printk(KERN_INFO "\t\tdevice stats\n");
+				pr_info("\t\tdevice stats\n");
 				break;
 			default:
-				printk(KERN_INFO "\t\tunknown persistent item\n");
+				pr_info("\t\tunknown persistent item\n");
 			}
 			break;
 		case BTRFS_TEMPORARY_ITEM_KEY:
-			printk(KERN_INFO "\t\ttemporary item objectid %llu offset %llu\n",
+			pr_info("\t\ttemporary item objectid %llu offset %llu\n",
 					key.objectid, key.offset);
 			switch (key.objectid) {
 			case BTRFS_BALANCE_OBJECTID:
-				printk(KERN_INFO "\t\tbalance status\n");
+				pr_info("\t\tbalance status\n");
 				break;
 			default:
-				printk(KERN_INFO "\t\tunknown temporary item\n");
+				pr_info("\t\tunknown temporary item\n");
 			}
 			break;
 		case BTRFS_DEV_REPLACE_KEY:
-			printk(KERN_INFO "\t\tdev replace\n");
+			pr_info("\t\tdev replace\n");
 			break;
 		case BTRFS_UUID_KEY_SUBVOL:
 		case BTRFS_UUID_KEY_RECEIVED_SUBVOL:
@@ -343,12 +328,13 @@ void btrfs_print_tree(struct btrfs_root *root, struct extent_buffer *c)
 		btrfs_print_leaf(root, c);
 		return;
 	}
-	btrfs_info(root->fs_info, "node %llu level %d total ptrs %d free spc %u",
-		btrfs_header_bytenr(c), level, nr,
-		(u32)BTRFS_NODEPTRS_PER_BLOCK(root) - nr);
+	btrfs_info(root->fs_info,
+		   "node %llu level %d total ptrs %d free spc %u",
+		   btrfs_header_bytenr(c), level, nr,
+		   (u32)BTRFS_NODEPTRS_PER_BLOCK(root) - nr);
 	for (i = 0; i < nr; i++) {
 		btrfs_node_key_to_cpu(c, &key, i);
-		printk(KERN_INFO "\tkey %d (%llu %u %llu) block %llu\n",
+		pr_info("\tkey %d (%llu %u %llu) block %llu\n",
 		       i, key.objectid, key.type, key.offset,
 		       btrfs_node_blockptr(c, i));
 	}
@@ -356,6 +342,13 @@ void btrfs_print_tree(struct btrfs_root *root, struct extent_buffer *c)
 		struct extent_buffer *next = read_tree_block(root,
 					btrfs_node_blockptr(c, i),
 					btrfs_node_ptr_generation(c, i));
+		if (IS_ERR(next)) {
+			continue;
+		} else if (!extent_buffer_uptodate(next)) {
+			free_extent_buffer(next);
+			continue;
+		}
+
 		if (btrfs_is_leaf(next) &&
 		   level != 1)
 			BUG();
diff --git a/fs/btrfs/props.c b/fs/btrfs/props.c
index 36992128c746..cf0b444ac4f3 100644
--- a/fs/btrfs/props.c
+++ b/fs/btrfs/props.c
@@ -350,6 +350,7 @@ int btrfs_subvol_inherit_props(struct btrfs_trans_handle *trans,
 			       struct btrfs_root *root,
 			       struct btrfs_root *parent_root)
 {
+	struct super_block *sb = root->fs_info->sb;
 	struct btrfs_key key;
 	struct inode *parent_inode, *child_inode;
 	int ret;
@@ -358,12 +359,11 @@ int btrfs_subvol_inherit_props(struct btrfs_trans_handle *trans,
 	key.type = BTRFS_INODE_ITEM_KEY;
 	key.offset = 0;
 
-	parent_inode = btrfs_iget(parent_root->fs_info->sb, &key,
-				  parent_root, NULL);
+	parent_inode = btrfs_iget(sb, &key, parent_root, NULL);
 	if (IS_ERR(parent_inode))
 		return PTR_ERR(parent_inode);
 
-	child_inode = btrfs_iget(root->fs_info->sb, &key, root, NULL);
+	child_inode = btrfs_iget(sb, &key, root, NULL);
 	if (IS_ERR(child_inode)) {
 		iput(parent_inode);
 		return PTR_ERR(child_inode);
diff --git a/fs/btrfs/qgroup.c b/fs/btrfs/qgroup.c
index 9d4c05b14f6e..11f4fffe503e 100644
--- a/fs/btrfs/qgroup.c
+++ b/fs/btrfs/qgroup.c
@@ -309,7 +309,7 @@ int btrfs_read_qgroup_config(struct btrfs_fs_info *fs_info)
 	u64 flags = 0;
 	u64 rescan_progress = 0;
 
-	if (!fs_info->quota_enabled)
+	if (!test_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags))
 		return 0;
 
 	fs_info->qgroup_ulist = ulist_alloc(GFP_NOFS);
@@ -360,8 +360,7 @@ int btrfs_read_qgroup_config(struct btrfs_fs_info *fs_info)
 			    fs_info->generation) {
 				flags |= BTRFS_QGROUP_STATUS_FLAG_INCONSISTENT;
 				btrfs_err(fs_info,
-					"qgroup generation mismatch, "
-					"marked as inconsistent");
+					"qgroup generation mismatch, marked as inconsistent");
 			}
 			fs_info->qgroup_flags = btrfs_qgroup_status_flags(l,
 									  ptr);
@@ -463,13 +462,11 @@ next2:
 	}
 out:
 	fs_info->qgroup_flags |= flags;
-	if (!(fs_info->qgroup_flags & BTRFS_QGROUP_STATUS_FLAG_ON)) {
-		fs_info->quota_enabled = 0;
-		fs_info->pending_quota_state = 0;
-	} else if (fs_info->qgroup_flags & BTRFS_QGROUP_STATUS_FLAG_RESCAN &&
-		   ret >= 0) {
+	if (!(fs_info->qgroup_flags & BTRFS_QGROUP_STATUS_FLAG_ON))
+		clear_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags);
+	else if (fs_info->qgroup_flags & BTRFS_QGROUP_STATUS_FLAG_RESCAN &&
+		 ret >= 0)
 		ret = qgroup_rescan_init(fs_info, rescan_progress, 0);
-	}
 	btrfs_free_path(path);
 
 	if (ret < 0) {
@@ -571,7 +568,7 @@ static int add_qgroup_item(struct btrfs_trans_handle *trans,
 	struct extent_buffer *leaf;
 	struct btrfs_key key;
 
-	if (btrfs_test_is_dummy_root(quota_root))
+	if (btrfs_is_testing(quota_root->fs_info))
 		return 0;
 
 	path = btrfs_alloc_path();
@@ -728,7 +725,7 @@ static int update_qgroup_info_item(struct btrfs_trans_handle *trans,
 	int ret;
 	int slot;
 
-	if (btrfs_test_is_dummy_root(root))
+	if (btrfs_is_testing(root->fs_info))
 		return 0;
 
 	key.objectid = 0;
@@ -847,7 +844,7 @@ static int btrfs_clean_quota_tree(struct btrfs_trans_handle *trans,
 	}
 	ret = 0;
 out:
-	root->fs_info->pending_quota_state = 0;
+	set_bit(BTRFS_FS_QUOTA_DISABLING, &root->fs_info->flags);
 	btrfs_free_path(path);
 	return ret;
 }
@@ -868,7 +865,7 @@ int btrfs_quota_enable(struct btrfs_trans_handle *trans,
 
 	mutex_lock(&fs_info->qgroup_ioctl_lock);
 	if (fs_info->quota_root) {
-		fs_info->pending_quota_state = 1;
+		set_bit(BTRFS_FS_QUOTA_ENABLING, &fs_info->flags);
 		goto out;
 	}
 
@@ -964,7 +961,7 @@ out_add_root:
 	}
 	spin_lock(&fs_info->qgroup_lock);
 	fs_info->quota_root = quota_root;
-	fs_info->pending_quota_state = 1;
+	set_bit(BTRFS_FS_QUOTA_ENABLING, &fs_info->flags);
 	spin_unlock(&fs_info->qgroup_lock);
 out_free_path:
 	btrfs_free_path(path);
@@ -993,9 +990,9 @@ int btrfs_quota_disable(struct btrfs_trans_handle *trans,
 	mutex_lock(&fs_info->qgroup_ioctl_lock);
 	if (!fs_info->quota_root)
 		goto out;
-	fs_info->quota_enabled = 0;
-	fs_info->pending_quota_state = 0;
-	btrfs_qgroup_wait_for_completion(fs_info);
+	clear_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags);
+	set_bit(BTRFS_FS_QUOTA_DISABLING, &fs_info->flags);
+	btrfs_qgroup_wait_for_completion(fs_info, false);
 	spin_lock(&fs_info->qgroup_lock);
 	quota_root = fs_info->quota_root;
 	fs_info->quota_root = NULL;
@@ -1453,9 +1450,9 @@ int btrfs_qgroup_prepare_account_extents(struct btrfs_trans_handle *trans,
 	return ret;
 }
 
-struct btrfs_qgroup_extent_record
-*btrfs_qgroup_insert_dirty_extent(struct btrfs_delayed_ref_root *delayed_refs,
-				  struct btrfs_qgroup_extent_record *record)
+int btrfs_qgroup_insert_dirty_extent_nolock(struct btrfs_fs_info *fs_info,
+				struct btrfs_delayed_ref_root *delayed_refs,
+				struct btrfs_qgroup_extent_record *record)
 {
 	struct rb_node **p = &delayed_refs->dirty_extent_root.rb_node;
 	struct rb_node *parent_node = NULL;
@@ -1463,7 +1460,7 @@ struct btrfs_qgroup_extent_record
 	u64 bytenr = record->bytenr;
 
 	assert_spin_locked(&delayed_refs->lock);
-	trace_btrfs_qgroup_insert_dirty_extent(record);
+	trace_btrfs_qgroup_insert_dirty_extent(fs_info, record);
 
 	while (*p) {
 		parent_node = *p;
@@ -1474,12 +1471,43 @@ struct btrfs_qgroup_extent_record
 		else if (bytenr > entry->bytenr)
 			p = &(*p)->rb_right;
 		else
-			return entry;
+			return 1;
 	}
 
 	rb_link_node(&record->node, parent_node, p);
 	rb_insert_color(&record->node, &delayed_refs->dirty_extent_root);
-	return NULL;
+	return 0;
+}
+
+int btrfs_qgroup_insert_dirty_extent(struct btrfs_trans_handle *trans,
+		struct btrfs_fs_info *fs_info, u64 bytenr, u64 num_bytes,
+		gfp_t gfp_flag)
+{
+	struct btrfs_qgroup_extent_record *record;
+	struct btrfs_delayed_ref_root *delayed_refs;
+	int ret;
+
+	if (!test_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags)
+	    || bytenr == 0 || num_bytes == 0)
+		return 0;
+	if (WARN_ON(trans == NULL))
+		return -EINVAL;
+	record = kmalloc(sizeof(*record), gfp_flag);
+	if (!record)
+		return -ENOMEM;
+
+	delayed_refs = &trans->transaction->delayed_refs;
+	record->bytenr = bytenr;
+	record->num_bytes = num_bytes;
+	record->old_roots = NULL;
+
+	spin_lock(&delayed_refs->lock);
+	ret = btrfs_qgroup_insert_dirty_extent_nolock(fs_info, delayed_refs,
+						      record);
+	spin_unlock(&delayed_refs->lock);
+	if (ret > 0)
+		kfree(record);
+	return 0;
 }
 
 #define UPDATE_NEW	0
@@ -1595,8 +1623,8 @@ static int qgroup_update_counters(struct btrfs_fs_info *fs_info,
 		cur_old_count = btrfs_qgroup_get_old_refcnt(qg, seq);
 		cur_new_count = btrfs_qgroup_get_new_refcnt(qg, seq);
 
-		trace_qgroup_update_counters(qg->qgroupid, cur_old_count,
-					     cur_new_count);
+		trace_qgroup_update_counters(fs_info, qg->qgroupid,
+					     cur_old_count, cur_new_count);
 
 		/* Rfer update part */
 		if (cur_old_count == 0 && cur_new_count > 0) {
@@ -1683,12 +1711,12 @@ btrfs_qgroup_account_extent(struct btrfs_trans_handle *trans,
 	if (old_roots)
 		nr_old_roots = old_roots->nnodes;
 
-	if (!fs_info->quota_enabled)
+	if (!test_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags))
 		goto out_free;
 	BUG_ON(!fs_info->quota_root);
 
-	trace_btrfs_qgroup_account_extent(bytenr, num_bytes, nr_old_roots,
-					  nr_new_roots);
+	trace_btrfs_qgroup_account_extent(fs_info, bytenr, num_bytes,
+					  nr_old_roots, nr_new_roots);
 
 	qgroups = ulist_alloc(GFP_NOFS);
 	if (!qgroups) {
@@ -1759,7 +1787,7 @@ int btrfs_qgroup_account_extents(struct btrfs_trans_handle *trans,
 		record = rb_entry(node, struct btrfs_qgroup_extent_record,
 				  node);
 
-		trace_btrfs_qgroup_account_extents(record);
+		trace_btrfs_qgroup_account_extents(fs_info, record);
 
 		if (!ret) {
 			/*
@@ -1803,10 +1831,14 @@ int btrfs_run_qgroups(struct btrfs_trans_handle *trans,
 	if (!quota_root)
 		goto out;
 
-	if (!fs_info->quota_enabled && fs_info->pending_quota_state)
+	if (!test_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags) &&
+	    test_bit(BTRFS_FS_QUOTA_ENABLING, &fs_info->flags))
 		start_rescan_worker = 1;
 
-	fs_info->quota_enabled = fs_info->pending_quota_state;
+	if (test_and_clear_bit(BTRFS_FS_QUOTA_ENABLING, &fs_info->flags))
+		set_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags);
+	if (test_and_clear_bit(BTRFS_FS_QUOTA_DISABLING, &fs_info->flags))
+		clear_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags);
 
 	spin_lock(&fs_info->qgroup_lock);
 	while (!list_empty(&fs_info->dirty_qgroups)) {
@@ -1825,7 +1857,7 @@ int btrfs_run_qgroups(struct btrfs_trans_handle *trans,
 					BTRFS_QGROUP_STATUS_FLAG_INCONSISTENT;
 		spin_lock(&fs_info->qgroup_lock);
 	}
-	if (fs_info->quota_enabled)
+	if (test_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags))
 		fs_info->qgroup_flags |= BTRFS_QGROUP_STATUS_FLAG_ON;
 	else
 		fs_info->qgroup_flags &= ~BTRFS_QGROUP_STATUS_FLAG_ON;
@@ -1870,7 +1902,7 @@ int btrfs_qgroup_inherit(struct btrfs_trans_handle *trans,
 	u64 nums;
 
 	mutex_lock(&fs_info->qgroup_ioctl_lock);
-	if (!fs_info->quota_enabled)
+	if (!test_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags))
 		goto out;
 
 	if (!quota_root) {
@@ -1961,8 +1993,9 @@ int btrfs_qgroup_inherit(struct btrfs_trans_handle *trans,
 		ret = update_qgroup_limit_item(trans, quota_root, dstgroup);
 		if (ret) {
 			fs_info->qgroup_flags |= BTRFS_QGROUP_STATUS_FLAG_INCONSISTENT;
-			btrfs_info(fs_info, "unable to update quota limit for %llu",
-			       dstgroup->qgroupid);
+			btrfs_info(fs_info,
+				   "unable to update quota limit for %llu",
+				   dstgroup->qgroupid);
 			goto unlock;
 		}
 	}
@@ -2195,9 +2228,8 @@ void assert_qgroups_uptodate(struct btrfs_trans_handle *trans)
 {
 	if (list_empty(&trans->qgroup_ref_list) && !trans->delayed_ref_elem.seq)
 		return;
-	btrfs_err(trans->root->fs_info,
-		"qgroups not uptodate in trans handle %p:  list is%s empty, "
-		"seq is %#x.%x",
+	btrfs_err(trans->fs_info,
+		"qgroups not uptodate in trans handle %p:  list is%s empty, seq is %#x.%x",
 		trans, list_empty(&trans->qgroup_ref_list) ? "" : " not",
 		(u32)(trans->delayed_ref_elem.seq >> 32),
 		(u32)trans->delayed_ref_elem.seq);
@@ -2225,10 +2257,11 @@ qgroup_rescan_leaf(struct btrfs_fs_info *fs_info, struct btrfs_path *path,
 					 &fs_info->qgroup_rescan_progress,
 					 path, 1, 0);
 
-	pr_debug("current progress key (%llu %u %llu), search_slot ret %d\n",
-		 fs_info->qgroup_rescan_progress.objectid,
-		 fs_info->qgroup_rescan_progress.type,
-		 fs_info->qgroup_rescan_progress.offset, ret);
+	btrfs_debug(fs_info,
+		"current progress key (%llu %u %llu), search_slot ret %d",
+		fs_info->qgroup_rescan_progress.objectid,
+		fs_info->qgroup_rescan_progress.type,
+		fs_info->qgroup_rescan_progress.offset, ret);
 
 	if (ret) {
 		/*
@@ -2302,6 +2335,10 @@ static void btrfs_qgroup_rescan_worker(struct btrfs_work *work)
 	int err = -ENOMEM;
 	int ret = 0;
 
+	mutex_lock(&fs_info->qgroup_rescan_lock);
+	fs_info->qgroup_rescan_running = true;
+	mutex_unlock(&fs_info->qgroup_rescan_lock);
+
 	path = btrfs_alloc_path();
 	if (!path)
 		goto out;
@@ -2313,7 +2350,7 @@ static void btrfs_qgroup_rescan_worker(struct btrfs_work *work)
 			err = PTR_ERR(trans);
 			break;
 		}
-		if (!fs_info->quota_enabled) {
+		if (!test_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags)) {
 			err = -EINTR;
 		} else {
 			err = qgroup_rescan_leaf(fs_info, path, trans);
@@ -2354,7 +2391,7 @@ out:
 	ret = update_qgroup_status_item(trans, fs_info, fs_info->quota_root);
 	if (ret < 0) {
 		err = ret;
-		btrfs_err(fs_info, "fail to update qgroup status: %d\n", err);
+		btrfs_err(fs_info, "fail to update qgroup status: %d", err);
 	}
 	btrfs_end_transaction(trans, fs_info->quota_root);
 
@@ -2368,6 +2405,9 @@ out:
 	}
 
 done:
+	mutex_lock(&fs_info->qgroup_rescan_lock);
+	fs_info->qgroup_rescan_running = false;
+	mutex_unlock(&fs_info->qgroup_rescan_lock);
 	complete_all(&fs_info->qgroup_rescan_completion);
 }
 
@@ -2486,20 +2526,26 @@ btrfs_qgroup_rescan(struct btrfs_fs_info *fs_info)
 	return 0;
 }
 
-int btrfs_qgroup_wait_for_completion(struct btrfs_fs_info *fs_info)
+int btrfs_qgroup_wait_for_completion(struct btrfs_fs_info *fs_info,
+				     bool interruptible)
 {
 	int running;
 	int ret = 0;
 
 	mutex_lock(&fs_info->qgroup_rescan_lock);
 	spin_lock(&fs_info->qgroup_lock);
-	running = fs_info->qgroup_flags & BTRFS_QGROUP_STATUS_FLAG_RESCAN;
+	running = fs_info->qgroup_rescan_running;
 	spin_unlock(&fs_info->qgroup_lock);
 	mutex_unlock(&fs_info->qgroup_rescan_lock);
 
-	if (running)
+	if (!running)
+		return 0;
+
+	if (interruptible)
 		ret = wait_for_completion_interruptible(
 					&fs_info->qgroup_rescan_completion);
+	else
+		wait_for_completion(&fs_info->qgroup_rescan_completion);
 
 	return ret;
 }
@@ -2535,8 +2581,8 @@ int btrfs_qgroup_reserve_data(struct inode *inode, u64 start, u64 len)
 	struct ulist_iterator uiter;
 	int ret;
 
-	if (!root->fs_info->quota_enabled || !is_fstree(root->objectid) ||
-	    len == 0)
+	if (!test_bit(BTRFS_FS_QUOTA_ENABLED, &root->fs_info->flags) ||
+	    !is_fstree(root->objectid) || len == 0)
 		return 0;
 
 	changeset.bytes_changed = 0;
@@ -2633,8 +2679,8 @@ int btrfs_qgroup_reserve_meta(struct btrfs_root *root, int num_bytes)
 {
 	int ret;
 
-	if (!root->fs_info->quota_enabled || !is_fstree(root->objectid) ||
-	    num_bytes == 0)
+	if (!test_bit(BTRFS_FS_QUOTA_ENABLED, &root->fs_info->flags) ||
+	    !is_fstree(root->objectid) || num_bytes == 0)
 		return 0;
 
 	BUG_ON(num_bytes != round_down(num_bytes, root->nodesize));
@@ -2649,7 +2695,8 @@ void btrfs_qgroup_free_meta_all(struct btrfs_root *root)
 {
 	int reserved;
 
-	if (!root->fs_info->quota_enabled || !is_fstree(root->objectid))
+	if (!test_bit(BTRFS_FS_QUOTA_ENABLED, &root->fs_info->flags) ||
+	    !is_fstree(root->objectid))
 		return;
 
 	reserved = atomic_xchg(&root->qgroup_meta_rsv, 0);
@@ -2660,7 +2707,8 @@ void btrfs_qgroup_free_meta_all(struct btrfs_root *root)
 
 void btrfs_qgroup_free_meta(struct btrfs_root *root, int num_bytes)
 {
-	if (!root->fs_info->quota_enabled || !is_fstree(root->objectid))
+	if (!test_bit(BTRFS_FS_QUOTA_ENABLED, &root->fs_info->flags) ||
+	    !is_fstree(root->objectid))
 		return;
 
 	BUG_ON(num_bytes != round_down(num_bytes, root->nodesize));
diff --git a/fs/btrfs/qgroup.h b/fs/btrfs/qgroup.h
index ecb2c143ef75..1bc64c864b62 100644
--- a/fs/btrfs/qgroup.h
+++ b/fs/btrfs/qgroup.h
@@ -46,7 +46,8 @@ int btrfs_quota_disable(struct btrfs_trans_handle *trans,
 			struct btrfs_fs_info *fs_info);
 int btrfs_qgroup_rescan(struct btrfs_fs_info *fs_info);
 void btrfs_qgroup_rescan_resume(struct btrfs_fs_info *fs_info);
-int btrfs_qgroup_wait_for_completion(struct btrfs_fs_info *fs_info);
+int btrfs_qgroup_wait_for_completion(struct btrfs_fs_info *fs_info,
+				     bool interruptible);
 int btrfs_add_qgroup_relation(struct btrfs_trans_handle *trans,
 			      struct btrfs_fs_info *fs_info, u64 src, u64 dst);
 int btrfs_del_qgroup_relation(struct btrfs_trans_handle *trans,
@@ -63,9 +64,35 @@ void btrfs_free_qgroup_config(struct btrfs_fs_info *fs_info);
 struct btrfs_delayed_extent_op;
 int btrfs_qgroup_prepare_account_extents(struct btrfs_trans_handle *trans,
 					 struct btrfs_fs_info *fs_info);
-struct btrfs_qgroup_extent_record
-*btrfs_qgroup_insert_dirty_extent(struct btrfs_delayed_ref_root *delayed_refs,
-				  struct btrfs_qgroup_extent_record *record);
+/*
+ * Insert one dirty extent record into @delayed_refs, informing qgroup to
+ * account that extent at commit trans time.
+ *
+ * No lock version, caller must acquire delayed ref lock and allocate memory.
+ *
+ * Return 0 for success insert
+ * Return >0 for existing record, caller can free @record safely.
+ * Error is not possible
+ */
+int btrfs_qgroup_insert_dirty_extent_nolock(
+		struct btrfs_fs_info *fs_info,
+		struct btrfs_delayed_ref_root *delayed_refs,
+		struct btrfs_qgroup_extent_record *record);
+
+/*
+ * Insert one dirty extent record into @delayed_refs, informing qgroup to
+ * account that extent at commit trans time.
+ *
+ * Better encapsulated version.
+ *
+ * Return 0 if the operation is done.
+ * Return <0 for error, like memory allocation failure or invalid parameter
+ * (NULL trans)
+ */
+int btrfs_qgroup_insert_dirty_extent(struct btrfs_trans_handle *trans,
+		struct btrfs_fs_info *fs_info, u64 bytenr, u64 num_bytes,
+		gfp_t gfp_flag);
+
 int
 btrfs_qgroup_account_extent(struct btrfs_trans_handle *trans,
 			    struct btrfs_fs_info *fs_info,
@@ -88,7 +115,7 @@ static inline void btrfs_qgroup_free_delayed_ref(struct btrfs_fs_info *fs_info,
 						 u64 ref_root, u64 num_bytes)
 {
 	btrfs_qgroup_free_refroot(fs_info, ref_root, num_bytes);
-	trace_btrfs_qgroup_free_delayed_ref(ref_root, num_bytes);
+	trace_btrfs_qgroup_free_delayed_ref(fs_info, ref_root, num_bytes);
 }
 void assert_qgroups_uptodate(struct btrfs_trans_handle *trans);
 
diff --git a/fs/btrfs/raid56.c b/fs/btrfs/raid56.c
index f8b6d411a034..d016d4a79864 100644
--- a/fs/btrfs/raid56.c
+++ b/fs/btrfs/raid56.c
@@ -1320,7 +1320,9 @@ write_data:
 
 		bio->bi_private = rbio;
 		bio->bi_end_io = raid_write_end_io;
-		submit_bio(WRITE, bio);
+		bio_set_op_attrs(bio, REQ_OP_WRITE, 0);
+
+		submit_bio(bio);
 	}
 	return;
 
@@ -1573,11 +1575,12 @@ static int raid56_rmw_stripe(struct btrfs_raid_bio *rbio)
 
 		bio->bi_private = rbio;
 		bio->bi_end_io = raid_rmw_end_io;
+		bio_set_op_attrs(bio, REQ_OP_READ, 0);
 
 		btrfs_bio_wq_end_io(rbio->fs_info, bio,
 				    BTRFS_WQ_ENDIO_RAID56);
 
-		submit_bio(READ, bio);
+		submit_bio(bio);
 	}
 	/* the actual write will happen once the reads are done */
 	return 0;
@@ -2097,11 +2100,12 @@ static int __raid56_parity_recover(struct btrfs_raid_bio *rbio)
 
 		bio->bi_private = rbio;
 		bio->bi_end_io = raid_recover_end_io;
+		bio_set_op_attrs(bio, REQ_OP_READ, 0);
 
 		btrfs_bio_wq_end_io(rbio->fs_info, bio,
 				    BTRFS_WQ_ENDIO_RAID56);
 
-		submit_bio(READ, bio);
+		submit_bio(bio);
 	}
 out:
 	return 0;
@@ -2139,7 +2143,10 @@ int raid56_parity_recover(struct btrfs_root *root, struct bio *bio,
 
 	rbio->faila = find_logical_bio_stripe(rbio, bio);
 	if (rbio->faila == -1) {
-		BUG();
+		btrfs_warn(root->fs_info,
+	"%s could not find the bad stripe in raid56 so that we cannot recover any more (bio has logical %llu len %llu, bbio has map_type %llu)",
+			   __func__, (u64)bio->bi_iter.bi_sector << 9,
+			   (u64)bio->bi_iter.bi_size, bbio->map_type);
 		if (generic_io)
 			btrfs_put_bbio(bbio);
 		kfree(rbio);
@@ -2433,7 +2440,9 @@ submit_write:
 
 		bio->bi_private = rbio;
 		bio->bi_end_io = raid_write_end_io;
-		submit_bio(WRITE, bio);
+		bio_set_op_attrs(bio, REQ_OP_WRITE, 0);
+
+		submit_bio(bio);
 	}
 	return;
 
@@ -2610,11 +2619,12 @@ static void raid56_parity_scrub_stripe(struct btrfs_raid_bio *rbio)
 
 		bio->bi_private = rbio;
 		bio->bi_end_io = raid56_parity_scrub_end_io;
+		bio_set_op_attrs(bio, REQ_OP_READ, 0);
 
 		btrfs_bio_wq_end_io(rbio->fs_info, bio,
 				    BTRFS_WQ_ENDIO_RAID56);
 
-		submit_bio(READ, bio);
+		submit_bio(bio);
 	}
 	/* the actual write will happen once the reads are done */
 	return;
diff --git a/fs/btrfs/reada.c b/fs/btrfs/reada.c
index 298631eaee78..75bab76739be 100644
--- a/fs/btrfs/reada.c
+++ b/fs/btrfs/reada.c
@@ -761,12 +761,14 @@ static void __reada_start_machine(struct btrfs_fs_info *fs_info)
 
 	do {
 		enqueued = 0;
+		mutex_lock(&fs_devices->device_list_mutex);
 		list_for_each_entry(device, &fs_devices->devices, dev_list) {
 			if (atomic_read(&device->reada_in_flight) <
 			    MAX_IN_FLIGHT)
 				enqueued += reada_start_machine_dev(fs_info,
 								    device);
 		}
+		mutex_unlock(&fs_devices->device_list_mutex);
 		total += enqueued;
 	} while (enqueued && total < 10000);
 
@@ -818,7 +820,7 @@ static void dump_devs(struct btrfs_fs_info *fs_info, int all)
 
 	spin_lock(&fs_info->reada_lock);
 	list_for_each_entry(device, &fs_devices->devices, dev_list) {
-		printk(KERN_DEBUG "dev %lld has %d in flight\n", device->devid,
+		btrfs_debug(fs_info, "dev %lld has %d in flight", device->devid,
 			atomic_read(&device->reada_in_flight));
 		index = 0;
 		while (1) {
@@ -827,17 +829,17 @@ static void dump_devs(struct btrfs_fs_info *fs_info, int all)
 						     (void **)&zone, index, 1);
 			if (ret == 0)
 				break;
-			printk(KERN_DEBUG "  zone %llu-%llu elems %llu locked "
-				"%d devs", zone->start, zone->end, zone->elems,
-				zone->locked);
+			pr_debug("  zone %llu-%llu elems %llu locked %d devs",
+				    zone->start, zone->end, zone->elems,
+				    zone->locked);
 			for (j = 0; j < zone->ndevs; ++j) {
-				printk(KERN_CONT " %lld",
+				pr_cont(" %lld",
 					zone->devs[j]->devid);
 			}
 			if (device->reada_curr_zone == zone)
-				printk(KERN_CONT " curr off %llu",
+				pr_cont(" curr off %llu",
 					device->reada_next - zone->start);
-			printk(KERN_CONT "\n");
+			pr_cont("\n");
 			index = (zone->end >> PAGE_SHIFT) + 1;
 		}
 		cnt = 0;
@@ -849,21 +851,20 @@ static void dump_devs(struct btrfs_fs_info *fs_info, int all)
 						     (void **)&re, index, 1);
 			if (ret == 0)
 				break;
-			printk(KERN_DEBUG
-				"  re: logical %llu size %u empty %d scheduled %d",
+			pr_debug("  re: logical %llu size %u empty %d scheduled %d",
 				re->logical, fs_info->tree_root->nodesize,
 				list_empty(&re->extctl), re->scheduled);
 
 			for (i = 0; i < re->nzones; ++i) {
-				printk(KERN_CONT " zone %llu-%llu devs",
+				pr_cont(" zone %llu-%llu devs",
 					re->zones[i]->start,
 					re->zones[i]->end);
 				for (j = 0; j < re->zones[i]->ndevs; ++j) {
-					printk(KERN_CONT " %lld",
+					pr_cont(" %lld",
 						re->zones[i]->devs[j]->devid);
 				}
 			}
-			printk(KERN_CONT "\n");
+			pr_cont("\n");
 			index = (re->logical >> PAGE_SHIFT) + 1;
 			if (++cnt > 15)
 				break;
@@ -883,20 +884,19 @@ static void dump_devs(struct btrfs_fs_info *fs_info, int all)
 			index = (re->logical >> PAGE_SHIFT) + 1;
 			continue;
 		}
-		printk(KERN_DEBUG
-			"re: logical %llu size %u list empty %d scheduled %d",
+		pr_debug("re: logical %llu size %u list empty %d scheduled %d",
 			re->logical, fs_info->tree_root->nodesize,
 			list_empty(&re->extctl), re->scheduled);
 		for (i = 0; i < re->nzones; ++i) {
-			printk(KERN_CONT " zone %llu-%llu devs",
+			pr_cont(" zone %llu-%llu devs",
 				re->zones[i]->start,
 				re->zones[i]->end);
 			for (j = 0; j < re->zones[i]->ndevs; ++j) {
-				printk(KERN_CONT " %lld",
+				pr_cont(" %lld",
 				       re->zones[i]->devs[j]->devid);
 			}
 		}
-		printk(KERN_CONT "\n");
+		pr_cont("\n");
 		index = (re->logical >> PAGE_SHIFT) + 1;
 	}
 	spin_unlock(&fs_info->reada_lock);
diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c
index 0477dca154ed..0ec8ffa37ab0 100644
--- a/fs/btrfs/relocation.c
+++ b/fs/btrfs/relocation.c
@@ -31,6 +31,7 @@
 #include "async-thread.h"
 #include "free-space-cache.h"
 #include "inode-map.h"
+#include "qgroup.h"
 
 /*
  * backref_node, mapping_node and tree_block start with this
@@ -235,12 +236,12 @@ static void backref_cache_cleanup(struct backref_cache *cache)
 	cache->last_trans = 0;
 
 	for (i = 0; i < BTRFS_MAX_LEVEL; i++)
-		BUG_ON(!list_empty(&cache->pending[i]));
-	BUG_ON(!list_empty(&cache->changed));
-	BUG_ON(!list_empty(&cache->detached));
-	BUG_ON(!RB_EMPTY_ROOT(&cache->rb_root));
-	BUG_ON(cache->nr_nodes);
-	BUG_ON(cache->nr_edges);
+		ASSERT(list_empty(&cache->pending[i]));
+	ASSERT(list_empty(&cache->changed));
+	ASSERT(list_empty(&cache->detached));
+	ASSERT(RB_EMPTY_ROOT(&cache->rb_root));
+	ASSERT(!cache->nr_nodes);
+	ASSERT(!cache->nr_edges);
 }
 
 static struct backref_node *alloc_backref_node(struct backref_cache *cache)
@@ -336,8 +337,9 @@ static void backref_tree_panic(struct rb_node *rb_node, int errno, u64 bytenr)
 					      rb_node);
 	if (bnode->root)
 		fs_info = bnode->root->fs_info;
-	btrfs_panic(fs_info, errno, "Inconsistency in backref cache "
-		    "found at offset %llu", bytenr);
+	btrfs_panic(fs_info, errno,
+		    "Inconsistency in backref cache found at offset %llu",
+		    bytenr);
 }
 
 /*
@@ -922,9 +924,16 @@ again:
 			path2->slots[level]--;
 
 		eb = path2->nodes[level];
-		WARN_ON(btrfs_node_blockptr(eb, path2->slots[level]) !=
-			cur->bytenr);
-
+		if (btrfs_node_blockptr(eb, path2->slots[level]) !=
+		    cur->bytenr) {
+			btrfs_err(root->fs_info,
+	"couldn't find block (%llu) (level %d) in tree (%llu) with key (%llu %u %llu)",
+				  cur->bytenr, level - 1, root->objectid,
+				  node_key->objectid, node_key->type,
+				  node_key->offset);
+			err = -ENOENT;
+			goto out;
+		}
 		lower = cur;
 		need_check = true;
 		for (; level < BTRFS_MAX_LEVEL; level++) {
@@ -1171,8 +1180,12 @@ out:
 			lower = list_entry(useless.next,
 					   struct backref_node, list);
 			list_del_init(&lower->list);
+			if (lower == node)
+				node = NULL;
 			free_backref_node(cache, lower);
 		}
+
+		free_backref_node(cache, node);
 		return ERR_PTR(err);
 	}
 	ASSERT(!node || !node->detached);
@@ -1291,9 +1304,9 @@ static int __must_check __add_reloc_root(struct btrfs_root *root)
 			      node->bytenr, &node->rb_node);
 	spin_unlock(&rc->reloc_root_tree.lock);
 	if (rb_node) {
-		btrfs_panic(root->fs_info, -EEXIST, "Duplicate root found "
-			    "for start=%llu while inserting into relocation "
-			    "tree", node->bytenr);
+		btrfs_panic(root->fs_info, -EEXIST,
+			    "Duplicate root found for start=%llu while inserting into relocation tree",
+			    node->bytenr);
 		kfree(node);
 		return -EEXIST;
 	}
@@ -1719,7 +1732,7 @@ int replace_file_extents(struct btrfs_trans_handle *trans,
 					   btrfs_header_owner(leaf),
 					   key.objectid, key.offset);
 		if (ret) {
-			btrfs_abort_transaction(trans, root, ret);
+			btrfs_abort_transaction(trans, ret);
 			break;
 		}
 
@@ -1727,7 +1740,7 @@ int replace_file_extents(struct btrfs_trans_handle *trans,
 					parent, btrfs_header_owner(leaf),
 					key.objectid, key.offset);
 		if (ret) {
-			btrfs_abort_transaction(trans, root, ret);
+			btrfs_abort_transaction(trans, ret);
 			break;
 		}
 	}
@@ -2345,6 +2358,10 @@ void free_reloc_roots(struct list_head *list)
 	while (!list_empty(list)) {
 		reloc_root = list_entry(list->next, struct btrfs_root,
 					root_list);
+		free_extent_buffer(reloc_root->node);
+		free_extent_buffer(reloc_root->commit_root);
+		reloc_root->node = NULL;
+		reloc_root->commit_root = NULL;
 		__del_reloc_root(reloc_root);
 	}
 }
@@ -2604,25 +2621,28 @@ static int reserve_metadata_space(struct btrfs_trans_handle *trans,
 
 	trans->block_rsv = rc->block_rsv;
 	rc->reserved_bytes += num_bytes;
+
+	/*
+	 * We are under a transaction here so we can only do limited flushing.
+	 * If we get an enospc just kick back -EAGAIN so we know to drop the
+	 * transaction and try to refill when we can flush all the things.
+	 */
 	ret = btrfs_block_rsv_refill(root, rc->block_rsv, num_bytes,
-				BTRFS_RESERVE_FLUSH_ALL);
+				BTRFS_RESERVE_FLUSH_LIMIT);
 	if (ret) {
-		if (ret == -EAGAIN) {
-			tmp = rc->extent_root->nodesize *
-				RELOCATION_RESERVED_NODES;
-			while (tmp <= rc->reserved_bytes)
-				tmp <<= 1;
-			/*
-			 * only one thread can access block_rsv at this point,
-			 * so we don't need hold lock to protect block_rsv.
-			 * we expand more reservation size here to allow enough
-			 * space for relocation and we will return earlier in
-			 * enospc case.
-			 */
-			rc->block_rsv->size = tmp + rc->extent_root->nodesize *
-					      RELOCATION_RESERVED_NODES;
-		}
-		return ret;
+		tmp = rc->extent_root->nodesize * RELOCATION_RESERVED_NODES;
+		while (tmp <= rc->reserved_bytes)
+			tmp <<= 1;
+		/*
+		 * only one thread can access block_rsv at this point,
+		 * so we don't need hold lock to protect block_rsv.
+		 * we expand more reservation size here to allow enough
+		 * space for relocation and we will return eailer in
+		 * enospc case.
+		 */
+		rc->block_rsv->size = tmp + rc->extent_root->nodesize *
+			RELOCATION_RESERVED_NODES;
+		return -EAGAIN;
 	}
 
 	return 0;
@@ -2678,11 +2698,15 @@ static int do_relocation(struct btrfs_trans_handle *trans,
 
 		if (!upper->eb) {
 			ret = btrfs_search_slot(trans, root, key, path, 0, 1);
-			if (ret < 0) {
-				err = ret;
+			if (ret) {
+				if (ret < 0)
+					err = ret;
+				else
+					err = -ENOENT;
+
+				btrfs_release_path(path);
 				break;
 			}
-			BUG_ON(ret > 0);
 
 			if (!upper->eb) {
 				upper->eb = path->nodes[upper->level];
@@ -3030,15 +3054,19 @@ int prealloc_file_extent_cluster(struct inode *inode,
 	u64 num_bytes;
 	int nr = 0;
 	int ret = 0;
+	u64 prealloc_start = cluster->start - offset;
+	u64 prealloc_end = cluster->end - offset;
+	u64 cur_offset;
 
 	BUG_ON(cluster->start != cluster->boundary[0]);
 	inode_lock(inode);
 
-	ret = btrfs_check_data_free_space(inode, cluster->start,
-					  cluster->end + 1 - cluster->start);
+	ret = btrfs_check_data_free_space(inode, prealloc_start,
+					  prealloc_end + 1 - prealloc_start);
 	if (ret)
 		goto out;
 
+	cur_offset = prealloc_start;
 	while (nr < cluster->nr) {
 		start = cluster->boundary[nr] - offset;
 		if (nr + 1 < cluster->nr)
@@ -3048,16 +3076,21 @@ int prealloc_file_extent_cluster(struct inode *inode,
 
 		lock_extent(&BTRFS_I(inode)->io_tree, start, end);
 		num_bytes = end + 1 - start;
+		if (cur_offset < start)
+			btrfs_free_reserved_data_space(inode, cur_offset,
+					start - cur_offset);
 		ret = btrfs_prealloc_file_range(inode, 0, start,
 						num_bytes, num_bytes,
 						end + 1, &alloc_hint);
+		cur_offset = end + 1;
 		unlock_extent(&BTRFS_I(inode)->io_tree, start, end);
 		if (ret)
 			break;
 		nr++;
 	}
-	btrfs_free_reserved_data_space(inode, cluster->start,
-				       cluster->end + 1 - cluster->start);
+	if (cur_offset < prealloc_end)
+		btrfs_free_reserved_data_space(inode, cur_offset,
+				       prealloc_end + 1 - cur_offset);
 out:
 	inode_unlock(inode);
 	return ret;
@@ -3186,7 +3219,7 @@ static int relocate_file_extent_cluster(struct inode *inode,
 			nr++;
 		}
 
-		btrfs_set_extent_delalloc(inode, page_start, page_end, NULL);
+		btrfs_set_extent_delalloc(inode, page_start, page_end, NULL, 0);
 		set_page_dirty(page);
 
 		unlock_extent(&BTRFS_I(inode)->io_tree,
@@ -3871,6 +3904,7 @@ static noinline_for_stack
 int prepare_to_relocate(struct reloc_control *rc)
 {
 	struct btrfs_trans_handle *trans;
+	int ret;
 
 	rc->block_rsv = btrfs_alloc_block_rsv(rc->extent_root,
 					      BTRFS_BLOCK_RSV_TEMP);
@@ -3885,6 +3919,11 @@ int prepare_to_relocate(struct reloc_control *rc)
 	rc->reserved_bytes = 0;
 	rc->block_rsv->size = rc->extent_root->nodesize *
 			      RELOCATION_RESERVED_NODES;
+	ret = btrfs_block_rsv_refill(rc->extent_root,
+				     rc->block_rsv, rc->block_rsv->size,
+				     BTRFS_RESERVE_FLUSH_ALL);
+	if (ret)
+		return ret;
 
 	rc->create_reloc_tree = 1;
 	set_reloc_control(rc);
@@ -3903,6 +3942,90 @@ int prepare_to_relocate(struct reloc_control *rc)
 	return 0;
 }
 
+/*
+ * Qgroup fixer for data chunk relocation.
+ * The data relocation is done in the following steps
+ * 1) Copy data extents into data reloc tree
+ * 2) Create tree reloc tree(special snapshot) for related subvolumes
+ * 3) Modify file extents in tree reloc tree
+ * 4) Merge tree reloc tree with original fs tree, by swapping tree blocks
+ *
+ * The problem is, data and tree reloc tree are not accounted to qgroup,
+ * and 4) will only info qgroup to track tree blocks change, not file extents
+ * in the tree blocks.
+ *
+ * The good news is, related data extents are all in data reloc tree, so we
+ * only need to info qgroup to track all file extents in data reloc tree
+ * before commit trans.
+ */
+static int qgroup_fix_relocated_data_extents(struct btrfs_trans_handle *trans,
+					     struct reloc_control *rc)
+{
+	struct btrfs_fs_info *fs_info = rc->extent_root->fs_info;
+	struct inode *inode = rc->data_inode;
+	struct btrfs_root *data_reloc_root = BTRFS_I(inode)->root;
+	struct btrfs_path *path;
+	struct btrfs_key key;
+	int ret = 0;
+
+	if (!test_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags))
+		return 0;
+
+	/*
+	 * Only for stage where we update data pointers the qgroup fix is
+	 * valid.
+	 * For MOVING_DATA stage, we will miss the timing of swapping tree
+	 * blocks, and won't fix it.
+	 */
+	if (!(rc->stage == UPDATE_DATA_PTRS && rc->extents_found))
+		return 0;
+
+	path = btrfs_alloc_path();
+	if (!path)
+		return -ENOMEM;
+	key.objectid = btrfs_ino(inode);
+	key.type = BTRFS_EXTENT_DATA_KEY;
+	key.offset = 0;
+
+	ret = btrfs_search_slot(NULL, data_reloc_root, &key, path, 0, 0);
+	if (ret < 0)
+		goto out;
+
+	lock_extent(&BTRFS_I(inode)->io_tree, 0, (u64)-1);
+	while (1) {
+		struct btrfs_file_extent_item *fi;
+
+		btrfs_item_key_to_cpu(path->nodes[0], &key, path->slots[0]);
+		if (key.objectid > btrfs_ino(inode))
+			break;
+		if (key.type != BTRFS_EXTENT_DATA_KEY)
+			goto next;
+		fi = btrfs_item_ptr(path->nodes[0], path->slots[0],
+				    struct btrfs_file_extent_item);
+		if (btrfs_file_extent_type(path->nodes[0], fi) !=
+				BTRFS_FILE_EXTENT_REG)
+			goto next;
+		ret = btrfs_qgroup_insert_dirty_extent(trans, fs_info,
+			btrfs_file_extent_disk_bytenr(path->nodes[0], fi),
+			btrfs_file_extent_disk_num_bytes(path->nodes[0], fi),
+			GFP_NOFS);
+		if (ret < 0)
+			break;
+next:
+		ret = btrfs_next_item(data_reloc_root, path);
+		if (ret < 0)
+			break;
+		if (ret > 0) {
+			ret = 0;
+			break;
+		}
+	}
+	unlock_extent(&BTRFS_I(inode)->io_tree, 0 , (u64)-1);
+out:
+	btrfs_free_path(path);
+	return ret;
+}
+
 static noinline_for_stack int relocate_block_group(struct reloc_control *rc)
 {
 	struct rb_root blocks = RB_ROOT;
@@ -4089,10 +4212,18 @@ restart:
 
 	/* get rid of pinned extents */
 	trans = btrfs_join_transaction(rc->extent_root);
-	if (IS_ERR(trans))
+	if (IS_ERR(trans)) {
 		err = PTR_ERR(trans);
-	else
-		btrfs_commit_transaction(trans, rc->extent_root);
+		goto out_free;
+	}
+	ret = qgroup_fix_relocated_data_extents(trans, rc);
+	if (ret < 0) {
+		btrfs_abort_transaction(trans, ret);
+		if (!err)
+			err = ret;
+		goto out_free;
+	}
+	btrfs_commit_transaction(trans, rc->extent_root);
 out_free:
 	btrfs_free_block_rsv(rc->extent_root, rc->block_rsv);
 	btrfs_free_path(path);
@@ -4250,8 +4381,9 @@ int btrfs_relocate_block_group(struct btrfs_root *extent_root, u64 group_start)
 		goto out;
 	}
 
-	btrfs_info(extent_root->fs_info, "relocating block group %llu flags %llu",
-	       rc->block_group->key.objectid, rc->block_group->flags);
+	btrfs_info(extent_root->fs_info,
+		   "relocating block group %llu flags %llu",
+		   rc->block_group->key.objectid, rc->block_group->flags);
 
 	btrfs_wait_block_group_reservations(rc->block_group);
 	btrfs_wait_nocow_writers(rc->block_group);
@@ -4455,10 +4587,16 @@ int btrfs_recover_relocation(struct btrfs_root *root)
 	unset_reloc_control(rc);
 
 	trans = btrfs_join_transaction(rc->extent_root);
-	if (IS_ERR(trans))
+	if (IS_ERR(trans)) {
 		err = PTR_ERR(trans);
-	else
-		err = btrfs_commit_transaction(trans, rc->extent_root);
+		goto out_free;
+	}
+	err = qgroup_fix_relocated_data_extents(trans, rc);
+	if (err < 0) {
+		btrfs_abort_transaction(trans, err);
+		goto out_free;
+	}
+	err = btrfs_commit_transaction(trans, rc->extent_root);
 out_free:
 	kfree(rc);
 out:
@@ -4643,7 +4781,7 @@ int btrfs_reloc_post_snapshot(struct btrfs_trans_handle *trans,
 	if (rc->merge_reloc_tree) {
 		ret = btrfs_block_rsv_migrate(&pending->block_rsv,
 					      rc->block_rsv,
-					      rc->nodes_relocated);
+					      rc->nodes_relocated, 1);
 		if (ret)
 			return ret;
 	}
diff --git a/fs/btrfs/root-tree.c b/fs/btrfs/root-tree.c
index f1c30861d062..edae751e870c 100644
--- a/fs/btrfs/root-tree.c
+++ b/fs/btrfs/root-tree.c
@@ -46,12 +46,7 @@ static void btrfs_read_root_item(struct extent_buffer *eb, int slot,
 		!= btrfs_root_generation_v2(item)) {
 		if (btrfs_root_generation_v2(item) != 0) {
 			btrfs_warn(eb->fs_info,
-					"mismatching "
-					"generation and generation_v2 "
-					"found in root item. This root "
-					"was probably mounted with an "
-					"older kernel. Resetting all "
-					"new fields.");
+					"mismatching generation and generation_v2 found in root item. This root was probably mounted with an older kernel. Resetting all new fields.");
 		}
 		need_reset = 1;
 	}
@@ -150,14 +145,15 @@ int btrfs_update_root(struct btrfs_trans_handle *trans, struct btrfs_root
 
 	ret = btrfs_search_slot(trans, root, key, path, 0, 1);
 	if (ret < 0) {
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 		goto out;
 	}
 
 	if (ret != 0) {
 		btrfs_print_leaf(root, path->nodes[0]);
-		btrfs_crit(root->fs_info, "unable to update root key %llu %u %llu",
-		       key->objectid, key->type, key->offset);
+		btrfs_crit(root->fs_info,
+			   "unable to update root key %llu %u %llu",
+			   key->objectid, key->type, key->offset);
 		BUG_ON(1);
 	}
 
@@ -176,20 +172,20 @@ int btrfs_update_root(struct btrfs_trans_handle *trans, struct btrfs_root
 		ret = btrfs_search_slot(trans, root, key, path,
 				-1, 1);
 		if (ret < 0) {
-			btrfs_abort_transaction(trans, root, ret);
+			btrfs_abort_transaction(trans, ret);
 			goto out;
 		}
 
 		ret = btrfs_del_item(trans, root, path);
 		if (ret < 0) {
-			btrfs_abort_transaction(trans, root, ret);
+			btrfs_abort_transaction(trans, ret);
 			goto out;
 		}
 		btrfs_release_path(path);
 		ret = btrfs_insert_empty_item(trans, root, path,
 				key, sizeof(*item));
 		if (ret < 0) {
-			btrfs_abort_transaction(trans, root, ret);
+			btrfs_abort_transaction(trans, ret);
 			goto out;
 		}
 		l = path->nodes[0];
@@ -272,6 +268,23 @@ int btrfs_find_orphan_roots(struct btrfs_root *tree_root)
 		root_key.objectid = key.offset;
 		key.offset++;
 
+		/*
+		 * The root might have been inserted already, as before we look
+		 * for orphan roots, log replay might have happened, which
+		 * triggers a transaction commit and qgroup accounting, which
+		 * in turn reads and inserts fs roots while doing backref
+		 * walking.
+		 */
+		root = btrfs_lookup_fs_root(tree_root->fs_info,
+					    root_key.objectid);
+		if (root) {
+			WARN_ON(!test_bit(BTRFS_ROOT_ORPHAN_ITEM_INSERTED,
+					  &root->state));
+			if (btrfs_root_refs(&root->root_item) == 0)
+				btrfs_add_dead_root(root);
+			continue;
+		}
+
 		root = btrfs_read_fs_root(tree_root, &root_key);
 		err = PTR_ERR_OR_ZERO(root);
 		if (err && err != -ENOENT) {
@@ -285,8 +298,7 @@ int btrfs_find_orphan_roots(struct btrfs_root *tree_root)
 			if (IS_ERR(trans)) {
 				err = PTR_ERR(trans);
 				btrfs_handle_fs_error(tree_root->fs_info, err,
-					    "Failed to start trans to delete "
-					    "orphan item");
+					    "Failed to start trans to delete orphan item");
 				break;
 			}
 			err = btrfs_del_orphan_item(trans, tree_root,
@@ -294,8 +306,7 @@ int btrfs_find_orphan_roots(struct btrfs_root *tree_root)
 			btrfs_end_transaction(trans, tree_root);
 			if (err) {
 				btrfs_handle_fs_error(tree_root->fs_info, err,
-					    "Failed to delete root orphan "
-					    "item");
+					    "Failed to delete root orphan item");
 				break;
 			}
 			continue;
@@ -310,16 +321,8 @@ int btrfs_find_orphan_roots(struct btrfs_root *tree_root)
 		set_bit(BTRFS_ROOT_ORPHAN_ITEM_INSERTED, &root->state);
 
 		err = btrfs_insert_fs_root(root->fs_info, root);
-		/*
-		 * The root might have been inserted already, as before we look
-		 * for orphan roots, log replay might have happened, which
-		 * triggers a transaction commit and qgroup accounting, which
-		 * in turn reads and inserts fs roots while doing backref
-		 * walking.
-		 */
-		if (err == -EEXIST)
-			err = 0;
 		if (err) {
+			BUG_ON(err == -EEXIST);
 			btrfs_free_fs_root(root);
 			break;
 		}
@@ -448,7 +451,7 @@ again:
 	ret = btrfs_insert_empty_item(trans, tree_root, path, &key,
 				      sizeof(*ref) + name_len);
 	if (ret) {
-		btrfs_abort_transaction(trans, tree_root, ret);
+		btrfs_abort_transaction(trans, ret);
 		btrfs_free_path(path);
 		return ret;
 	}
diff --git a/fs/btrfs/scrub.c b/fs/btrfs/scrub.c
index 46d847f66e4b..fffb9ab8526e 100644
--- a/fs/btrfs/scrub.c
+++ b/fs/btrfs/scrub.c
@@ -575,23 +575,25 @@ static int scrub_print_warning_inode(u64 inum, u64 offset, u64 root,
 	 * hold all of the paths here
 	 */
 	for (i = 0; i < ipath->fspath->elem_cnt; ++i)
-		btrfs_warn_in_rcu(fs_info, "%s at logical %llu on dev "
-			"%s, sector %llu, root %llu, inode %llu, offset %llu, "
-			"length %llu, links %u (path: %s)", swarn->errstr,
-			swarn->logical, rcu_str_deref(swarn->dev->name),
-			(unsigned long long)swarn->sector, root, inum, offset,
-			min(isize - offset, (u64)PAGE_SIZE), nlink,
-			(char *)(unsigned long)ipath->fspath->val[i]);
+		btrfs_warn_in_rcu(fs_info,
+				  "%s at logical %llu on dev %s, sector %llu, root %llu, inode %llu, offset %llu, length %llu, links %u (path: %s)",
+				  swarn->errstr, swarn->logical,
+				  rcu_str_deref(swarn->dev->name),
+				  (unsigned long long)swarn->sector,
+				  root, inum, offset,
+				  min(isize - offset, (u64)PAGE_SIZE), nlink,
+				  (char *)(unsigned long)ipath->fspath->val[i]);
 
 	free_ipath(ipath);
 	return 0;
 
 err:
-	btrfs_warn_in_rcu(fs_info, "%s at logical %llu on dev "
-		"%s, sector %llu, root %llu, inode %llu, offset %llu: path "
-		"resolving failed with ret=%d", swarn->errstr,
-		swarn->logical, rcu_str_deref(swarn->dev->name),
-		(unsigned long long)swarn->sector, root, inum, offset, ret);
+	btrfs_warn_in_rcu(fs_info,
+			  "%s at logical %llu on dev %s, sector %llu, root %llu, inode %llu, offset %llu: path resolving failed with ret=%d",
+			  swarn->errstr, swarn->logical,
+			  rcu_str_deref(swarn->dev->name),
+			  (unsigned long long)swarn->sector,
+			  root, inum, offset, ret);
 
 	free_ipath(ipath);
 	return 0;
@@ -645,9 +647,8 @@ static void scrub_print_warning(const char *errstr, struct scrub_block *sblock)
 						      item_size, &ref_root,
 						      &ref_level);
 			btrfs_warn_in_rcu(fs_info,
-				"%s at logical %llu on dev %s, "
-				"sector %llu: metadata %s (level %d) in tree "
-				"%llu", errstr, swarn.logical,
+				"%s at logical %llu on dev %s, sector %llu: metadata %s (level %d) in tree %llu",
+				errstr, swarn.logical,
 				rcu_str_deref(dev->name),
 				(unsigned long long)swarn.sector,
 				ref_level ? "node" : "leaf",
@@ -1504,8 +1505,9 @@ static void scrub_recheck_block(struct btrfs_fs_info *fs_info,
 				sblock->no_io_error_seen = 0;
 		} else {
 			bio->bi_iter.bi_sector = page->physical >> 9;
+			bio_set_op_attrs(bio, REQ_OP_READ, 0);
 
-			if (btrfsic_submit_bio_wait(READ, bio))
+			if (btrfsic_submit_bio_wait(bio))
 				sblock->no_io_error_seen = 0;
 		}
 
@@ -1573,8 +1575,7 @@ static int scrub_repair_page_from_good_copy(struct scrub_block *sblock_bad,
 
 		if (!page_bad->dev->bdev) {
 			btrfs_warn_rl(sblock_bad->sctx->dev_root->fs_info,
-				"scrub_repair_page_from_good_copy(bdev == NULL) "
-				"is unexpected");
+				"scrub_repair_page_from_good_copy(bdev == NULL) is unexpected");
 			return -EIO;
 		}
 
@@ -1583,6 +1584,7 @@ static int scrub_repair_page_from_good_copy(struct scrub_block *sblock_bad,
 			return -EIO;
 		bio->bi_bdev = page_bad->dev->bdev;
 		bio->bi_iter.bi_sector = page_bad->physical >> 9;
+		bio_set_op_attrs(bio, REQ_OP_WRITE, 0);
 
 		ret = bio_add_page(bio, page_good->page, PAGE_SIZE, 0);
 		if (PAGE_SIZE != ret) {
@@ -1590,7 +1592,7 @@ static int scrub_repair_page_from_good_copy(struct scrub_block *sblock_bad,
 			return -EIO;
 		}
 
-		if (btrfsic_submit_bio_wait(WRITE, bio)) {
+		if (btrfsic_submit_bio_wait(bio)) {
 			btrfs_dev_stat_inc_and_print(page_bad->dev,
 				BTRFS_DEV_STAT_WRITE_ERRS);
 			btrfs_dev_replace_stats_inc(
@@ -1684,6 +1686,7 @@ again:
 		bio->bi_end_io = scrub_wr_bio_end_io;
 		bio->bi_bdev = sbio->dev->bdev;
 		bio->bi_iter.bi_sector = sbio->physical >> 9;
+		bio_set_op_attrs(bio, REQ_OP_WRITE, 0);
 		sbio->err = 0;
 	} else if (sbio->physical + sbio->page_count * PAGE_SIZE !=
 		   spage->physical_for_dev_replace ||
@@ -1731,7 +1734,7 @@ static void scrub_wr_submit(struct scrub_ctx *sctx)
 	 * orders the requests before sending them to the driver which
 	 * doubled the write performance on spinning disks when measured
 	 * with Linux 3.5 */
-	btrfsic_submit_bio(WRITE, sbio->bio);
+	btrfsic_submit_bio(sbio->bio);
 }
 
 static void scrub_wr_bio_end_io(struct bio *bio)
@@ -2041,7 +2044,7 @@ static void scrub_submit(struct scrub_ctx *sctx)
 	sbio = sctx->bios[sctx->curr];
 	sctx->curr = -1;
 	scrub_pending_bio_inc(sctx);
-	btrfsic_submit_bio(READ, sbio->bio);
+	btrfsic_submit_bio(sbio->bio);
 }
 
 static int scrub_add_page_to_rd_bio(struct scrub_ctx *sctx,
@@ -2088,6 +2091,7 @@ again:
 		bio->bi_end_io = scrub_bio_end_io;
 		bio->bi_bdev = sbio->dev->bdev;
 		bio->bi_iter.bi_sector = sbio->physical >> 9;
+		bio_set_op_attrs(bio, REQ_OP_READ, 0);
 		sbio->err = 0;
 	} else if (sbio->physical + sbio->page_count * PAGE_SIZE !=
 		   spage->physical ||
@@ -2957,7 +2961,8 @@ static noinline_for_stack int scrub_raid56_parity(struct scrub_ctx *sctx,
 			    (key.objectid < logic_start ||
 			     key.objectid + bytes >
 			     logic_start + map->stripe_len)) {
-				btrfs_err(fs_info, "scrub: tree block %llu spanning stripes, ignored. logical=%llu",
+				btrfs_err(fs_info,
+					  "scrub: tree block %llu spanning stripes, ignored. logical=%llu",
 					  key.objectid, logic_start);
 				spin_lock(&sctx->stat_lock);
 				sctx->stat.uncorrectable_errors++;
@@ -3308,8 +3313,7 @@ static noinline_for_stack int scrub_stripe(struct scrub_ctx *sctx,
 			     key.objectid + bytes >
 			     logical + map->stripe_len)) {
 				btrfs_err(fs_info,
-					   "scrub: tree block %llu spanning "
-					   "stripes, ignored. logical=%llu",
+					   "scrub: tree block %llu spanning stripes, ignored. logical=%llu",
 				       key.objectid, logical);
 				spin_lock(&sctx->stat_lock);
 				sctx->stat.uncorrectable_errors++;
@@ -3582,6 +3586,46 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx,
 		 */
 		scrub_pause_on(fs_info);
 		ret = btrfs_inc_block_group_ro(root, cache);
+		if (!ret && is_dev_replace) {
+			/*
+			 * If we are doing a device replace wait for any tasks
+			 * that started dellaloc right before we set the block
+			 * group to RO mode, as they might have just allocated
+			 * an extent from it or decided they could do a nocow
+			 * write. And if any such tasks did that, wait for their
+			 * ordered extents to complete and then commit the
+			 * current transaction, so that we can later see the new
+			 * extent items in the extent tree - the ordered extents
+			 * create delayed data references (for cow writes) when
+			 * they complete, which will be run and insert the
+			 * corresponding extent items into the extent tree when
+			 * we commit the transaction they used when running
+			 * inode.c:btrfs_finish_ordered_io(). We later use
+			 * the commit root of the extent tree to find extents
+			 * to copy from the srcdev into the tgtdev, and we don't
+			 * want to miss any new extents.
+			 */
+			btrfs_wait_block_group_reservations(cache);
+			btrfs_wait_nocow_writers(cache);
+			ret = btrfs_wait_ordered_roots(fs_info, -1,
+						       cache->key.objectid,
+						       cache->key.offset);
+			if (ret > 0) {
+				struct btrfs_trans_handle *trans;
+
+				trans = btrfs_join_transaction(root);
+				if (IS_ERR(trans))
+					ret = PTR_ERR(trans);
+				else
+					ret = btrfs_commit_transaction(trans,
+								       root);
+				if (ret) {
+					scrub_pause_off(fs_info);
+					btrfs_put_block_group(cache);
+					break;
+				}
+			}
+		}
 		scrub_pause_off(fs_info);
 
 		if (ret == 0) {
@@ -3596,15 +3640,18 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx,
 			 */
 			ro_set = 0;
 		} else {
-			btrfs_warn(fs_info, "failed setting block group ro, ret=%d\n",
+			btrfs_warn(fs_info,
+				   "failed setting block group ro, ret=%d\n",
 				   ret);
 			btrfs_put_block_group(cache);
 			break;
 		}
 
+		btrfs_dev_replace_lock(&fs_info->dev_replace, 1);
 		dev_replace->cursor_right = found_key.offset + length;
 		dev_replace->cursor_left = found_key.offset;
 		dev_replace->item_needs_writeback = 1;
+		btrfs_dev_replace_unlock(&fs_info->dev_replace, 1);
 		ret = scrub_chunk(sctx, scrub_dev, chunk_offset, length,
 				  found_key.offset, cache, is_dev_replace);
 
@@ -3640,6 +3687,11 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx,
 
 		scrub_pause_off(fs_info);
 
+		btrfs_dev_replace_lock(&fs_info->dev_replace, 1);
+		dev_replace->cursor_left = dev_replace->cursor_right;
+		dev_replace->item_needs_writeback = 1;
+		btrfs_dev_replace_unlock(&fs_info->dev_replace, 1);
+
 		if (ro_set)
 			btrfs_dec_block_group_ro(root, cache);
 
@@ -3677,9 +3729,6 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx,
 			ret = -ENOMEM;
 			break;
 		}
-
-		dev_replace->cursor_left = dev_replace->cursor_right;
-		dev_replace->item_needs_writeback = 1;
 skip:
 		key.offset = found_key.offset + length;
 		btrfs_release_path(path);
@@ -3737,27 +3786,27 @@ static noinline_for_stack int scrub_workers_get(struct btrfs_fs_info *fs_info,
 	if (fs_info->scrub_workers_refcnt == 0) {
 		if (is_dev_replace)
 			fs_info->scrub_workers =
-				btrfs_alloc_workqueue("scrub", flags,
+				btrfs_alloc_workqueue(fs_info, "scrub", flags,
 						      1, 4);
 		else
 			fs_info->scrub_workers =
-				btrfs_alloc_workqueue("scrub", flags,
+				btrfs_alloc_workqueue(fs_info, "scrub", flags,
 						      max_active, 4);
 		if (!fs_info->scrub_workers)
 			goto fail_scrub_workers;
 
 		fs_info->scrub_wr_completion_workers =
-			btrfs_alloc_workqueue("scrubwrc", flags,
+			btrfs_alloc_workqueue(fs_info, "scrubwrc", flags,
 					      max_active, 2);
 		if (!fs_info->scrub_wr_completion_workers)
 			goto fail_scrub_wr_completion_workers;
 
 		fs_info->scrub_nocow_workers =
-			btrfs_alloc_workqueue("scrubnc", flags, 1, 0);
+			btrfs_alloc_workqueue(fs_info, "scrubnc", flags, 1, 0);
 		if (!fs_info->scrub_nocow_workers)
 			goto fail_scrub_nocow_workers;
 		fs_info->scrub_parity_workers =
-			btrfs_alloc_workqueue("scrubparity", flags,
+			btrfs_alloc_workqueue(fs_info, "scrubparity", flags,
 					      max_active, 2);
 		if (!fs_info->scrub_parity_workers)
 			goto fail_scrub_parity_workers;
@@ -3812,9 +3861,8 @@ int btrfs_scrub_dev(struct btrfs_fs_info *fs_info, u64 devid, u64 start,
 
 	if (fs_info->chunk_root->sectorsize != PAGE_SIZE) {
 		/* not supported for data w/o checksums */
-		btrfs_err(fs_info,
-			   "scrub: size assumption sectorsize != PAGE_SIZE "
-			   "(%d != %lu) fails",
+		btrfs_err_rl(fs_info,
+			   "scrub: size assumption sectorsize != PAGE_SIZE (%d != %lu) fails",
 		       fs_info->chunk_root->sectorsize, PAGE_SIZE);
 		return -EINVAL;
 	}
@@ -3827,8 +3875,8 @@ int btrfs_scrub_dev(struct btrfs_fs_info *fs_info, u64 devid, u64 start,
 		 * would exhaust the array bounds of pagev member in
 		 * struct scrub_block
 		 */
-		btrfs_err(fs_info, "scrub: size assumption nodesize and sectorsize "
-			   "<= SCRUB_MAX_PAGES_PER_BLOCK (%d <= %d && %d <= %d) fails",
+		btrfs_err(fs_info,
+			  "scrub: size assumption nodesize and sectorsize <= SCRUB_MAX_PAGES_PER_BLOCK (%d <= %d && %d <= %d) fails",
 		       fs_info->chunk_root->nodesize,
 		       SCRUB_MAX_PAGES_PER_BLOCK,
 		       fs_info->chunk_root->sectorsize,
@@ -4154,10 +4202,10 @@ static void copy_nocow_pages_worker(struct btrfs_work *work)
 	ret = iterate_inodes_from_logical(logical, fs_info, path,
 					  record_inode_for_nocow, nocow_ctx);
 	if (ret != 0 && ret != -ENOENT) {
-		btrfs_warn(fs_info, "iterate_inodes_from_logical() failed: log %llu, "
-			"phys %llu, len %llu, mir %u, ret %d",
-			logical, physical_for_dev_replace, len, mirror_num,
-			ret);
+		btrfs_warn(fs_info,
+			   "iterate_inodes_from_logical() failed: log %llu, phys %llu, len %llu, mir %u, ret %d",
+			   logical, physical_for_dev_replace, len, mirror_num,
+			   ret);
 		not_written = 1;
 		goto out;
 	}
@@ -4392,6 +4440,7 @@ static int write_page_nocow(struct scrub_ctx *sctx,
 	bio->bi_iter.bi_size = 0;
 	bio->bi_iter.bi_sector = physical_for_dev_replace >> 9;
 	bio->bi_bdev = dev->bdev;
+	bio_set_op_attrs(bio, REQ_OP_WRITE, WRITE_SYNC);
 	ret = bio_add_page(bio, page, PAGE_SIZE, 0);
 	if (ret != PAGE_SIZE) {
 leave_with_eio:
@@ -4400,7 +4449,7 @@ leave_with_eio:
 		return -EIO;
 	}
 
-	if (btrfsic_submit_bio_wait(WRITE_SYNC, bio))
+	if (btrfsic_submit_bio_wait(bio))
 		goto leave_with_eio;
 
 	bio_put(bio);
diff --git a/fs/btrfs/send.c b/fs/btrfs/send.c
index b71dd298385c..01bc36cec26e 100644
--- a/fs/btrfs/send.c
+++ b/fs/btrfs/send.c
@@ -36,10 +36,6 @@
 #include "transaction.h"
 #include "compression.h"
 
-static int g_verbose = 0;
-
-#define verbose_printk(...) if (g_verbose) printk(__VA_ARGS__)
-
 /*
  * A fs_path is a helper to dynamically build path names with unknown size.
  * It reallocates the internal buffer on demand.
@@ -231,7 +227,6 @@ struct pending_dir_move {
 	u64 parent_ino;
 	u64 ino;
 	u64 gen;
-	bool is_orphan;
 	struct list_head update_refs;
 };
 
@@ -274,6 +269,39 @@ struct name_cache_entry {
 	char name[];
 };
 
+static void inconsistent_snapshot_error(struct send_ctx *sctx,
+					enum btrfs_compare_tree_result result,
+					const char *what)
+{
+	const char *result_string;
+
+	switch (result) {
+	case BTRFS_COMPARE_TREE_NEW:
+		result_string = "new";
+		break;
+	case BTRFS_COMPARE_TREE_DELETED:
+		result_string = "deleted";
+		break;
+	case BTRFS_COMPARE_TREE_CHANGED:
+		result_string = "updated";
+		break;
+	case BTRFS_COMPARE_TREE_SAME:
+		ASSERT(0);
+		result_string = "unchanged";
+		break;
+	default:
+		ASSERT(0);
+		result_string = "unexpected";
+	}
+
+	btrfs_err(sctx->send_root->fs_info,
+		  "Send: inconsistent snapshot, found %s %s for inode %llu without updated inode item, send root is %llu, parent root is %llu",
+		  result_string, what, sctx->cmp_key->objectid,
+		  sctx->send_root->root_key.objectid,
+		  (sctx->parent_root ?
+		   sctx->parent_root->root_key.objectid : 0));
+}
+
 static int is_waiting_for_move(struct send_ctx *sctx, u64 ino);
 
 static struct waiting_dir_move *
@@ -695,9 +723,10 @@ static int send_cmd(struct send_ctx *sctx)
 static int send_rename(struct send_ctx *sctx,
 		     struct fs_path *from, struct fs_path *to)
 {
+	struct btrfs_fs_info *fs_info = sctx->send_root->fs_info;
 	int ret;
 
-verbose_printk("btrfs: send_rename %s -> %s\n", from->start, to->start);
+	btrfs_debug(fs_info, "send_rename %s -> %s", from->start, to->start);
 
 	ret = begin_cmd(sctx, BTRFS_SEND_C_RENAME);
 	if (ret < 0)
@@ -719,9 +748,10 @@ out:
 static int send_link(struct send_ctx *sctx,
 		     struct fs_path *path, struct fs_path *lnk)
 {
+	struct btrfs_fs_info *fs_info = sctx->send_root->fs_info;
 	int ret;
 
-verbose_printk("btrfs: send_link %s -> %s\n", path->start, lnk->start);
+	btrfs_debug(fs_info, "send_link %s -> %s", path->start, lnk->start);
 
 	ret = begin_cmd(sctx, BTRFS_SEND_C_LINK);
 	if (ret < 0)
@@ -742,9 +772,10 @@ out:
  */
 static int send_unlink(struct send_ctx *sctx, struct fs_path *path)
 {
+	struct btrfs_fs_info *fs_info = sctx->send_root->fs_info;
 	int ret;
 
-verbose_printk("btrfs: send_unlink %s\n", path->start);
+	btrfs_debug(fs_info, "send_unlink %s", path->start);
 
 	ret = begin_cmd(sctx, BTRFS_SEND_C_UNLINK);
 	if (ret < 0)
@@ -764,9 +795,10 @@ out:
  */
 static int send_rmdir(struct send_ctx *sctx, struct fs_path *path)
 {
+	struct btrfs_fs_info *fs_info = sctx->send_root->fs_info;
 	int ret;
 
-verbose_printk("btrfs: send_rmdir %s\n", path->start);
+	btrfs_debug(fs_info, "send_rmdir %s", path->start);
 
 	ret = begin_cmd(sctx, BTRFS_SEND_C_RMDIR);
 	if (ret < 0)
@@ -1281,6 +1313,7 @@ static int find_extent_clone(struct send_ctx *sctx,
 			     u64 ino_size,
 			     struct clone_root **found)
 {
+	struct btrfs_fs_info *fs_info = sctx->send_root->fs_info;
 	int ret;
 	int extent_type;
 	u64 logical;
@@ -1339,10 +1372,10 @@ static int find_extent_clone(struct send_ctx *sctx,
 	}
 	logical = disk_byte + btrfs_file_extent_offset(eb, fi);
 
-	down_read(&sctx->send_root->fs_info->commit_root_sem);
-	ret = extent_from_logical(sctx->send_root->fs_info, disk_byte, tmp_path,
+	down_read(&fs_info->commit_root_sem);
+	ret = extent_from_logical(fs_info, disk_byte, tmp_path,
 				  &found_key, &flags);
-	up_read(&sctx->send_root->fs_info->commit_root_sem);
+	up_read(&fs_info->commit_root_sem);
 	btrfs_release_path(tmp_path);
 
 	if (ret < 0)
@@ -1397,7 +1430,7 @@ static int find_extent_clone(struct send_ctx *sctx,
 		extent_item_pos = logical - found_key.objectid;
 	else
 		extent_item_pos = 0;
-	ret = iterate_extent_inodes(sctx->send_root->fs_info,
+	ret = iterate_extent_inodes(fs_info,
 					found_key.objectid, extent_item_pos, 1,
 					__iterate_backrefs, backref_ctx);
 
@@ -1407,20 +1440,18 @@ static int find_extent_clone(struct send_ctx *sctx,
 	if (!backref_ctx->found_itself) {
 		/* found a bug in backref code? */
 		ret = -EIO;
-		btrfs_err(sctx->send_root->fs_info, "did not find backref in "
-				"send_root. inode=%llu, offset=%llu, "
-				"disk_byte=%llu found extent=%llu",
-				ino, data_offset, disk_byte, found_key.objectid);
+		btrfs_err(fs_info,
+			  "did not find backref in send_root. inode=%llu, offset=%llu, disk_byte=%llu found extent=%llu",
+			  ino, data_offset, disk_byte, found_key.objectid);
 		goto out;
 	}
 
-verbose_printk(KERN_DEBUG "btrfs: find_extent_clone: data_offset=%llu, "
-		"ino=%llu, "
-		"num_bytes=%llu, logical=%llu\n",
-		data_offset, ino, num_bytes, logical);
+	btrfs_debug(fs_info,
+		    "find_extent_clone: data_offset=%llu, ino=%llu, num_bytes=%llu, logical=%llu",
+		    data_offset, ino, num_bytes, logical);
 
 	if (!backref_ctx->found)
-		verbose_printk("btrfs:    no clones found\n");
+		btrfs_debug(fs_info, "no clones found");
 
 	cur_clone_root = NULL;
 	for (i = 0; i < sctx->clone_roots_cnt; i++) {
@@ -1861,7 +1892,8 @@ static int will_overwrite_ref(struct send_ctx *sctx, u64 dir, u64 dir_gen,
 	 * was already unlinked/moved, so we can safely assume that we will not
 	 * overwrite anything at this point in time.
 	 */
-	if (other_inode > sctx->send_progress) {
+	if (other_inode > sctx->send_progress ||
+	    is_waiting_for_move(sctx, other_inode)) {
 		ret = get_inode_info(sctx->parent_root, other_inode, NULL,
 				who_gen, NULL, NULL, NULL, NULL);
 		if (ret < 0)
@@ -2390,10 +2422,11 @@ out:
 
 static int send_truncate(struct send_ctx *sctx, u64 ino, u64 gen, u64 size)
 {
+	struct btrfs_fs_info *fs_info = sctx->send_root->fs_info;
 	int ret = 0;
 	struct fs_path *p;
 
-verbose_printk("btrfs: send_truncate %llu size=%llu\n", ino, size);
+	btrfs_debug(fs_info, "send_truncate %llu size=%llu", ino, size);
 
 	p = fs_path_alloc();
 	if (!p)
@@ -2419,10 +2452,11 @@ out:
 
 static int send_chmod(struct send_ctx *sctx, u64 ino, u64 gen, u64 mode)
 {
+	struct btrfs_fs_info *fs_info = sctx->send_root->fs_info;
 	int ret = 0;
 	struct fs_path *p;
 
-verbose_printk("btrfs: send_chmod %llu mode=%llu\n", ino, mode);
+	btrfs_debug(fs_info, "send_chmod %llu mode=%llu", ino, mode);
 
 	p = fs_path_alloc();
 	if (!p)
@@ -2448,10 +2482,12 @@ out:
 
 static int send_chown(struct send_ctx *sctx, u64 ino, u64 gen, u64 uid, u64 gid)
 {
+	struct btrfs_fs_info *fs_info = sctx->send_root->fs_info;
 	int ret = 0;
 	struct fs_path *p;
 
-verbose_printk("btrfs: send_chown %llu uid=%llu, gid=%llu\n", ino, uid, gid);
+	btrfs_debug(fs_info, "send_chown %llu uid=%llu, gid=%llu",
+		    ino, uid, gid);
 
 	p = fs_path_alloc();
 	if (!p)
@@ -2478,6 +2514,7 @@ out:
 
 static int send_utimes(struct send_ctx *sctx, u64 ino, u64 gen)
 {
+	struct btrfs_fs_info *fs_info = sctx->send_root->fs_info;
 	int ret = 0;
 	struct fs_path *p = NULL;
 	struct btrfs_inode_item *ii;
@@ -2486,7 +2523,7 @@ static int send_utimes(struct send_ctx *sctx, u64 ino, u64 gen)
 	struct btrfs_key key;
 	int slot;
 
-verbose_printk("btrfs: send_utimes %llu\n", ino);
+	btrfs_debug(fs_info, "send_utimes %llu", ino);
 
 	p = fs_path_alloc();
 	if (!p)
@@ -2502,6 +2539,8 @@ verbose_printk("btrfs: send_utimes %llu\n", ino);
 	key.type = BTRFS_INODE_ITEM_KEY;
 	key.offset = 0;
 	ret = btrfs_search_slot(NULL, sctx->send_root, &key, path, 0, 0);
+	if (ret > 0)
+		ret = -ENOENT;
 	if (ret < 0)
 		goto out;
 
@@ -2538,6 +2577,7 @@ out:
  */
 static int send_create_inode(struct send_ctx *sctx, u64 ino)
 {
+	struct btrfs_fs_info *fs_info = sctx->send_root->fs_info;
 	int ret = 0;
 	struct fs_path *p;
 	int cmd;
@@ -2545,7 +2585,7 @@ static int send_create_inode(struct send_ctx *sctx, u64 ino)
 	u64 mode;
 	u64 rdev;
 
-verbose_printk("btrfs: send_create_inode %llu\n", ino);
+	btrfs_debug(fs_info, "send_create_inode %llu", ino);
 
 	p = fs_path_alloc();
 	if (!p)
@@ -2947,6 +2987,10 @@ static int can_rmdir(struct send_ctx *sctx, u64 dir, u64 dir_gen,
 		}
 
 		if (loc.objectid > send_progress) {
+			struct orphan_dir_info *odi;
+
+			odi = get_orphan_dir_info(sctx, dir);
+			free_orphan_dir_info(sctx, odi);
 			ret = 0;
 			goto out;
 		}
@@ -3047,7 +3091,6 @@ static int add_pending_dir_move(struct send_ctx *sctx,
 	pm->parent_ino = parent_ino;
 	pm->ino = ino;
 	pm->gen = ino_gen;
-	pm->is_orphan = is_orphan;
 	INIT_LIST_HEAD(&pm->list);
 	INIT_LIST_HEAD(&pm->update_refs);
 	RB_CLEAR_NODE(&pm->node);
@@ -3113,6 +3156,48 @@ static struct pending_dir_move *get_pending_dir_moves(struct send_ctx *sctx,
 	return NULL;
 }
 
+static int path_loop(struct send_ctx *sctx, struct fs_path *name,
+		     u64 ino, u64 gen, u64 *ancestor_ino)
+{
+	int ret = 0;
+	u64 parent_inode = 0;
+	u64 parent_gen = 0;
+	u64 start_ino = ino;
+
+	*ancestor_ino = 0;
+	while (ino != BTRFS_FIRST_FREE_OBJECTID) {
+		fs_path_reset(name);
+
+		if (is_waiting_for_rm(sctx, ino))
+			break;
+		if (is_waiting_for_move(sctx, ino)) {
+			if (*ancestor_ino == 0)
+				*ancestor_ino = ino;
+			ret = get_first_ref(sctx->parent_root, ino,
+					    &parent_inode, &parent_gen, name);
+		} else {
+			ret = __get_cur_name_and_parent(sctx, ino, gen,
+							&parent_inode,
+							&parent_gen, name);
+			if (ret > 0) {
+				ret = 0;
+				break;
+			}
+		}
+		if (ret < 0)
+			break;
+		if (parent_inode == start_ino) {
+			ret = 1;
+			if (*ancestor_ino == 0)
+				*ancestor_ino = ino;
+			break;
+		}
+		ino = parent_inode;
+		gen = parent_gen;
+	}
+	return ret;
+}
+
 static int apply_dir_move(struct send_ctx *sctx, struct pending_dir_move *pm)
 {
 	struct fs_path *from_path = NULL;
@@ -3123,6 +3208,8 @@ static int apply_dir_move(struct send_ctx *sctx, struct pending_dir_move *pm)
 	u64 parent_ino, parent_gen;
 	struct waiting_dir_move *dm = NULL;
 	u64 rmdir_ino = 0;
+	u64 ancestor;
+	bool is_orphan;
 	int ret;
 
 	name = fs_path_alloc();
@@ -3135,9 +3222,10 @@ static int apply_dir_move(struct send_ctx *sctx, struct pending_dir_move *pm)
 	dm = get_waiting_dir_move(sctx, pm->ino);
 	ASSERT(dm);
 	rmdir_ino = dm->rmdir_ino;
+	is_orphan = dm->orphanized;
 	free_waiting_dir_move(sctx, dm);
 
-	if (pm->is_orphan) {
+	if (is_orphan) {
 		ret = gen_unique_name(sctx, pm->ino,
 				      pm->gen, from_path);
 	} else {
@@ -3155,6 +3243,24 @@ static int apply_dir_move(struct send_ctx *sctx, struct pending_dir_move *pm)
 		goto out;
 
 	sctx->send_progress = sctx->cur_ino + 1;
+	ret = path_loop(sctx, name, pm->ino, pm->gen, &ancestor);
+	if (ret < 0)
+		goto out;
+	if (ret) {
+		LIST_HEAD(deleted_refs);
+		ASSERT(ancestor > BTRFS_FIRST_FREE_OBJECTID);
+		ret = add_pending_dir_move(sctx, pm->ino, pm->gen, ancestor,
+					   &pm->update_refs, &deleted_refs,
+					   is_orphan);
+		if (ret < 0)
+			goto out;
+		if (rmdir_ino) {
+			dm = get_waiting_dir_move(sctx, pm->ino);
+			ASSERT(dm);
+			dm->rmdir_ino = rmdir_ino;
+		}
+		goto out;
+	}
 	fs_path_reset(name);
 	to_path = name;
 	name = NULL;
@@ -3174,7 +3280,7 @@ static int apply_dir_move(struct send_ctx *sctx, struct pending_dir_move *pm)
 			/* already deleted */
 			goto finish;
 		}
-		ret = can_rmdir(sctx, rmdir_ino, odi->gen, sctx->cur_ino + 1);
+		ret = can_rmdir(sctx, rmdir_ino, odi->gen, sctx->cur_ino);
 		if (ret < 0)
 			goto out;
 		if (!ret)
@@ -3204,8 +3310,18 @@ finish:
 	 * and old parent(s).
 	 */
 	list_for_each_entry(cur, &pm->update_refs, list) {
-		if (cur->dir == rmdir_ino)
+		/*
+		 * The parent inode might have been deleted in the send snapshot
+		 */
+		ret = get_inode_info(sctx->send_root, cur->dir, NULL,
+				     NULL, NULL, NULL, NULL, NULL);
+		if (ret == -ENOENT) {
+			ret = 0;
 			continue;
+		}
+		if (ret < 0)
+			goto out;
+
 		ret = send_utimes(sctx, cur->dir, cur->dir_gen);
 		if (ret < 0)
 			goto out;
@@ -3325,6 +3441,7 @@ static int wait_for_dest_dir_move(struct send_ctx *sctx,
 	u64 left_gen;
 	u64 right_gen;
 	int ret = 0;
+	struct waiting_dir_move *wdm;
 
 	if (RB_EMPTY_ROOT(&sctx->waiting_dir_moves))
 		return 0;
@@ -3383,7 +3500,8 @@ static int wait_for_dest_dir_move(struct send_ctx *sctx,
 		goto out;
 	}
 
-	if (is_waiting_for_move(sctx, di_key.objectid)) {
+	wdm = get_waiting_dir_move(sctx, di_key.objectid);
+	if (wdm && !wdm->orphanized) {
 		ret = add_pending_dir_move(sctx,
 					   sctx->cur_ino,
 					   sctx->cur_inode_gen,
@@ -3470,7 +3588,8 @@ static int wait_for_parent_move(struct send_ctx *sctx,
 			ret = is_ancestor(sctx->parent_root,
 					  sctx->cur_ino, sctx->cur_inode_gen,
 					  ino, path_before);
-			break;
+			if (ret)
+				break;
 		}
 
 		fs_path_reset(path_before);
@@ -3524,6 +3643,7 @@ out:
  */
 static int process_recorded_refs(struct send_ctx *sctx, int *pending_move)
 {
+	struct btrfs_fs_info *fs_info = sctx->send_root->fs_info;
 	int ret = 0;
 	struct recorded_ref *cur;
 	struct recorded_ref *cur2;
@@ -3536,7 +3656,7 @@ static int process_recorded_refs(struct send_ctx *sctx, int *pending_move)
 	u64 last_dir_ino_rm = 0;
 	bool can_rename = true;
 
-verbose_printk("btrfs: process_recorded_refs %llu\n", sctx->cur_ino);
+	btrfs_debug(fs_info, "process_recorded_refs %llu", sctx->cur_ino);
 
 	/*
 	 * This should never happen as the root dir always has the same ref
@@ -3643,11 +3763,26 @@ verbose_printk("btrfs: process_recorded_refs %llu\n", sctx->cur_ino);
 				goto out;
 			if (ret) {
 				struct name_cache_entry *nce;
+				struct waiting_dir_move *wdm;
 
 				ret = orphanize_inode(sctx, ow_inode, ow_gen,
 						cur->full_path);
 				if (ret < 0)
 					goto out;
+
+				/*
+				 * If ow_inode has its rename operation delayed
+				 * make sure that its orphanized name is used in
+				 * the source path when performing its rename
+				 * operation.
+				 */
+				if (is_waiting_for_move(sctx, ow_inode)) {
+					wdm = get_waiting_dir_move(sctx,
+								   ow_inode);
+					ASSERT(wdm);
+					wdm->orphanized = true;
+				}
+
 				/*
 				 * Make sure we clear our orphanized inode's
 				 * name from the name cache. This is because the
@@ -3663,6 +3798,19 @@ verbose_printk("btrfs: process_recorded_refs %llu\n", sctx->cur_ino);
 					name_cache_delete(sctx, nce);
 					kfree(nce);
 				}
+
+				/*
+				 * ow_inode might currently be an ancestor of
+				 * cur_ino, therefore compute valid_path (the
+				 * current path of cur_ino) again because it
+				 * might contain the pre-orphanization name of
+				 * ow_inode, which is no longer valid.
+				 */
+				fs_path_reset(valid_path);
+				ret = get_cur_path(sctx, sctx->cur_ino,
+					   sctx->cur_inode_gen, valid_path);
+				if (ret < 0)
+					goto out;
 			} else {
 				ret = send_unlink(sctx, cur->full_path);
 				if (ret < 0)
@@ -4126,10 +4274,12 @@ static int process_all_refs(struct send_ctx *sctx,
 	}
 	btrfs_release_path(path);
 
+	/*
+	 * We don't actually care about pending_move as we are simply
+	 * re-creating this inode and will be rename'ing it into place once we
+	 * rename the parent directory.
+	 */
 	ret = process_recorded_refs(sctx, &pending_move);
-	/* Only applicable to an incremental send. */
-	ASSERT(pending_move == 0);
-
 out:
 	btrfs_free_path(path);
 	return ret;
@@ -4185,7 +4335,7 @@ static int __process_new_xattr(int num, struct btrfs_key *di_key,
 	int ret;
 	struct send_ctx *sctx = ctx;
 	struct fs_path *p;
-	posix_acl_xattr_header dummy_acl;
+	struct posix_acl_xattr_header dummy_acl;
 
 	p = fs_path_alloc();
 	if (!p)
@@ -4254,12 +4404,8 @@ static int process_new_xattr(struct send_ctx *sctx)
 
 static int process_deleted_xattr(struct send_ctx *sctx)
 {
-	int ret;
-
-	ret = iterate_dir_item(sctx->parent_root, sctx->right_path,
-			       sctx->cmp_key, __process_deleted_xattr, sctx);
-
-	return ret;
+	return iterate_dir_item(sctx->parent_root, sctx->right_path,
+				sctx->cmp_key, __process_deleted_xattr, sctx);
 }
 
 struct find_xattr_ctx {
@@ -4520,6 +4666,7 @@ out:
  */
 static int send_write(struct send_ctx *sctx, u64 offset, u32 len)
 {
+	struct btrfs_fs_info *fs_info = sctx->send_root->fs_info;
 	int ret = 0;
 	struct fs_path *p;
 	ssize_t num_read = 0;
@@ -4528,7 +4675,7 @@ static int send_write(struct send_ctx *sctx, u64 offset, u32 len)
 	if (!p)
 		return -ENOMEM;
 
-verbose_printk("btrfs: send_write offset=%llu, len=%d\n", offset, len);
+	btrfs_debug(fs_info, "send_write offset=%llu, len=%d", offset, len);
 
 	num_read = fill_read_buf(sctx, offset, len);
 	if (num_read <= 0) {
@@ -4570,10 +4717,10 @@ static int send_clone(struct send_ctx *sctx,
 	struct fs_path *p;
 	u64 gen;
 
-verbose_printk("btrfs: send_clone offset=%llu, len=%d, clone_root=%llu, "
-	       "clone_inode=%llu, clone_offset=%llu\n", offset, len,
-		clone_root->root->objectid, clone_root->ino,
-		clone_root->offset);
+	btrfs_debug(sctx->send_root->fs_info,
+		    "send_clone offset=%llu, len=%d, clone_root=%llu, clone_inode=%llu, clone_offset=%llu",
+		    offset, len, clone_root->root->objectid, clone_root->ino,
+		    clone_root->offset);
 
 	p = fs_path_alloc();
 	if (!p)
@@ -5602,7 +5749,10 @@ static int changed_ref(struct send_ctx *sctx,
 {
 	int ret = 0;
 
-	BUG_ON(sctx->cur_ino != sctx->cmp_key->objectid);
+	if (sctx->cur_ino != sctx->cmp_key->objectid) {
+		inconsistent_snapshot_error(sctx, result, "reference");
+		return -EIO;
+	}
 
 	if (!sctx->cur_inode_new_gen &&
 	    sctx->cur_ino != BTRFS_FIRST_FREE_OBJECTID) {
@@ -5627,7 +5777,10 @@ static int changed_xattr(struct send_ctx *sctx,
 {
 	int ret = 0;
 
-	BUG_ON(sctx->cur_ino != sctx->cmp_key->objectid);
+	if (sctx->cur_ino != sctx->cmp_key->objectid) {
+		inconsistent_snapshot_error(sctx, result, "xattr");
+		return -EIO;
+	}
 
 	if (!sctx->cur_inode_new_gen && !sctx->cur_inode_deleted) {
 		if (result == BTRFS_COMPARE_TREE_NEW)
@@ -5651,7 +5804,10 @@ static int changed_extent(struct send_ctx *sctx,
 {
 	int ret = 0;
 
-	BUG_ON(sctx->cur_ino != sctx->cmp_key->objectid);
+	if (sctx->cur_ino != sctx->cmp_key->objectid) {
+		inconsistent_snapshot_error(sctx, result, "extent");
+		return -EIO;
+	}
 
 	if (!sctx->cur_inode_new_gen && !sctx->cur_inode_deleted) {
 		if (result != BTRFS_COMPARE_TREE_DELETED)
diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c
index 4e59a91a11e0..74ed5aae6cea 100644
--- a/fs/btrfs/super.c
+++ b/fs/btrfs/super.c
@@ -151,12 +151,11 @@ void __btrfs_handle_fs_error(struct btrfs_fs_info *fs_info, const char *function
 		vaf.fmt = fmt;
 		vaf.va = &args;
 
-		printk(KERN_CRIT
-			"BTRFS: error (device %s) in %s:%d: errno=%d %s (%pV)\n",
+		pr_crit("BTRFS: error (device %s) in %s:%d: errno=%d %s (%pV)\n",
 			sb->s_id, function, line, errno, errstr, &vaf);
 		va_end(args);
 	} else {
-		printk(KERN_CRIT "BTRFS: error (device %s) in %s:%d: errno=%d %s\n",
+		pr_crit("BTRFS: error (device %s) in %s:%d: errno=%d %s\n",
 			sb->s_id, function, line, errno, errstr);
 	}
 #endif
@@ -184,6 +183,22 @@ static const char * const logtypes[] = {
 	"debug",
 };
 
+
+/*
+ * Use one ratelimit state per log level so that a flood of less important
+ * messages doesn't cause more important ones to be dropped.
+ */
+static struct ratelimit_state printk_limits[] = {
+	RATELIMIT_STATE_INIT(printk_limits[0], DEFAULT_RATELIMIT_INTERVAL, 100),
+	RATELIMIT_STATE_INIT(printk_limits[1], DEFAULT_RATELIMIT_INTERVAL, 100),
+	RATELIMIT_STATE_INIT(printk_limits[2], DEFAULT_RATELIMIT_INTERVAL, 100),
+	RATELIMIT_STATE_INIT(printk_limits[3], DEFAULT_RATELIMIT_INTERVAL, 100),
+	RATELIMIT_STATE_INIT(printk_limits[4], DEFAULT_RATELIMIT_INTERVAL, 100),
+	RATELIMIT_STATE_INIT(printk_limits[5], DEFAULT_RATELIMIT_INTERVAL, 100),
+	RATELIMIT_STATE_INIT(printk_limits[6], DEFAULT_RATELIMIT_INTERVAL, 100),
+	RATELIMIT_STATE_INIT(printk_limits[7], DEFAULT_RATELIMIT_INTERVAL, 100),
+};
+
 void btrfs_printk(const struct btrfs_fs_info *fs_info, const char *fmt, ...)
 {
 	struct super_block *sb = fs_info->sb;
@@ -192,6 +207,7 @@ void btrfs_printk(const struct btrfs_fs_info *fs_info, const char *fmt, ...)
 	va_list args;
 	const char *type = logtypes[4];
 	int kern_level;
+	struct ratelimit_state *ratelimit;
 
 	va_start(args, fmt);
 
@@ -202,13 +218,18 @@ void btrfs_printk(const struct btrfs_fs_info *fs_info, const char *fmt, ...)
 		lvl[size] = '\0';
 		fmt += size;
 		type = logtypes[kern_level - '0'];
-	} else
+		ratelimit = &printk_limits[kern_level - '0'];
+	} else {
 		*lvl = '\0';
+		/* Default to debug output */
+		ratelimit = &printk_limits[7];
+	}
 
 	vaf.fmt = fmt;
 	vaf.va = &args;
 
-	printk("%sBTRFS %s (device %s): %pV\n", lvl, type, sb->s_id, &vaf);
+	if (__ratelimit(ratelimit))
+		printk("%sBTRFS %s (device %s): %pV\n", lvl, type, sb->s_id, &vaf);
 
 	va_end(args);
 }
@@ -229,26 +250,28 @@ void btrfs_printk(const struct btrfs_fs_info *fs_info, const char *fmt, ...)
  */
 __cold
 void __btrfs_abort_transaction(struct btrfs_trans_handle *trans,
-			       struct btrfs_root *root, const char *function,
+			       const char *function,
 			       unsigned int line, int errno)
 {
+	struct btrfs_fs_info *fs_info = trans->fs_info;
+
 	trans->aborted = errno;
 	/* Nothing used. The other threads that have joined this
 	 * transaction may be able to continue. */
-	if (!trans->blocks_used && list_empty(&trans->new_bgs)) {
+	if (!trans->dirty && list_empty(&trans->new_bgs)) {
 		const char *errstr;
 
 		errstr = btrfs_decode_error(errno);
-		btrfs_warn(root->fs_info,
+		btrfs_warn(fs_info,
 		           "%s:%d: Aborting unused transaction(%s).",
 		           function, line, errstr);
 		return;
 	}
 	ACCESS_ONCE(trans->transaction->aborted) = errno;
 	/* Wake up anybody who may be waiting on this transaction */
-	wake_up(&root->fs_info->transaction_wait);
-	wake_up(&root->fs_info->transaction_blocked_wait);
-	__btrfs_handle_fs_error(root->fs_info, function, line, errno, NULL);
+	wake_up(&fs_info->transaction_wait);
+	wake_up(&fs_info->transaction_blocked_wait);
+	__btrfs_handle_fs_error(fs_info, function, line, errno, NULL);
 }
 /*
  * __btrfs_panic decodes unexpected, fatal errors from the caller,
@@ -432,27 +455,30 @@ int btrfs_parse_options(struct btrfs_root *root, char *options,
 			 */
 			break;
 		case Opt_nodatasum:
-			btrfs_set_and_info(root, NODATASUM,
+			btrfs_set_and_info(info, NODATASUM,
 					   "setting nodatasum");
 			break;
 		case Opt_datasum:
-			if (btrfs_test_opt(root, NODATASUM)) {
-				if (btrfs_test_opt(root, NODATACOW))
-					btrfs_info(root->fs_info, "setting datasum, datacow enabled");
+			if (btrfs_test_opt(info, NODATASUM)) {
+				if (btrfs_test_opt(info, NODATACOW))
+					btrfs_info(root->fs_info,
+						   "setting datasum, datacow enabled");
 				else
-					btrfs_info(root->fs_info, "setting datasum");
+					btrfs_info(root->fs_info,
+						   "setting datasum");
 			}
 			btrfs_clear_opt(info->mount_opt, NODATACOW);
 			btrfs_clear_opt(info->mount_opt, NODATASUM);
 			break;
 		case Opt_nodatacow:
-			if (!btrfs_test_opt(root, NODATACOW)) {
-				if (!btrfs_test_opt(root, COMPRESS) ||
-				    !btrfs_test_opt(root, FORCE_COMPRESS)) {
+			if (!btrfs_test_opt(info, NODATACOW)) {
+				if (!btrfs_test_opt(info, COMPRESS) ||
+				    !btrfs_test_opt(info, FORCE_COMPRESS)) {
 					btrfs_info(root->fs_info,
 						   "setting nodatacow, compression disabled");
 				} else {
-					btrfs_info(root->fs_info, "setting nodatacow");
+					btrfs_info(root->fs_info,
+						   "setting nodatacow");
 				}
 			}
 			btrfs_clear_opt(info->mount_opt, COMPRESS);
@@ -461,7 +487,7 @@ int btrfs_parse_options(struct btrfs_root *root, char *options,
 			btrfs_set_opt(info->mount_opt, NODATASUM);
 			break;
 		case Opt_datacow:
-			btrfs_clear_and_info(root, NODATACOW,
+			btrfs_clear_and_info(info, NODATACOW,
 					     "setting datacow");
 			break;
 		case Opt_compress_force:
@@ -470,10 +496,11 @@ int btrfs_parse_options(struct btrfs_root *root, char *options,
 			/* Fallthrough */
 		case Opt_compress:
 		case Opt_compress_type:
-			saved_compress_type = btrfs_test_opt(root, COMPRESS) ?
+			saved_compress_type = btrfs_test_opt(info,
+							     COMPRESS) ?
 				info->compress_type : BTRFS_COMPRESS_NONE;
 			saved_compress_force =
-				btrfs_test_opt(root, FORCE_COMPRESS);
+				btrfs_test_opt(info, FORCE_COMPRESS);
 			if (token == Opt_compress ||
 			    token == Opt_compress_force ||
 			    strcmp(args[0].from, "zlib") == 0) {
@@ -513,10 +540,10 @@ int btrfs_parse_options(struct btrfs_root *root, char *options,
 				 */
 				btrfs_clear_opt(info->mount_opt, FORCE_COMPRESS);
 			}
-			if ((btrfs_test_opt(root, COMPRESS) &&
+			if ((btrfs_test_opt(info, COMPRESS) &&
 			     (info->compress_type != saved_compress_type ||
 			      compress_force != saved_compress_force)) ||
-			    (!btrfs_test_opt(root, COMPRESS) &&
+			    (!btrfs_test_opt(info, COMPRESS) &&
 			     no_compress == 1)) {
 				btrfs_info(root->fs_info,
 					   "%s %s compression",
@@ -526,25 +553,25 @@ int btrfs_parse_options(struct btrfs_root *root, char *options,
 			compress_force = false;
 			break;
 		case Opt_ssd:
-			btrfs_set_and_info(root, SSD,
+			btrfs_set_and_info(info, SSD,
 					   "use ssd allocation scheme");
 			break;
 		case Opt_ssd_spread:
-			btrfs_set_and_info(root, SSD_SPREAD,
+			btrfs_set_and_info(info, SSD_SPREAD,
 					   "use spread ssd allocation scheme");
 			btrfs_set_opt(info->mount_opt, SSD);
 			break;
 		case Opt_nossd:
-			btrfs_set_and_info(root, NOSSD,
+			btrfs_set_and_info(info, NOSSD,
 					     "not using ssd allocation scheme");
 			btrfs_clear_opt(info->mount_opt, SSD);
 			break;
 		case Opt_barrier:
-			btrfs_clear_and_info(root, NOBARRIER,
+			btrfs_clear_and_info(info, NOBARRIER,
 					     "turning on barriers");
 			break;
 		case Opt_nobarrier:
-			btrfs_set_and_info(root, NOBARRIER,
+			btrfs_set_and_info(info, NOBARRIER,
 					   "turning off barriers");
 			break;
 		case Opt_thread_pool:
@@ -583,8 +610,9 @@ int btrfs_parse_options(struct btrfs_root *root, char *options,
 				info->alloc_start = memparse(num, NULL);
 				mutex_unlock(&info->chunk_mutex);
 				kfree(num);
-				btrfs_info(root->fs_info, "allocations start at %llu",
-					info->alloc_start);
+				btrfs_info(root->fs_info,
+					   "allocations start at %llu",
+					   info->alloc_start);
 			} else {
 				ret = -ENOMEM;
 				goto out;
@@ -604,24 +632,24 @@ int btrfs_parse_options(struct btrfs_root *root, char *options,
 			root->fs_info->sb->s_flags &= ~MS_POSIXACL;
 			break;
 		case Opt_notreelog:
-			btrfs_set_and_info(root, NOTREELOG,
+			btrfs_set_and_info(info, NOTREELOG,
 					   "disabling tree log");
 			break;
 		case Opt_treelog:
-			btrfs_clear_and_info(root, NOTREELOG,
+			btrfs_clear_and_info(info, NOTREELOG,
 					     "enabling tree log");
 			break;
 		case Opt_norecovery:
 		case Opt_nologreplay:
-			btrfs_set_and_info(root, NOLOGREPLAY,
+			btrfs_set_and_info(info, NOLOGREPLAY,
 					   "disabling log replay at mount time");
 			break;
 		case Opt_flushoncommit:
-			btrfs_set_and_info(root, FLUSHONCOMMIT,
+			btrfs_set_and_info(info, FLUSHONCOMMIT,
 					   "turning on flush-on-commit");
 			break;
 		case Opt_noflushoncommit:
-			btrfs_clear_and_info(root, FLUSHONCOMMIT,
+			btrfs_clear_and_info(info, FLUSHONCOMMIT,
 					     "turning off flush-on-commit");
 			break;
 		case Opt_ratio:
@@ -638,11 +666,11 @@ int btrfs_parse_options(struct btrfs_root *root, char *options,
 			}
 			break;
 		case Opt_discard:
-			btrfs_set_and_info(root, DISCARD,
+			btrfs_set_and_info(info, DISCARD,
 					   "turning on discard");
 			break;
 		case Opt_nodiscard:
-			btrfs_clear_and_info(root, DISCARD,
+			btrfs_clear_and_info(info, DISCARD,
 					     "turning off discard");
 			break;
 		case Opt_space_cache:
@@ -651,12 +679,13 @@ int btrfs_parse_options(struct btrfs_root *root, char *options,
 			    strcmp(args[0].from, "v1") == 0) {
 				btrfs_clear_opt(root->fs_info->mount_opt,
 						FREE_SPACE_TREE);
-				btrfs_set_and_info(root, SPACE_CACHE,
+				btrfs_set_and_info(info, SPACE_CACHE,
 						   "enabling disk space caching");
 			} else if (strcmp(args[0].from, "v2") == 0) {
 				btrfs_clear_opt(root->fs_info->mount_opt,
 						SPACE_CACHE);
-				btrfs_set_and_info(root, FREE_SPACE_TREE,
+				btrfs_set_and_info(info,
+						   FREE_SPACE_TREE,
 						   "enabling free space tree");
 			} else {
 				ret = -EINVAL;
@@ -667,12 +696,14 @@ int btrfs_parse_options(struct btrfs_root *root, char *options,
 			btrfs_set_opt(info->mount_opt, RESCAN_UUID_TREE);
 			break;
 		case Opt_no_space_cache:
-			if (btrfs_test_opt(root, SPACE_CACHE)) {
-				btrfs_clear_and_info(root, SPACE_CACHE,
+			if (btrfs_test_opt(info, SPACE_CACHE)) {
+				btrfs_clear_and_info(info,
+						     SPACE_CACHE,
 						     "disabling disk space caching");
 			}
-			if (btrfs_test_opt(root, FREE_SPACE_TREE)) {
-				btrfs_clear_and_info(root, FREE_SPACE_TREE,
+			if (btrfs_test_opt(info, FREE_SPACE_TREE)) {
+				btrfs_clear_and_info(info,
+						     FREE_SPACE_TREE,
 						     "disabling free space tree");
 			}
 			break;
@@ -685,7 +716,7 @@ int btrfs_parse_options(struct btrfs_root *root, char *options,
 					     "disabling inode map caching");
 			break;
 		case Opt_clear_cache:
-			btrfs_set_and_info(root, CLEAR_CACHE,
+			btrfs_set_and_info(info, CLEAR_CACHE,
 					   "force clearing of disk cache");
 			break;
 		case Opt_user_subvol_rm_allowed:
@@ -698,11 +729,11 @@ int btrfs_parse_options(struct btrfs_root *root, char *options,
 			btrfs_clear_opt(info->mount_opt, ENOSPC_DEBUG);
 			break;
 		case Opt_defrag:
-			btrfs_set_and_info(root, AUTO_DEFRAG,
+			btrfs_set_and_info(info, AUTO_DEFRAG,
 					   "enabling auto defrag");
 			break;
 		case Opt_nodefrag:
-			btrfs_clear_and_info(root, AUTO_DEFRAG,
+			btrfs_clear_and_info(info, AUTO_DEFRAG,
 					     "disabling auto defrag");
 			break;
 		case Opt_recovery:
@@ -734,8 +765,9 @@ int btrfs_parse_options(struct btrfs_root *root, char *options,
 				goto out;
 			} else if (intarg >= 0) {
 				info->check_integrity_print_mask = intarg;
-				btrfs_info(root->fs_info, "check_integrity_print_mask 0x%x",
-				       info->check_integrity_print_mask);
+				btrfs_info(root->fs_info,
+					   "check_integrity_print_mask 0x%x",
+					   info->check_integrity_print_mask);
 			} else {
 				ret = -EINVAL;
 				goto out;
@@ -766,19 +798,22 @@ int btrfs_parse_options(struct btrfs_root *root, char *options,
 			intarg = 0;
 			ret = match_int(&args[0], &intarg);
 			if (ret < 0) {
-				btrfs_err(root->fs_info, "invalid commit interval");
+				btrfs_err(root->fs_info,
+					  "invalid commit interval");
 				ret = -EINVAL;
 				goto out;
 			}
 			if (intarg > 0) {
 				if (intarg > 300) {
-					btrfs_warn(root->fs_info, "excessive commit interval %d",
-							intarg);
+					btrfs_warn(root->fs_info,
+						"excessive commit interval %d",
+						intarg);
 				}
 				info->commit_interval = intarg;
 			} else {
-				btrfs_info(root->fs_info, "using default commit interval %ds",
-				    BTRFS_DEFAULT_COMMIT_INTERVAL);
+				btrfs_info(root->fs_info,
+					   "using default commit interval %ds",
+					   BTRFS_DEFAULT_COMMIT_INTERVAL);
 				info->commit_interval = BTRFS_DEFAULT_COMMIT_INTERVAL;
 			}
 			break;
@@ -799,7 +834,8 @@ int btrfs_parse_options(struct btrfs_root *root, char *options,
 			break;
 #endif
 		case Opt_err:
-			btrfs_info(root->fs_info, "unrecognized mount option '%s'", p);
+			btrfs_info(root->fs_info,
+				   "unrecognized mount option '%s'", p);
 			ret = -EINVAL;
 			goto out;
 		default:
@@ -810,22 +846,22 @@ check:
 	/*
 	 * Extra check for current option against current flag
 	 */
-	if (btrfs_test_opt(root, NOLOGREPLAY) && !(new_flags & MS_RDONLY)) {
+	if (btrfs_test_opt(info, NOLOGREPLAY) && !(new_flags & MS_RDONLY)) {
 		btrfs_err(root->fs_info,
 			  "nologreplay must be used with ro mount option");
 		ret = -EINVAL;
 	}
 out:
 	if (btrfs_fs_compat_ro(root->fs_info, FREE_SPACE_TREE) &&
-	    !btrfs_test_opt(root, FREE_SPACE_TREE) &&
-	    !btrfs_test_opt(root, CLEAR_CACHE)) {
+	    !btrfs_test_opt(info, FREE_SPACE_TREE) &&
+	    !btrfs_test_opt(info, CLEAR_CACHE)) {
 		btrfs_err(root->fs_info, "cannot disable free space tree");
 		ret = -EINVAL;
 
 	}
-	if (!ret && btrfs_test_opt(root, SPACE_CACHE))
+	if (!ret && btrfs_test_opt(info, SPACE_CACHE))
 		btrfs_info(root->fs_info, "disk space caching is enabled");
-	if (!ret && btrfs_test_opt(root, FREE_SPACE_TREE))
+	if (!ret && btrfs_test_opt(info, FREE_SPACE_TREE))
 		btrfs_info(root->fs_info, "using free space tree");
 	kfree(orig);
 	return ret;
@@ -888,9 +924,7 @@ static int btrfs_parse_early_options(const char *options, fmode_t flags,
 			}
 			break;
 		case Opt_subvolrootid:
-			printk(KERN_WARNING
-				"BTRFS: 'subvolrootid' mount option is deprecated and has "
-				"no effect\n");
+			pr_warn("BTRFS: 'subvolrootid' mount option is deprecated and has no effect\n");
 			break;
 		case Opt_device:
 			device_name = match_strdup(&args[0]);
@@ -1114,7 +1148,7 @@ static int btrfs_fill_super(struct super_block *sb,
 	sb->s_iflags |= SB_I_CGROUPWB;
 	err = open_ctree(sb, fs_devices, (char *)data);
 	if (err) {
-		printk(KERN_ERR "BTRFS: open_ctree failed\n");
+		btrfs_err(fs_info, "open_ctree failed");
 		return err;
 	}
 
@@ -1149,7 +1183,7 @@ int btrfs_sync_fs(struct super_block *sb, int wait)
 	struct btrfs_fs_info *fs_info = btrfs_sb(sb);
 	struct btrfs_root *root = fs_info->tree_root;
 
-	trace_btrfs_sync_fs(wait);
+	trace_btrfs_sync_fs(fs_info, wait);
 
 	if (!wait) {
 		filemap_flush(fs_info->btree_inode->i_mapping);
@@ -1192,13 +1226,13 @@ static int btrfs_show_options(struct seq_file *seq, struct dentry *dentry)
 	struct btrfs_root *root = info->tree_root;
 	char *compress_type;
 
-	if (btrfs_test_opt(root, DEGRADED))
+	if (btrfs_test_opt(info, DEGRADED))
 		seq_puts(seq, ",degraded");
-	if (btrfs_test_opt(root, NODATASUM))
+	if (btrfs_test_opt(info, NODATASUM))
 		seq_puts(seq, ",nodatasum");
-	if (btrfs_test_opt(root, NODATACOW))
+	if (btrfs_test_opt(info, NODATACOW))
 		seq_puts(seq, ",nodatacow");
-	if (btrfs_test_opt(root, NOBARRIER))
+	if (btrfs_test_opt(info, NOBARRIER))
 		seq_puts(seq, ",nobarrier");
 	if (info->max_inline != BTRFS_DEFAULT_MAX_INLINE)
 		seq_printf(seq, ",max_inline=%llu", info->max_inline);
@@ -1207,56 +1241,56 @@ static int btrfs_show_options(struct seq_file *seq, struct dentry *dentry)
 	if (info->thread_pool_size !=  min_t(unsigned long,
 					     num_online_cpus() + 2, 8))
 		seq_printf(seq, ",thread_pool=%d", info->thread_pool_size);
-	if (btrfs_test_opt(root, COMPRESS)) {
+	if (btrfs_test_opt(info, COMPRESS)) {
 		if (info->compress_type == BTRFS_COMPRESS_ZLIB)
 			compress_type = "zlib";
 		else
 			compress_type = "lzo";
-		if (btrfs_test_opt(root, FORCE_COMPRESS))
+		if (btrfs_test_opt(info, FORCE_COMPRESS))
 			seq_printf(seq, ",compress-force=%s", compress_type);
 		else
 			seq_printf(seq, ",compress=%s", compress_type);
 	}
-	if (btrfs_test_opt(root, NOSSD))
+	if (btrfs_test_opt(info, NOSSD))
 		seq_puts(seq, ",nossd");
-	if (btrfs_test_opt(root, SSD_SPREAD))
+	if (btrfs_test_opt(info, SSD_SPREAD))
 		seq_puts(seq, ",ssd_spread");
-	else if (btrfs_test_opt(root, SSD))
+	else if (btrfs_test_opt(info, SSD))
 		seq_puts(seq, ",ssd");
-	if (btrfs_test_opt(root, NOTREELOG))
+	if (btrfs_test_opt(info, NOTREELOG))
 		seq_puts(seq, ",notreelog");
-	if (btrfs_test_opt(root, NOLOGREPLAY))
+	if (btrfs_test_opt(info, NOLOGREPLAY))
 		seq_puts(seq, ",nologreplay");
-	if (btrfs_test_opt(root, FLUSHONCOMMIT))
+	if (btrfs_test_opt(info, FLUSHONCOMMIT))
 		seq_puts(seq, ",flushoncommit");
-	if (btrfs_test_opt(root, DISCARD))
+	if (btrfs_test_opt(info, DISCARD))
 		seq_puts(seq, ",discard");
 	if (!(root->fs_info->sb->s_flags & MS_POSIXACL))
 		seq_puts(seq, ",noacl");
-	if (btrfs_test_opt(root, SPACE_CACHE))
+	if (btrfs_test_opt(info, SPACE_CACHE))
 		seq_puts(seq, ",space_cache");
-	else if (btrfs_test_opt(root, FREE_SPACE_TREE))
+	else if (btrfs_test_opt(info, FREE_SPACE_TREE))
 		seq_puts(seq, ",space_cache=v2");
 	else
 		seq_puts(seq, ",nospace_cache");
-	if (btrfs_test_opt(root, RESCAN_UUID_TREE))
+	if (btrfs_test_opt(info, RESCAN_UUID_TREE))
 		seq_puts(seq, ",rescan_uuid_tree");
-	if (btrfs_test_opt(root, CLEAR_CACHE))
+	if (btrfs_test_opt(info, CLEAR_CACHE))
 		seq_puts(seq, ",clear_cache");
-	if (btrfs_test_opt(root, USER_SUBVOL_RM_ALLOWED))
+	if (btrfs_test_opt(info, USER_SUBVOL_RM_ALLOWED))
 		seq_puts(seq, ",user_subvol_rm_allowed");
-	if (btrfs_test_opt(root, ENOSPC_DEBUG))
+	if (btrfs_test_opt(info, ENOSPC_DEBUG))
 		seq_puts(seq, ",enospc_debug");
-	if (btrfs_test_opt(root, AUTO_DEFRAG))
+	if (btrfs_test_opt(info, AUTO_DEFRAG))
 		seq_puts(seq, ",autodefrag");
-	if (btrfs_test_opt(root, INODE_MAP_CACHE))
+	if (btrfs_test_opt(info, INODE_MAP_CACHE))
 		seq_puts(seq, ",inode_cache");
-	if (btrfs_test_opt(root, SKIP_BALANCE))
+	if (btrfs_test_opt(info, SKIP_BALANCE))
 		seq_puts(seq, ",skip_balance");
 #ifdef CONFIG_BTRFS_FS_CHECK_INTEGRITY
-	if (btrfs_test_opt(root, CHECK_INTEGRITY_INCLUDING_EXTENT_DATA))
+	if (btrfs_test_opt(info, CHECK_INTEGRITY_INCLUDING_EXTENT_DATA))
 		seq_puts(seq, ",check_int_data");
-	else if (btrfs_test_opt(root, CHECK_INTEGRITY))
+	else if (btrfs_test_opt(info, CHECK_INTEGRITY))
 		seq_puts(seq, ",check_int");
 	if (info->check_integrity_print_mask)
 		seq_printf(seq, ",check_int_print_mask=%d",
@@ -1265,14 +1299,14 @@ static int btrfs_show_options(struct seq_file *seq, struct dentry *dentry)
 	if (info->metadata_ratio)
 		seq_printf(seq, ",metadata_ratio=%d",
 				info->metadata_ratio);
-	if (btrfs_test_opt(root, PANIC_ON_FATAL_ERROR))
+	if (btrfs_test_opt(info, PANIC_ON_FATAL_ERROR))
 		seq_puts(seq, ",fatal_errors=panic");
 	if (info->commit_interval != BTRFS_DEFAULT_COMMIT_INTERVAL)
 		seq_printf(seq, ",commit=%d", info->commit_interval);
 #ifdef CONFIG_BTRFS_DEBUG
-	if (btrfs_test_opt(root, FRAGMENT_DATA))
+	if (btrfs_test_opt(info, FRAGMENT_DATA))
 		seq_puts(seq, ",fragment=data");
-	if (btrfs_test_opt(root, FRAGMENT_METADATA))
+	if (btrfs_test_opt(info, FRAGMENT_METADATA))
 		seq_puts(seq, ",fragment=metadata");
 #endif
 	seq_printf(seq, ",subvolid=%llu",
@@ -1412,12 +1446,13 @@ static struct dentry *mount_subvol(const char *subvol_name, u64 subvol_objectid,
 
 	if (!IS_ERR(root)) {
 		struct super_block *s = root->d_sb;
+		struct btrfs_fs_info *fs_info = btrfs_sb(s);
 		struct inode *root_inode = d_inode(root);
 		u64 root_objectid = BTRFS_I(root_inode)->root->root_key.objectid;
 
 		ret = 0;
 		if (!is_subvolume_inode(root_inode)) {
-			pr_err("BTRFS: '%s' is not a valid subvolume\n",
+			btrfs_err(fs_info, "'%s' is not a valid subvolume",
 			       subvol_name);
 			ret = -EINVAL;
 		}
@@ -1427,8 +1462,9 @@ static struct dentry *mount_subvol(const char *subvol_name, u64 subvol_objectid,
 			 * subvolume which was passed by ID is renamed and
 			 * another subvolume is renamed over the old location.
 			 */
-			pr_err("BTRFS: subvol '%s' does not match subvolid %llu\n",
-			       subvol_name, subvol_objectid);
+			btrfs_err(fs_info,
+				  "subvol '%s' does not match subvolid %llu",
+				  subvol_name, subvol_objectid);
 			ret = -EINVAL;
 		}
 		if (ret) {
@@ -1802,11 +1838,15 @@ static int btrfs_remount(struct super_block *sb, int *flags, char *data)
 			btrfs_info(fs_info, "creating UUID tree");
 			ret = btrfs_create_uuid_tree(fs_info);
 			if (ret) {
-				btrfs_warn(fs_info, "failed to create the UUID tree %d", ret);
+				btrfs_warn(fs_info,
+					   "failed to create the UUID tree %d",
+					   ret);
 				goto restore;
 			}
 		}
 		sb->s_flags &= ~MS_RDONLY;
+
+		set_bit(BTRFS_FS_OPEN, &fs_info->flags);
 	}
 out:
 	wake_up_process(fs_info->transaction_kthread);
@@ -2028,9 +2068,6 @@ static int btrfs_calc_avail_data_space(struct btrfs_root *root, u64 *free_bytes)
  * chunk).
  *
  * If metadata is exhausted, f_bavail will be 0.
- *
- * FIXME: not accurate for mixed block groups, total and free/used are ok,
- * available appears slightly larger.
  */
 static int btrfs_statfs(struct dentry *dentry, struct kstatfs *buf)
 {
@@ -2214,6 +2251,13 @@ static int btrfs_freeze(struct super_block *sb)
 	struct btrfs_trans_handle *trans;
 	struct btrfs_root *root = btrfs_sb(sb)->tree_root;
 
+	root->fs_info->fs_frozen = 1;
+	/*
+	 * We don't need a barrier here, we'll wait for any transaction that
+	 * could be in progress on other threads (and do delayed iputs that
+	 * we want to avoid on a frozen filesystem), or do the commit
+	 * ourselves.
+	 */
 	trans = btrfs_attach_transaction_barrier(root);
 	if (IS_ERR(trans)) {
 		/* no transaction, don't bother */
@@ -2224,6 +2268,14 @@ static int btrfs_freeze(struct super_block *sb)
 	return btrfs_commit_transaction(trans, root);
 }
 
+static int btrfs_unfreeze(struct super_block *sb)
+{
+	struct btrfs_root *root = btrfs_sb(sb)->tree_root;
+
+	root->fs_info->fs_frozen = 0;
+	return 0;
+}
+
 static int btrfs_show_devname(struct seq_file *m, struct dentry *root)
 {
 	struct btrfs_fs_info *fs_info = btrfs_sb(root->d_sb);
@@ -2272,6 +2324,7 @@ static const struct super_operations btrfs_super_ops = {
 	.statfs		= btrfs_statfs,
 	.remount_fs	= btrfs_remount,
 	.freeze_fs	= btrfs_freeze,
+	.unfreeze_fs	= btrfs_unfreeze,
 };
 
 static const struct file_operations btrfs_ctl_fops = {
@@ -2303,7 +2356,7 @@ static void btrfs_interface_exit(void)
 
 static void btrfs_print_mod_info(void)
 {
-	printk(KERN_INFO "Btrfs loaded"
+	pr_info("Btrfs loaded, crc32c=%s"
 #ifdef CONFIG_BTRFS_DEBUG
 			", debug=on"
 #endif
@@ -2313,36 +2366,8 @@ static void btrfs_print_mod_info(void)
 #ifdef CONFIG_BTRFS_FS_CHECK_INTEGRITY
 			", integrity-checker=on"
 #endif
-			"\n");
-}
-
-static int btrfs_run_sanity_tests(void)
-{
-	int ret;
-
-	ret = btrfs_init_test_fs();
-	if (ret)
-		return ret;
-
-	ret = btrfs_test_free_space_cache();
-	if (ret)
-		goto out;
-	ret = btrfs_test_extent_buffer_operations();
-	if (ret)
-		goto out;
-	ret = btrfs_test_extent_io();
-	if (ret)
-		goto out;
-	ret = btrfs_test_inodes();
-	if (ret)
-		goto out;
-	ret = btrfs_test_qgroups();
-	if (ret)
-		goto out;
-	ret = btrfs_test_free_space_tree();
-out:
-	btrfs_destroy_test_fs();
-	return ret;
+			"\n",
+			btrfs_crc32c_impl());
 }
 
 static int __init init_btrfs_fs(void)
diff --git a/fs/btrfs/sysfs.c b/fs/btrfs/sysfs.c
index 4879656bda3c..1f157fba8940 100644
--- a/fs/btrfs/sysfs.c
+++ b/fs/btrfs/sysfs.c
@@ -77,7 +77,7 @@ static int can_modify_feature(struct btrfs_feature_attr *fa)
 		clear = BTRFS_FEATURE_INCOMPAT_SAFE_CLEAR;
 		break;
 	default:
-		printk(KERN_WARNING "btrfs: sysfs: unknown feature set %d\n",
+		pr_warn("btrfs: sysfs: unknown feature set %d\n",
 				fa->feature_set);
 		return 0;
 	}
@@ -326,6 +326,7 @@ SPACE_INFO_ATTR(bytes_used);
 SPACE_INFO_ATTR(bytes_pinned);
 SPACE_INFO_ATTR(bytes_reserved);
 SPACE_INFO_ATTR(bytes_may_use);
+SPACE_INFO_ATTR(bytes_readonly);
 SPACE_INFO_ATTR(disk_used);
 SPACE_INFO_ATTR(disk_total);
 BTRFS_ATTR(total_bytes_pinned, btrfs_space_info_show_total_bytes_pinned);
@@ -337,6 +338,7 @@ static struct attribute *space_info_attrs[] = {
 	BTRFS_ATTR_PTR(bytes_pinned),
 	BTRFS_ATTR_PTR(bytes_reserved),
 	BTRFS_ATTR_PTR(bytes_may_use),
+	BTRFS_ATTR_PTR(bytes_readonly),
 	BTRFS_ATTR_PTR(disk_used),
 	BTRFS_ATTR_PTR(disk_total),
 	BTRFS_ATTR_PTR(total_bytes_pinned),
@@ -428,7 +430,8 @@ static ssize_t btrfs_sectorsize_show(struct kobject *kobj,
 {
 	struct btrfs_fs_info *fs_info = to_fs_info(kobj);
 
-	return snprintf(buf, PAGE_SIZE, "%u\n", fs_info->super_copy->sectorsize);
+	return snprintf(buf, PAGE_SIZE, "%u\n",
+			fs_info->super_copy->sectorsize);
 }
 
 BTRFS_ATTR(sectorsize, btrfs_sectorsize_show);
@@ -438,7 +441,8 @@ static ssize_t btrfs_clone_alignment_show(struct kobject *kobj,
 {
 	struct btrfs_fs_info *fs_info = to_fs_info(kobj);
 
-	return snprintf(buf, PAGE_SIZE, "%u\n", fs_info->super_copy->sectorsize);
+	return snprintf(buf, PAGE_SIZE, "%u\n",
+			fs_info->super_copy->sectorsize);
 }
 
 BTRFS_ATTR(clone_alignment, btrfs_clone_alignment_show);
@@ -834,9 +838,18 @@ static int btrfs_init_debugfs(void)
 	if (!btrfs_debugfs_root_dentry)
 		return -ENOMEM;
 
-	debugfs_create_u64("test", S_IRUGO | S_IWUGO, btrfs_debugfs_root_dentry,
+	/*
+	 * Example code, how to export data through debugfs.
+	 *
+	 * file:        /sys/kernel/debug/btrfs/test
+	 * contents of: btrfs_debugfs_test
+	 */
+#ifdef CONFIG_BTRFS_DEBUG
+	debugfs_create_u64("test", S_IRUGO | S_IWUSR, btrfs_debugfs_root_dentry,
 			&btrfs_debugfs_test);
 #endif
+
+#endif
 	return 0;
 }
 
diff --git a/fs/btrfs/tests/btrfs-tests.c b/fs/btrfs/tests/btrfs-tests.c
index f54bf450bad3..bf62ad919a95 100644
--- a/fs/btrfs/tests/btrfs-tests.c
+++ b/fs/btrfs/tests/btrfs-tests.c
@@ -54,7 +54,7 @@ struct inode *btrfs_new_test_inode(void)
 	return new_inode(test_mnt->mnt_sb);
 }
 
-int btrfs_init_test_fs(void)
+static int btrfs_init_test_fs(void)
 {
 	int ret;
 
@@ -68,12 +68,12 @@ int btrfs_init_test_fs(void)
 	if (IS_ERR(test_mnt)) {
 		printk(KERN_ERR "btrfs: cannot mount test file system\n");
 		unregister_filesystem(&test_type);
-		return ret;
+		return PTR_ERR(test_mnt);
 	}
 	return 0;
 }
 
-void btrfs_destroy_test_fs(void)
+static void btrfs_destroy_test_fs(void)
 {
 	kern_unmount(test_mnt);
 	unregister_filesystem(&test_type);
@@ -128,14 +128,27 @@ struct btrfs_fs_info *btrfs_alloc_dummy_fs_info(void)
 	extent_io_tree_init(&fs_info->freed_extents[0], NULL);
 	extent_io_tree_init(&fs_info->freed_extents[1], NULL);
 	fs_info->pinned_extents = &fs_info->freed_extents[0];
+	set_bit(BTRFS_FS_STATE_DUMMY_FS_INFO, &fs_info->fs_state);
+
+	test_mnt->mnt_sb->s_fs_info = fs_info;
+
 	return fs_info;
 }
 
-static void btrfs_free_dummy_fs_info(struct btrfs_fs_info *fs_info)
+void btrfs_free_dummy_fs_info(struct btrfs_fs_info *fs_info)
 {
 	struct radix_tree_iter iter;
 	void **slot;
 
+	if (!fs_info)
+		return;
+
+	if (WARN_ON(!test_bit(BTRFS_FS_STATE_DUMMY_FS_INFO,
+			      &fs_info->fs_state)))
+		return;
+
+	test_mnt->mnt_sb->s_fs_info = NULL;
+
 	spin_lock(&fs_info->buffer_lock);
 	radix_tree_for_each_slot(slot, &fs_info->buffer_radix, &iter, 0) {
 		struct extent_buffer *eb;
@@ -167,15 +180,16 @@ void btrfs_free_dummy_root(struct btrfs_root *root)
 {
 	if (!root)
 		return;
+	/* Will be freed by btrfs_free_fs_roots */
+	if (WARN_ON(test_bit(BTRFS_ROOT_IN_RADIX, &root->state)))
+		return;
 	if (root->node)
 		free_extent_buffer(root->node);
-	if (root->fs_info)
-		btrfs_free_dummy_fs_info(root->fs_info);
 	kfree(root);
 }
 
 struct btrfs_block_group_cache *
-btrfs_alloc_dummy_block_group(unsigned long length)
+btrfs_alloc_dummy_block_group(unsigned long length, u32 sectorsize)
 {
 	struct btrfs_block_group_cache *cache;
 
@@ -192,8 +206,8 @@ btrfs_alloc_dummy_block_group(unsigned long length)
 	cache->key.objectid = 0;
 	cache->key.offset = length;
 	cache->key.type = BTRFS_BLOCK_GROUP_ITEM_KEY;
-	cache->sectorsize = 4096;
-	cache->full_stripe_len = 4096;
+	cache->sectorsize = sectorsize;
+	cache->full_stripe_len = sectorsize;
 
 	INIT_LIST_HEAD(&cache->list);
 	INIT_LIST_HEAD(&cache->cluster_list);
@@ -220,3 +234,46 @@ void btrfs_init_dummy_trans(struct btrfs_trans_handle *trans)
 	INIT_LIST_HEAD(&trans->qgroup_ref_list);
 	trans->type = __TRANS_DUMMY;
 }
+
+int btrfs_run_sanity_tests(void)
+{
+	int ret, i;
+	u32 sectorsize, nodesize;
+	u32 test_sectorsize[] = {
+		PAGE_SIZE,
+	};
+	ret = btrfs_init_test_fs();
+	if (ret)
+		return ret;
+	for (i = 0; i < ARRAY_SIZE(test_sectorsize); i++) {
+		sectorsize = test_sectorsize[i];
+		for (nodesize = sectorsize;
+		     nodesize <= BTRFS_MAX_METADATA_BLOCKSIZE;
+		     nodesize <<= 1) {
+			pr_info("BTRFS: selftest: sectorsize: %u  nodesize: %u\n",
+				sectorsize, nodesize);
+			ret = btrfs_test_free_space_cache(sectorsize, nodesize);
+			if (ret)
+				goto out;
+			ret = btrfs_test_extent_buffer_operations(sectorsize,
+				nodesize);
+			if (ret)
+				goto out;
+			ret = btrfs_test_extent_io(sectorsize, nodesize);
+			if (ret)
+				goto out;
+			ret = btrfs_test_inodes(sectorsize, nodesize);
+			if (ret)
+				goto out;
+			ret = btrfs_test_qgroups(sectorsize, nodesize);
+			if (ret)
+				goto out;
+			ret = btrfs_test_free_space_tree(sectorsize, nodesize);
+			if (ret)
+				goto out;
+		}
+	}
+out:
+	btrfs_destroy_test_fs();
+	return ret;
+}
diff --git a/fs/btrfs/tests/btrfs-tests.h b/fs/btrfs/tests/btrfs-tests.h
index 054b8c73c951..b17ffbe8f9f3 100644
--- a/fs/btrfs/tests/btrfs-tests.h
+++ b/fs/btrfs/tests/btrfs-tests.h
@@ -20,56 +20,29 @@
 #define __BTRFS_TESTS
 
 #ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
+int btrfs_run_sanity_tests(void);
 
 #define test_msg(fmt, ...) pr_info("BTRFS: selftest: " fmt, ##__VA_ARGS__)
 
 struct btrfs_root;
 struct btrfs_trans_handle;
 
-int btrfs_test_free_space_cache(void);
-int btrfs_test_extent_buffer_operations(void);
-int btrfs_test_extent_io(void);
-int btrfs_test_inodes(void);
-int btrfs_test_qgroups(void);
-int btrfs_test_free_space_tree(void);
-int btrfs_init_test_fs(void);
-void btrfs_destroy_test_fs(void);
+int btrfs_test_extent_buffer_operations(u32 sectorsize, u32 nodesize);
+int btrfs_test_free_space_cache(u32 sectorsize, u32 nodesize);
+int btrfs_test_extent_io(u32 sectorsize, u32 nodesize);
+int btrfs_test_inodes(u32 sectorsize, u32 nodesize);
+int btrfs_test_qgroups(u32 sectorsize, u32 nodesize);
+int btrfs_test_free_space_tree(u32 sectorsize, u32 nodesize);
 struct inode *btrfs_new_test_inode(void);
 struct btrfs_fs_info *btrfs_alloc_dummy_fs_info(void);
+void btrfs_free_dummy_fs_info(struct btrfs_fs_info *fs_info);
 void btrfs_free_dummy_root(struct btrfs_root *root);
 struct btrfs_block_group_cache *
-btrfs_alloc_dummy_block_group(unsigned long length);
+btrfs_alloc_dummy_block_group(unsigned long length, u32 sectorsize);
 void btrfs_free_dummy_block_group(struct btrfs_block_group_cache *cache);
 void btrfs_init_dummy_trans(struct btrfs_trans_handle *trans);
 #else
-static inline int btrfs_test_free_space_cache(void)
-{
-	return 0;
-}
-static inline int btrfs_test_extent_buffer_operations(void)
-{
-	return 0;
-}
-static inline int btrfs_init_test_fs(void)
-{
-	return 0;
-}
-static inline void btrfs_destroy_test_fs(void)
-{
-}
-static inline int btrfs_test_extent_io(void)
-{
-	return 0;
-}
-static inline int btrfs_test_inodes(void)
-{
-	return 0;
-}
-static inline int btrfs_test_qgroups(void)
-{
-	return 0;
-}
-static inline int btrfs_test_free_space_tree(void)
+static inline int btrfs_run_sanity_tests(void)
 {
 	return 0;
 }
diff --git a/fs/btrfs/tests/extent-buffer-tests.c b/fs/btrfs/tests/extent-buffer-tests.c
index f51963a8f929..199569174637 100644
--- a/fs/btrfs/tests/extent-buffer-tests.c
+++ b/fs/btrfs/tests/extent-buffer-tests.c
@@ -22,10 +22,11 @@
 #include "../extent_io.h"
 #include "../disk-io.h"
 
-static int test_btrfs_split_item(void)
+static int test_btrfs_split_item(u32 sectorsize, u32 nodesize)
 {
-	struct btrfs_path *path;
-	struct btrfs_root *root;
+	struct btrfs_fs_info *fs_info;
+	struct btrfs_path *path = NULL;
+	struct btrfs_root *root = NULL;
 	struct extent_buffer *eb;
 	struct btrfs_item *item;
 	char *value = "mary had a little lamb";
@@ -40,20 +41,28 @@ static int test_btrfs_split_item(void)
 
 	test_msg("Running btrfs_split_item tests\n");
 
-	root = btrfs_alloc_dummy_root();
+	fs_info = btrfs_alloc_dummy_fs_info();
+	if (!fs_info) {
+		test_msg("Could not allocate fs_info\n");
+		return -ENOMEM;
+	}
+
+	root = btrfs_alloc_dummy_root(fs_info, sectorsize, nodesize);
 	if (IS_ERR(root)) {
 		test_msg("Could not allocate root\n");
-		return PTR_ERR(root);
+		ret = PTR_ERR(root);
+		goto out;
 	}
 
 	path = btrfs_alloc_path();
 	if (!path) {
 		test_msg("Could not allocate path\n");
-		kfree(root);
-		return -ENOMEM;
+		ret = -ENOMEM;
+		goto out;
 	}
 
-	path->nodes[0] = eb = alloc_dummy_extent_buffer(NULL, 4096);
+	path->nodes[0] = eb = alloc_dummy_extent_buffer(NULL, nodesize,
+							nodesize);
 	if (!eb) {
 		test_msg("Could not allocate dummy buffer\n");
 		ret = -ENOMEM;
@@ -218,12 +227,13 @@ static int test_btrfs_split_item(void)
 	}
 out:
 	btrfs_free_path(path);
-	kfree(root);
+	btrfs_free_dummy_root(root);
+	btrfs_free_dummy_fs_info(fs_info);
 	return ret;
 }
 
-int btrfs_test_extent_buffer_operations(void)
+int btrfs_test_extent_buffer_operations(u32 sectorsize, u32 nodesize)
 {
-	test_msg("Running extent buffer operation tests");
-	return test_btrfs_split_item();
+	test_msg("Running extent buffer operation tests\n");
+	return test_btrfs_split_item(sectorsize, nodesize);
 }
diff --git a/fs/btrfs/tests/extent-io-tests.c b/fs/btrfs/tests/extent-io-tests.c
index 55724607f79b..caad80bb9bd0 100644
--- a/fs/btrfs/tests/extent-io-tests.c
+++ b/fs/btrfs/tests/extent-io-tests.c
@@ -21,6 +21,7 @@
 #include <linux/slab.h>
 #include <linux/sizes.h>
 #include "btrfs-tests.h"
+#include "../ctree.h"
 #include "../extent_io.h"
 
 #define PROCESS_UNLOCK		(1 << 0)
@@ -65,7 +66,7 @@ static noinline int process_page_range(struct inode *inode, u64 start, u64 end,
 	return count;
 }
 
-static int test_find_delalloc(void)
+static int test_find_delalloc(u32 sectorsize)
 {
 	struct inode *inode;
 	struct extent_io_tree tmp;
@@ -113,7 +114,7 @@ static int test_find_delalloc(void)
 	 * |--- delalloc ---|
 	 * |---  search  ---|
 	 */
-	set_extent_delalloc(&tmp, 0, 4095, NULL);
+	set_extent_delalloc(&tmp, 0, sectorsize - 1, NULL);
 	start = 0;
 	end = 0;
 	found = find_lock_delalloc_range(inode, &tmp, locked_page, &start,
@@ -122,9 +123,9 @@ static int test_find_delalloc(void)
 		test_msg("Should have found at least one delalloc\n");
 		goto out_bits;
 	}
-	if (start != 0 || end != 4095) {
-		test_msg("Expected start 0 end 4095, got start %Lu end %Lu\n",
-			 start, end);
+	if (start != 0 || end != (sectorsize - 1)) {
+		test_msg("Expected start 0 end %u, got start %llu end %llu\n",
+			sectorsize - 1, start, end);
 		goto out_bits;
 	}
 	unlock_extent(&tmp, start, end);
@@ -144,7 +145,7 @@ static int test_find_delalloc(void)
 		test_msg("Couldn't find the locked page\n");
 		goto out_bits;
 	}
-	set_extent_delalloc(&tmp, 4096, max_bytes - 1, NULL);
+	set_extent_delalloc(&tmp, sectorsize, max_bytes - 1, NULL);
 	start = test_start;
 	end = 0;
 	found = find_lock_delalloc_range(inode, &tmp, locked_page, &start,
@@ -172,7 +173,7 @@ static int test_find_delalloc(void)
 	 * |--- delalloc ---|
 	 *                    |--- search ---|
 	 */
-	test_start = max_bytes + 4096;
+	test_start = max_bytes + sectorsize;
 	locked_page = find_lock_page(inode->i_mapping, test_start >>
 				     PAGE_SHIFT);
 	if (!locked_page) {
@@ -272,10 +273,37 @@ out:
 	return ret;
 }
 
+static int check_eb_bitmap(unsigned long *bitmap, struct extent_buffer *eb,
+			   unsigned long len)
+{
+	unsigned long i;
+
+	for (i = 0; i < len * BITS_PER_BYTE; i++) {
+		int bit, bit1;
+
+		bit = !!test_bit(i, bitmap);
+		bit1 = !!extent_buffer_test_bit(eb, 0, i);
+		if (bit1 != bit) {
+			test_msg("Bits do not match\n");
+			return -EINVAL;
+		}
+
+		bit1 = !!extent_buffer_test_bit(eb, i / BITS_PER_BYTE,
+						i % BITS_PER_BYTE);
+		if (bit1 != bit) {
+			test_msg("Offset bits do not match\n");
+			return -EINVAL;
+		}
+	}
+	return 0;
+}
+
 static int __test_eb_bitmaps(unsigned long *bitmap, struct extent_buffer *eb,
 			     unsigned long len)
 {
-	unsigned long i, x;
+	unsigned long i, j;
+	u32 x;
+	int ret;
 
 	memset(bitmap, 0, len);
 	memset_extent_buffer(eb, 0, 0, len);
@@ -286,37 +314,45 @@ static int __test_eb_bitmaps(unsigned long *bitmap, struct extent_buffer *eb,
 
 	bitmap_set(bitmap, 0, len * BITS_PER_BYTE);
 	extent_buffer_bitmap_set(eb, 0, 0, len * BITS_PER_BYTE);
-	if (memcmp_extent_buffer(eb, bitmap, 0, len) != 0) {
+	ret = check_eb_bitmap(bitmap, eb, len);
+	if (ret) {
 		test_msg("Setting all bits failed\n");
-		return -EINVAL;
+		return ret;
 	}
 
 	bitmap_clear(bitmap, 0, len * BITS_PER_BYTE);
 	extent_buffer_bitmap_clear(eb, 0, 0, len * BITS_PER_BYTE);
-	if (memcmp_extent_buffer(eb, bitmap, 0, len) != 0) {
+	ret = check_eb_bitmap(bitmap, eb, len);
+	if (ret) {
 		test_msg("Clearing all bits failed\n");
-		return -EINVAL;
+		return ret;
 	}
 
-	bitmap_set(bitmap, (PAGE_SIZE - sizeof(long) / 2) * BITS_PER_BYTE,
-		   sizeof(long) * BITS_PER_BYTE);
-	extent_buffer_bitmap_set(eb, PAGE_SIZE - sizeof(long) / 2, 0,
-				 sizeof(long) * BITS_PER_BYTE);
-	if (memcmp_extent_buffer(eb, bitmap, 0, len) != 0) {
-		test_msg("Setting straddling pages failed\n");
-		return -EINVAL;
-	}
+	/* Straddling pages test */
+	if (len > PAGE_SIZE) {
+		bitmap_set(bitmap,
+			(PAGE_SIZE - sizeof(long) / 2) * BITS_PER_BYTE,
+			sizeof(long) * BITS_PER_BYTE);
+		extent_buffer_bitmap_set(eb, PAGE_SIZE - sizeof(long) / 2, 0,
+					sizeof(long) * BITS_PER_BYTE);
+		ret = check_eb_bitmap(bitmap, eb, len);
+		if (ret) {
+			test_msg("Setting straddling pages failed\n");
+			return ret;
+		}
 
-	bitmap_set(bitmap, 0, len * BITS_PER_BYTE);
-	bitmap_clear(bitmap,
-		     (PAGE_SIZE - sizeof(long) / 2) * BITS_PER_BYTE,
-		     sizeof(long) * BITS_PER_BYTE);
-	extent_buffer_bitmap_set(eb, 0, 0, len * BITS_PER_BYTE);
-	extent_buffer_bitmap_clear(eb, PAGE_SIZE - sizeof(long) / 2, 0,
-				   sizeof(long) * BITS_PER_BYTE);
-	if (memcmp_extent_buffer(eb, bitmap, 0, len) != 0) {
-		test_msg("Clearing straddling pages failed\n");
-		return -EINVAL;
+		bitmap_set(bitmap, 0, len * BITS_PER_BYTE);
+		bitmap_clear(bitmap,
+			(PAGE_SIZE - sizeof(long) / 2) * BITS_PER_BYTE,
+			sizeof(long) * BITS_PER_BYTE);
+		extent_buffer_bitmap_set(eb, 0, 0, len * BITS_PER_BYTE);
+		extent_buffer_bitmap_clear(eb, PAGE_SIZE - sizeof(long) / 2, 0,
+					sizeof(long) * BITS_PER_BYTE);
+		ret = check_eb_bitmap(bitmap, eb, len);
+		if (ret) {
+			test_msg("Clearing straddling pages failed\n");
+			return ret;
+		}
 	}
 
 	/*
@@ -324,42 +360,43 @@ static int __test_eb_bitmaps(unsigned long *bitmap, struct extent_buffer *eb,
 	 * something repetitive that could miss some hypothetical off-by-n bug.
 	 */
 	x = 0;
-	for (i = 0; i < len / sizeof(long); i++) {
-		x = (0x19660dULL * (u64)x + 0x3c6ef35fULL) & 0xffffffffUL;
-		bitmap[i] = x;
-	}
-	write_extent_buffer(eb, bitmap, 0, len);
-
-	for (i = 0; i < len * BITS_PER_BYTE; i++) {
-		int bit, bit1;
-
-		bit = !!test_bit(i, bitmap);
-		bit1 = !!extent_buffer_test_bit(eb, 0, i);
-		if (bit1 != bit) {
-			test_msg("Testing bit pattern failed\n");
-			return -EINVAL;
+	bitmap_clear(bitmap, 0, len * BITS_PER_BYTE);
+	extent_buffer_bitmap_clear(eb, 0, 0, len * BITS_PER_BYTE);
+	for (i = 0; i < len * BITS_PER_BYTE / 32; i++) {
+		x = (0x19660dULL * (u64)x + 0x3c6ef35fULL) & 0xffffffffU;
+		for (j = 0; j < 32; j++) {
+			if (x & (1U << j)) {
+				bitmap_set(bitmap, i * 32 + j, 1);
+				extent_buffer_bitmap_set(eb, 0, i * 32 + j, 1);
+			}
 		}
+	}
 
-		bit1 = !!extent_buffer_test_bit(eb, i / BITS_PER_BYTE,
-						i % BITS_PER_BYTE);
-		if (bit1 != bit) {
-			test_msg("Testing bit pattern with offset failed\n");
-			return -EINVAL;
-		}
+	ret = check_eb_bitmap(bitmap, eb, len);
+	if (ret) {
+		test_msg("Random bit pattern failed\n");
+		return ret;
 	}
 
 	return 0;
 }
 
-static int test_eb_bitmaps(void)
+static int test_eb_bitmaps(u32 sectorsize, u32 nodesize)
 {
-	unsigned long len = PAGE_SIZE * 4;
+	unsigned long len;
 	unsigned long *bitmap;
 	struct extent_buffer *eb;
 	int ret;
 
 	test_msg("Running extent buffer bitmap tests\n");
 
+	/*
+	 * In ppc64, sectorsize can be 64K, thus 4 * 64K will be larger than
+	 * BTRFS_MAX_METADATA_BLOCKSIZE.
+	 */
+	len = (sectorsize < BTRFS_MAX_METADATA_BLOCKSIZE)
+		? sectorsize * 4 : sectorsize;
+
 	bitmap = kmalloc(len, GFP_KERNEL);
 	if (!bitmap) {
 		test_msg("Couldn't allocate test bitmap\n");
@@ -379,7 +416,7 @@ static int test_eb_bitmaps(void)
 
 	/* Do it over again with an extent buffer which isn't page-aligned. */
 	free_extent_buffer(eb);
-	eb = __alloc_dummy_extent_buffer(NULL, PAGE_SIZE / 2, len);
+	eb = __alloc_dummy_extent_buffer(NULL, nodesize / 2, len);
 	if (!eb) {
 		test_msg("Couldn't allocate test extent buffer\n");
 		kfree(bitmap);
@@ -393,17 +430,17 @@ out:
 	return ret;
 }
 
-int btrfs_test_extent_io(void)
+int btrfs_test_extent_io(u32 sectorsize, u32 nodesize)
 {
 	int ret;
 
 	test_msg("Running extent I/O tests\n");
 
-	ret = test_find_delalloc();
+	ret = test_find_delalloc(sectorsize);
 	if (ret)
 		goto out;
 
-	ret = test_eb_bitmaps();
+	ret = test_eb_bitmaps(sectorsize, nodesize);
 out:
 	test_msg("Extent I/O tests finished\n");
 	return ret;
diff --git a/fs/btrfs/tests/free-space-tests.c b/fs/btrfs/tests/free-space-tests.c
index 0eeb8f3d6b67..3221c8dee272 100644
--- a/fs/btrfs/tests/free-space-tests.c
+++ b/fs/btrfs/tests/free-space-tests.c
@@ -22,7 +22,7 @@
 #include "../disk-io.h"
 #include "../free-space-cache.h"
 
-#define BITS_PER_BITMAP		(PAGE_SIZE * 8)
+#define BITS_PER_BITMAP		(PAGE_SIZE * 8UL)
 
 /*
  * This test just does basic sanity checking, making sure we can add an extent
@@ -99,7 +99,8 @@ static int test_extents(struct btrfs_block_group_cache *cache)
 	return 0;
 }
 
-static int test_bitmaps(struct btrfs_block_group_cache *cache)
+static int test_bitmaps(struct btrfs_block_group_cache *cache,
+			u32 sectorsize)
 {
 	u64 next_bitmap_offset;
 	int ret;
@@ -139,7 +140,7 @@ static int test_bitmaps(struct btrfs_block_group_cache *cache)
 	 * The first bitmap we have starts at offset 0 so the next one is just
 	 * at the end of the first bitmap.
 	 */
-	next_bitmap_offset = (u64)(BITS_PER_BITMAP * 4096);
+	next_bitmap_offset = (u64)(BITS_PER_BITMAP * sectorsize);
 
 	/* Test a bit straddling two bitmaps */
 	ret = test_add_free_space_entry(cache, next_bitmap_offset - SZ_2M,
@@ -167,9 +168,10 @@ static int test_bitmaps(struct btrfs_block_group_cache *cache)
 }
 
 /* This is the high grade jackassery */
-static int test_bitmaps_and_extents(struct btrfs_block_group_cache *cache)
+static int test_bitmaps_and_extents(struct btrfs_block_group_cache *cache,
+				    u32 sectorsize)
 {
-	u64 bitmap_offset = (u64)(BITS_PER_BITMAP * 4096);
+	u64 bitmap_offset = (u64)(BITS_PER_BITMAP * sectorsize);
 	int ret;
 
 	test_msg("Running bitmap and extent tests\n");
@@ -401,7 +403,8 @@ static int check_cache_empty(struct btrfs_block_group_cache *cache)
  * requests.
  */
 static int
-test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache)
+test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache,
+				       u32 sectorsize)
 {
 	int ret;
 	u64 offset;
@@ -539,7 +542,7 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache)
 	 * The goal is to test that the bitmap entry space stealing doesn't
 	 * steal this space region.
 	 */
-	ret = btrfs_add_free_space(cache, SZ_128M + SZ_16M, 4096);
+	ret = btrfs_add_free_space(cache, SZ_128M + SZ_16M, sectorsize);
 	if (ret) {
 		test_msg("Error adding free space: %d\n", ret);
 		return ret;
@@ -597,8 +600,8 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache)
 		return -ENOENT;
 	}
 
-	if (cache->free_space_ctl->free_space != (SZ_1M + 4096)) {
-		test_msg("Cache free space is not 1Mb + 4Kb\n");
+	if (cache->free_space_ctl->free_space != (SZ_1M + sectorsize)) {
+		test_msg("Cache free space is not 1Mb + %u\n", sectorsize);
 		return -EINVAL;
 	}
 
@@ -611,22 +614,25 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache)
 		return -EINVAL;
 	}
 
-	/* All that remains is a 4Kb free space region in a bitmap. Confirm. */
+	/*
+	 * All that remains is a sectorsize free space region in a bitmap.
+	 * Confirm.
+	 */
 	ret = check_num_extents_and_bitmaps(cache, 1, 1);
 	if (ret)
 		return ret;
 
-	if (cache->free_space_ctl->free_space != 4096) {
-		test_msg("Cache free space is not 4Kb\n");
+	if (cache->free_space_ctl->free_space != sectorsize) {
+		test_msg("Cache free space is not %u\n", sectorsize);
 		return -EINVAL;
 	}
 
 	offset = btrfs_find_space_for_alloc(cache,
-					    0, 4096, 0,
+					    0, sectorsize, 0,
 					    &max_extent_size);
 	if (offset != (SZ_128M + SZ_16M)) {
-		test_msg("Failed to allocate 4Kb from space cache, returned offset is: %llu\n",
-			 offset);
+		test_msg("Failed to allocate %u, returned offset : %llu\n",
+			 sectorsize, offset);
 		return -EINVAL;
 	}
 
@@ -733,7 +739,7 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache)
 	 * The goal is to test that the bitmap entry space stealing doesn't
 	 * steal this space region.
 	 */
-	ret = btrfs_add_free_space(cache, SZ_32M, 8192);
+	ret = btrfs_add_free_space(cache, SZ_32M, 2 * sectorsize);
 	if (ret) {
 		test_msg("Error adding free space: %d\n", ret);
 		return ret;
@@ -757,7 +763,7 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache)
 
 	/*
 	 * Confirm that our extent entry didn't stole all free space from the
-	 * bitmap, because of the small 8Kb free space region.
+	 * bitmap, because of the small 2 * sectorsize free space region.
 	 */
 	ret = check_num_extents_and_bitmaps(cache, 2, 1);
 	if (ret)
@@ -783,8 +789,8 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache)
 		return -ENOENT;
 	}
 
-	if (cache->free_space_ctl->free_space != (SZ_1M + 8192)) {
-		test_msg("Cache free space is not 1Mb + 8Kb\n");
+	if (cache->free_space_ctl->free_space != (SZ_1M + 2 * sectorsize)) {
+		test_msg("Cache free space is not 1Mb + %u\n", 2 * sectorsize);
 		return -EINVAL;
 	}
 
@@ -796,21 +802,25 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache)
 		return -EINVAL;
 	}
 
-	/* All that remains is a 8Kb free space region in a bitmap. Confirm. */
+	/*
+	 * All that remains is 2 * sectorsize free space region
+	 * in a bitmap. Confirm.
+	 */
 	ret = check_num_extents_and_bitmaps(cache, 1, 1);
 	if (ret)
 		return ret;
 
-	if (cache->free_space_ctl->free_space != 8192) {
-		test_msg("Cache free space is not 8Kb\n");
+	if (cache->free_space_ctl->free_space != 2 * sectorsize) {
+		test_msg("Cache free space is not %u\n", 2 * sectorsize);
 		return -EINVAL;
 	}
 
 	offset = btrfs_find_space_for_alloc(cache,
-					    0, 8192, 0,
+					    0, 2 * sectorsize, 0,
 					    &max_extent_size);
 	if (offset != SZ_32M) {
-		test_msg("Failed to allocate 8Kb from space cache, returned offset is: %llu\n",
+		test_msg("Failed to allocate %u, offset: %llu\n",
+			 2 * sectorsize,
 			 offset);
 		return -EINVAL;
 	}
@@ -825,29 +835,38 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache)
 	return 0;
 }
 
-int btrfs_test_free_space_cache(void)
+int btrfs_test_free_space_cache(u32 sectorsize, u32 nodesize)
 {
+	struct btrfs_fs_info *fs_info;
 	struct btrfs_block_group_cache *cache;
 	struct btrfs_root *root = NULL;
 	int ret = -ENOMEM;
 
 	test_msg("Running btrfs free space cache tests\n");
 
-	cache = btrfs_alloc_dummy_block_group(1024 * 1024 * 1024);
+	/*
+	 * For ppc64 (with 64k page size), bytes per bitmap might be
+	 * larger than 1G.  To make bitmap test available in ppc64,
+	 * alloc dummy block group whose size cross bitmaps.
+	 */
+	cache = btrfs_alloc_dummy_block_group(BITS_PER_BITMAP * sectorsize
+					+ PAGE_SIZE, sectorsize);
 	if (!cache) {
 		test_msg("Couldn't run the tests\n");
 		return 0;
 	}
 
-	root = btrfs_alloc_dummy_root();
-	if (IS_ERR(root)) {
-		ret = PTR_ERR(root);
+	fs_info = btrfs_alloc_dummy_fs_info();
+	if (!fs_info) {
+		ret = -ENOMEM;
 		goto out;
 	}
 
-	root->fs_info = btrfs_alloc_dummy_fs_info();
-	if (!root->fs_info)
+	root = btrfs_alloc_dummy_root(fs_info, sectorsize, nodesize);
+	if (IS_ERR(root)) {
+		ret = PTR_ERR(root);
 		goto out;
+	}
 
 	root->fs_info->extent_root = root;
 	cache->fs_info = root->fs_info;
@@ -855,17 +874,18 @@ int btrfs_test_free_space_cache(void)
 	ret = test_extents(cache);
 	if (ret)
 		goto out;
-	ret = test_bitmaps(cache);
+	ret = test_bitmaps(cache, sectorsize);
 	if (ret)
 		goto out;
-	ret = test_bitmaps_and_extents(cache);
+	ret = test_bitmaps_and_extents(cache, sectorsize);
 	if (ret)
 		goto out;
 
-	ret = test_steal_space_from_bitmap_to_extent(cache);
+	ret = test_steal_space_from_bitmap_to_extent(cache, sectorsize);
 out:
 	btrfs_free_dummy_block_group(cache);
 	btrfs_free_dummy_root(root);
+	btrfs_free_dummy_fs_info(fs_info);
 	test_msg("Free space cache tests finished\n");
 	return ret;
 }
diff --git a/fs/btrfs/tests/free-space-tree-tests.c b/fs/btrfs/tests/free-space-tree-tests.c
index 7cea4462acd5..6e144048a72e 100644
--- a/fs/btrfs/tests/free-space-tree-tests.c
+++ b/fs/btrfs/tests/free-space-tree-tests.c
@@ -16,6 +16,7 @@
  * Boston, MA 021110-1307, USA.
  */
 
+#include <linux/types.h>
 #include "btrfs-tests.h"
 #include "../ctree.h"
 #include "../disk-io.h"
@@ -23,20 +24,15 @@
 #include "../transaction.h"
 
 struct free_space_extent {
-	u64 start, length;
+	u64 start;
+	u64 length;
 };
 
-/*
- * The test cases align their operations to this in order to hit some of the
- * edge cases in the bitmap code.
- */
-#define BITMAP_RANGE (BTRFS_FREE_SPACE_BITMAP_BITS * 4096)
-
 static int __check_free_space_extents(struct btrfs_trans_handle *trans,
 				      struct btrfs_fs_info *fs_info,
 				      struct btrfs_block_group_cache *cache,
 				      struct btrfs_path *path,
-				      struct free_space_extent *extents,
+				      const struct free_space_extent * const extents,
 				      unsigned int num_extents)
 {
 	struct btrfs_free_space_info *info;
@@ -125,7 +121,7 @@ static int check_free_space_extents(struct btrfs_trans_handle *trans,
 				    struct btrfs_fs_info *fs_info,
 				    struct btrfs_block_group_cache *cache,
 				    struct btrfs_path *path,
-				    struct free_space_extent *extents,
+				    const struct free_space_extent * const extents,
 				    unsigned int num_extents)
 {
 	struct btrfs_free_space_info *info;
@@ -167,9 +163,10 @@ static int check_free_space_extents(struct btrfs_trans_handle *trans,
 static int test_empty_block_group(struct btrfs_trans_handle *trans,
 				  struct btrfs_fs_info *fs_info,
 				  struct btrfs_block_group_cache *cache,
-				  struct btrfs_path *path)
+				  struct btrfs_path *path,
+				  u32 alignment)
 {
-	struct free_space_extent extents[] = {
+	const struct free_space_extent extents[] = {
 		{cache->key.objectid, cache->key.offset},
 	};
 
@@ -180,9 +177,10 @@ static int test_empty_block_group(struct btrfs_trans_handle *trans,
 static int test_remove_all(struct btrfs_trans_handle *trans,
 			   struct btrfs_fs_info *fs_info,
 			   struct btrfs_block_group_cache *cache,
-			   struct btrfs_path *path)
+			   struct btrfs_path *path,
+			   u32 alignment)
 {
-	struct free_space_extent extents[] = {};
+	const struct free_space_extent extents[] = {};
 	int ret;
 
 	ret = __remove_from_free_space_tree(trans, fs_info, cache, path,
@@ -200,16 +198,17 @@ static int test_remove_all(struct btrfs_trans_handle *trans,
 static int test_remove_beginning(struct btrfs_trans_handle *trans,
 				 struct btrfs_fs_info *fs_info,
 				 struct btrfs_block_group_cache *cache,
-				 struct btrfs_path *path)
+				 struct btrfs_path *path,
+				 u32 alignment)
 {
-	struct free_space_extent extents[] = {
-		{cache->key.objectid + BITMAP_RANGE,
-			cache->key.offset - BITMAP_RANGE},
+	const struct free_space_extent extents[] = {
+		{cache->key.objectid + alignment,
+			cache->key.offset - alignment},
 	};
 	int ret;
 
 	ret = __remove_from_free_space_tree(trans, fs_info, cache, path,
-					    cache->key.objectid, BITMAP_RANGE);
+					    cache->key.objectid, alignment);
 	if (ret) {
 		test_msg("Could not remove free space\n");
 		return ret;
@@ -223,17 +222,18 @@ static int test_remove_beginning(struct btrfs_trans_handle *trans,
 static int test_remove_end(struct btrfs_trans_handle *trans,
 			   struct btrfs_fs_info *fs_info,
 			   struct btrfs_block_group_cache *cache,
-			   struct btrfs_path *path)
+			   struct btrfs_path *path,
+			   u32 alignment)
 {
-	struct free_space_extent extents[] = {
-		{cache->key.objectid, cache->key.offset - BITMAP_RANGE},
+	const struct free_space_extent extents[] = {
+		{cache->key.objectid, cache->key.offset - alignment},
 	};
 	int ret;
 
 	ret = __remove_from_free_space_tree(trans, fs_info, cache, path,
 					    cache->key.objectid +
-					    cache->key.offset - BITMAP_RANGE,
-					    BITMAP_RANGE);
+					    cache->key.offset - alignment,
+					    alignment);
 	if (ret) {
 		test_msg("Could not remove free space\n");
 		return ret;
@@ -246,18 +246,19 @@ static int test_remove_end(struct btrfs_trans_handle *trans,
 static int test_remove_middle(struct btrfs_trans_handle *trans,
 			      struct btrfs_fs_info *fs_info,
 			      struct btrfs_block_group_cache *cache,
-			      struct btrfs_path *path)
+			      struct btrfs_path *path,
+			      u32 alignment)
 {
-	struct free_space_extent extents[] = {
-		{cache->key.objectid, BITMAP_RANGE},
-		{cache->key.objectid + 2 * BITMAP_RANGE,
-			cache->key.offset - 2 * BITMAP_RANGE},
+	const struct free_space_extent extents[] = {
+		{cache->key.objectid, alignment},
+		{cache->key.objectid + 2 * alignment,
+			cache->key.offset - 2 * alignment},
 	};
 	int ret;
 
 	ret = __remove_from_free_space_tree(trans, fs_info, cache, path,
-					    cache->key.objectid + BITMAP_RANGE,
-					    BITMAP_RANGE);
+					    cache->key.objectid + alignment,
+					    alignment);
 	if (ret) {
 		test_msg("Could not remove free space\n");
 		return ret;
@@ -270,10 +271,11 @@ static int test_remove_middle(struct btrfs_trans_handle *trans,
 static int test_merge_left(struct btrfs_trans_handle *trans,
 			   struct btrfs_fs_info *fs_info,
 			   struct btrfs_block_group_cache *cache,
-			   struct btrfs_path *path)
+			   struct btrfs_path *path,
+			   u32 alignment)
 {
-	struct free_space_extent extents[] = {
-		{cache->key.objectid, 2 * BITMAP_RANGE},
+	const struct free_space_extent extents[] = {
+		{cache->key.objectid, 2 * alignment},
 	};
 	int ret;
 
@@ -286,15 +288,15 @@ static int test_merge_left(struct btrfs_trans_handle *trans,
 	}
 
 	ret = __add_to_free_space_tree(trans, fs_info, cache, path,
-				       cache->key.objectid, BITMAP_RANGE);
+				       cache->key.objectid, alignment);
 	if (ret) {
 		test_msg("Could not add free space\n");
 		return ret;
 	}
 
 	ret = __add_to_free_space_tree(trans, fs_info, cache, path,
-				       cache->key.objectid + BITMAP_RANGE,
-				       BITMAP_RANGE);
+				       cache->key.objectid + alignment,
+				       alignment);
 	if (ret) {
 		test_msg("Could not add free space\n");
 		return ret;
@@ -307,10 +309,11 @@ static int test_merge_left(struct btrfs_trans_handle *trans,
 static int test_merge_right(struct btrfs_trans_handle *trans,
 			   struct btrfs_fs_info *fs_info,
 			   struct btrfs_block_group_cache *cache,
-			   struct btrfs_path *path)
+			   struct btrfs_path *path,
+			   u32 alignment)
 {
-	struct free_space_extent extents[] = {
-		{cache->key.objectid + BITMAP_RANGE, 2 * BITMAP_RANGE},
+	const struct free_space_extent extents[] = {
+		{cache->key.objectid + alignment, 2 * alignment},
 	};
 	int ret;
 
@@ -323,16 +326,16 @@ static int test_merge_right(struct btrfs_trans_handle *trans,
 	}
 
 	ret = __add_to_free_space_tree(trans, fs_info, cache, path,
-				       cache->key.objectid + 2 * BITMAP_RANGE,
-				       BITMAP_RANGE);
+				       cache->key.objectid + 2 * alignment,
+				       alignment);
 	if (ret) {
 		test_msg("Could not add free space\n");
 		return ret;
 	}
 
 	ret = __add_to_free_space_tree(trans, fs_info, cache, path,
-				       cache->key.objectid + BITMAP_RANGE,
-				       BITMAP_RANGE);
+				       cache->key.objectid + alignment,
+				       alignment);
 	if (ret) {
 		test_msg("Could not add free space\n");
 		return ret;
@@ -345,10 +348,11 @@ static int test_merge_right(struct btrfs_trans_handle *trans,
 static int test_merge_both(struct btrfs_trans_handle *trans,
 			   struct btrfs_fs_info *fs_info,
 			   struct btrfs_block_group_cache *cache,
-			   struct btrfs_path *path)
+			   struct btrfs_path *path,
+			   u32 alignment)
 {
-	struct free_space_extent extents[] = {
-		{cache->key.objectid, 3 * BITMAP_RANGE},
+	const struct free_space_extent extents[] = {
+		{cache->key.objectid, 3 * alignment},
 	};
 	int ret;
 
@@ -361,23 +365,23 @@ static int test_merge_both(struct btrfs_trans_handle *trans,
 	}
 
 	ret = __add_to_free_space_tree(trans, fs_info, cache, path,
-				       cache->key.objectid, BITMAP_RANGE);
+				       cache->key.objectid, alignment);
 	if (ret) {
 		test_msg("Could not add free space\n");
 		return ret;
 	}
 
 	ret = __add_to_free_space_tree(trans, fs_info, cache, path,
-				       cache->key.objectid + 2 * BITMAP_RANGE,
-				       BITMAP_RANGE);
+				       cache->key.objectid + 2 * alignment,
+				       alignment);
 	if (ret) {
 		test_msg("Could not add free space\n");
 		return ret;
 	}
 
 	ret = __add_to_free_space_tree(trans, fs_info, cache, path,
-				       cache->key.objectid + BITMAP_RANGE,
-				       BITMAP_RANGE);
+				       cache->key.objectid + alignment,
+				       alignment);
 	if (ret) {
 		test_msg("Could not add free space\n");
 		return ret;
@@ -390,12 +394,13 @@ static int test_merge_both(struct btrfs_trans_handle *trans,
 static int test_merge_none(struct btrfs_trans_handle *trans,
 			   struct btrfs_fs_info *fs_info,
 			   struct btrfs_block_group_cache *cache,
-			   struct btrfs_path *path)
+			   struct btrfs_path *path,
+			   u32 alignment)
 {
-	struct free_space_extent extents[] = {
-		{cache->key.objectid, BITMAP_RANGE},
-		{cache->key.objectid + 2 * BITMAP_RANGE, BITMAP_RANGE},
-		{cache->key.objectid + 4 * BITMAP_RANGE, BITMAP_RANGE},
+	const struct free_space_extent extents[] = {
+		{cache->key.objectid, alignment},
+		{cache->key.objectid + 2 * alignment, alignment},
+		{cache->key.objectid + 4 * alignment, alignment},
 	};
 	int ret;
 
@@ -408,23 +413,23 @@ static int test_merge_none(struct btrfs_trans_handle *trans,
 	}
 
 	ret = __add_to_free_space_tree(trans, fs_info, cache, path,
-				       cache->key.objectid, BITMAP_RANGE);
+				       cache->key.objectid, alignment);
 	if (ret) {
 		test_msg("Could not add free space\n");
 		return ret;
 	}
 
 	ret = __add_to_free_space_tree(trans, fs_info, cache, path,
-				       cache->key.objectid + 4 * BITMAP_RANGE,
-				       BITMAP_RANGE);
+				       cache->key.objectid + 4 * alignment,
+				       alignment);
 	if (ret) {
 		test_msg("Could not add free space\n");
 		return ret;
 	}
 
 	ret = __add_to_free_space_tree(trans, fs_info, cache, path,
-				       cache->key.objectid + 2 * BITMAP_RANGE,
-				       BITMAP_RANGE);
+				       cache->key.objectid + 2 * alignment,
+				       alignment);
 	if (ret) {
 		test_msg("Could not add free space\n");
 		return ret;
@@ -437,27 +442,30 @@ static int test_merge_none(struct btrfs_trans_handle *trans,
 typedef int (*test_func_t)(struct btrfs_trans_handle *,
 			   struct btrfs_fs_info *,
 			   struct btrfs_block_group_cache *,
-			   struct btrfs_path *);
+			   struct btrfs_path *,
+			   u32 alignment);
 
-static int run_test(test_func_t test_func, int bitmaps)
+static int run_test(test_func_t test_func, int bitmaps, u32 sectorsize,
+		    u32 nodesize, u32 alignment)
 {
+	struct btrfs_fs_info *fs_info;
 	struct btrfs_root *root = NULL;
 	struct btrfs_block_group_cache *cache = NULL;
 	struct btrfs_trans_handle trans;
 	struct btrfs_path *path = NULL;
 	int ret;
 
-	root = btrfs_alloc_dummy_root();
-	if (IS_ERR(root)) {
-		test_msg("Couldn't allocate dummy root\n");
-		ret = PTR_ERR(root);
+	fs_info = btrfs_alloc_dummy_fs_info();
+	if (!fs_info) {
+		test_msg("Couldn't allocate dummy fs info\n");
+		ret = -ENOMEM;
 		goto out;
 	}
 
-	root->fs_info = btrfs_alloc_dummy_fs_info();
-	if (!root->fs_info) {
-		test_msg("Couldn't allocate dummy fs info\n");
-		ret = -ENOMEM;
+	root = btrfs_alloc_dummy_root(fs_info, sectorsize, nodesize);
+	if (IS_ERR(root)) {
+		test_msg("Couldn't allocate dummy root\n");
+		ret = PTR_ERR(root);
 		goto out;
 	}
 
@@ -466,7 +474,8 @@ static int run_test(test_func_t test_func, int bitmaps)
 	root->fs_info->free_space_root = root;
 	root->fs_info->tree_root = root;
 
-	root->node = alloc_test_extent_buffer(root->fs_info, 4096);
+	root->node = alloc_test_extent_buffer(root->fs_info,
+		nodesize, nodesize);
 	if (!root->node) {
 		test_msg("Couldn't allocate dummy buffer\n");
 		ret = -ENOMEM;
@@ -474,9 +483,9 @@ static int run_test(test_func_t test_func, int bitmaps)
 	}
 	btrfs_set_header_level(root->node, 0);
 	btrfs_set_header_nritems(root->node, 0);
-	root->alloc_bytenr += 8192;
+	root->alloc_bytenr += 2 * nodesize;
 
-	cache = btrfs_alloc_dummy_block_group(8 * BITMAP_RANGE);
+	cache = btrfs_alloc_dummy_block_group(8 * alignment, sectorsize);
 	if (!cache) {
 		test_msg("Couldn't allocate dummy block group cache\n");
 		ret = -ENOMEM;
@@ -510,7 +519,7 @@ static int run_test(test_func_t test_func, int bitmaps)
 		}
 	}
 
-	ret = test_func(&trans, root->fs_info, cache, path);
+	ret = test_func(&trans, root->fs_info, cache, path, alignment);
 	if (ret)
 		goto out;
 
@@ -531,20 +540,34 @@ out:
 	btrfs_free_path(path);
 	btrfs_free_dummy_block_group(cache);
 	btrfs_free_dummy_root(root);
+	btrfs_free_dummy_fs_info(fs_info);
 	return ret;
 }
 
-static int run_test_both_formats(test_func_t test_func)
+static int run_test_both_formats(test_func_t test_func, u32 sectorsize,
+				 u32 nodesize, u32 alignment)
 {
+	int test_ret = 0;
 	int ret;
 
-	ret = run_test(test_func, 0);
-	if (ret)
-		return ret;
-	return run_test(test_func, 1);
+	ret = run_test(test_func, 0, sectorsize, nodesize, alignment);
+	if (ret) {
+		test_msg("%pf failed with extents, sectorsize=%u, nodesize=%u, alignment=%u\n",
+			 test_func, sectorsize, nodesize, alignment);
+		test_ret = ret;
+	}
+
+	ret = run_test(test_func, 1, sectorsize, nodesize, alignment);
+	if (ret) {
+		test_msg("%pf failed with bitmaps, sectorsize=%u, nodesize=%u, alignment=%u\n",
+			 test_func, sectorsize, nodesize, alignment);
+		test_ret = ret;
+	}
+
+	return test_ret;
 }
 
-int btrfs_test_free_space_tree(void)
+int btrfs_test_free_space_tree(u32 sectorsize, u32 nodesize)
 {
 	test_func_t tests[] = {
 		test_empty_block_group,
@@ -557,16 +580,30 @@ int btrfs_test_free_space_tree(void)
 		test_merge_both,
 		test_merge_none,
 	};
+	u32 bitmap_alignment;
+	int test_ret = 0;
 	int i;
 
+	/*
+	 * Align some operations to a page to flush out bugs in the extent
+	 * buffer bitmap handling of highmem.
+	 */
+	bitmap_alignment = BTRFS_FREE_SPACE_BITMAP_BITS * PAGE_SIZE;
+
 	test_msg("Running free space tree tests\n");
 	for (i = 0; i < ARRAY_SIZE(tests); i++) {
-		int ret = run_test_both_formats(tests[i]);
-		if (ret) {
-			test_msg("%pf failed\n", tests[i]);
-			return ret;
-		}
+		int ret;
+
+		ret = run_test_both_formats(tests[i], sectorsize, nodesize,
+					    sectorsize);
+		if (ret)
+			test_ret = ret;
+
+		ret = run_test_both_formats(tests[i], sectorsize, nodesize,
+					    bitmap_alignment);
+		if (ret)
+			test_ret = ret;
 	}
 
-	return 0;
+	return test_ret;
 }
diff --git a/fs/btrfs/tests/inode-tests.c b/fs/btrfs/tests/inode-tests.c
index 8a25fe8b7c45..0bf46808ce8f 100644
--- a/fs/btrfs/tests/inode-tests.c
+++ b/fs/btrfs/tests/inode-tests.c
@@ -16,6 +16,7 @@
  * Boston, MA 021110-1307, USA.
  */
 
+#include <linux/types.h>
 #include "btrfs-tests.h"
 #include "../ctree.h"
 #include "../btrfs_inode.h"
@@ -86,19 +87,19 @@ static void insert_inode_item_key(struct btrfs_root *root)
  * diagram of how the extents will look though this may not be possible we still
  * want to make sure everything acts normally (the last number is not inclusive)
  *
- * [0 - 5][5 -  6][6 - 10][10 - 4096][  4096 - 8192 ][8192 - 12288]
- * [hole ][inline][ hole ][ regular ][regular1 split][    hole    ]
+ * [0 - 5][5 -  6][     6 - 4096     ][ 4096 - 4100][4100 - 8195][8195 - 12291]
+ * [hole ][inline][hole but no extent][  hole   ][   regular ][regular1 split]
  *
- * [ 12288 - 20480][20480 - 24576][  24576 - 28672  ][28672 - 36864][36864 - 45056]
- * [regular1 split][   prealloc1 ][prealloc1 written][   prealloc1 ][ compressed  ]
+ * [12291 - 16387][16387 - 24579][24579 - 28675][ 28675 - 32771][32771 - 36867 ]
+ * [    hole    ][regular1 split][   prealloc ][   prealloc1  ][prealloc1 written]
  *
- * [45056 - 49152][49152-53248][53248-61440][61440-65536][     65536+81920   ]
- * [ compressed1 ][  regular  ][compressed1][  regular  ][ hole but no extent]
+ * [36867 - 45059][45059 - 53251][53251 - 57347][57347 - 61443][61443- 69635]
+ * [  prealloc1  ][ compressed  ][ compressed1 ][    regular  ][ compressed1]
  *
- * [81920-86016]
- * [  regular  ]
+ * [69635-73731][   73731 - 86019   ][86019-90115]
+ * [  regular  ][ hole but no extent][  regular  ]
  */
-static void setup_file_extents(struct btrfs_root *root)
+static void setup_file_extents(struct btrfs_root *root, u32 sectorsize)
 {
 	int slot = 0;
 	u64 disk_bytenr = SZ_1M;
@@ -119,7 +120,7 @@ static void setup_file_extents(struct btrfs_root *root)
 	insert_extent(root, offset, 1, 1, 0, 0, 0, BTRFS_FILE_EXTENT_INLINE, 0,
 		      slot);
 	slot++;
-	offset = 4096;
+	offset = sectorsize;
 
 	/* Now another hole */
 	insert_extent(root, offset, 4, 4, 0, 0, 0, BTRFS_FILE_EXTENT_REG, 0,
@@ -128,100 +129,108 @@ static void setup_file_extents(struct btrfs_root *root)
 	offset += 4;
 
 	/* Now for a regular extent */
-	insert_extent(root, offset, 4095, 4095, 0, disk_bytenr, 4096,
-		      BTRFS_FILE_EXTENT_REG, 0, slot);
+	insert_extent(root, offset, sectorsize - 1, sectorsize - 1, 0,
+		      disk_bytenr, sectorsize, BTRFS_FILE_EXTENT_REG, 0, slot);
 	slot++;
-	disk_bytenr += 4096;
-	offset += 4095;
+	disk_bytenr += sectorsize;
+	offset += sectorsize - 1;
 
 	/*
 	 * Now for 3 extents that were split from a hole punch so we test
 	 * offsets properly.
 	 */
-	insert_extent(root, offset, 4096, 16384, 0, disk_bytenr, 16384,
-		      BTRFS_FILE_EXTENT_REG, 0, slot);
+	insert_extent(root, offset, sectorsize, 4 * sectorsize, 0, disk_bytenr,
+		      4 * sectorsize, BTRFS_FILE_EXTENT_REG, 0, slot);
 	slot++;
-	offset += 4096;
-	insert_extent(root, offset, 4096, 4096, 0, 0, 0, BTRFS_FILE_EXTENT_REG,
-		      0, slot);
+	offset += sectorsize;
+	insert_extent(root, offset, sectorsize, sectorsize, 0, 0, 0,
+		      BTRFS_FILE_EXTENT_REG, 0, slot);
 	slot++;
-	offset += 4096;
-	insert_extent(root, offset, 8192, 16384, 8192, disk_bytenr, 16384,
+	offset += sectorsize;
+	insert_extent(root, offset, 2 * sectorsize, 4 * sectorsize,
+		      2 * sectorsize, disk_bytenr, 4 * sectorsize,
 		      BTRFS_FILE_EXTENT_REG, 0, slot);
 	slot++;
-	offset += 8192;
-	disk_bytenr += 16384;
+	offset += 2 * sectorsize;
+	disk_bytenr += 4 * sectorsize;
 
 	/* Now for a unwritten prealloc extent */
-	insert_extent(root, offset, 4096, 4096, 0, disk_bytenr, 4096,
-		      BTRFS_FILE_EXTENT_PREALLOC, 0, slot);
+	insert_extent(root, offset, sectorsize, sectorsize, 0, disk_bytenr,
+		sectorsize, BTRFS_FILE_EXTENT_PREALLOC, 0, slot);
 	slot++;
-	offset += 4096;
+	offset += sectorsize;
 
 	/*
 	 * We want to jack up disk_bytenr a little more so the em stuff doesn't
 	 * merge our records.
 	 */
-	disk_bytenr += 8192;
+	disk_bytenr += 2 * sectorsize;
 
 	/*
 	 * Now for a partially written prealloc extent, basically the same as
 	 * the hole punch example above.  Ram_bytes never changes when you mark
 	 * extents written btw.
 	 */
-	insert_extent(root, offset, 4096, 16384, 0, disk_bytenr, 16384,
-		      BTRFS_FILE_EXTENT_PREALLOC, 0, slot);
+	insert_extent(root, offset, sectorsize, 4 * sectorsize, 0, disk_bytenr,
+		      4 * sectorsize, BTRFS_FILE_EXTENT_PREALLOC, 0, slot);
 	slot++;
-	offset += 4096;
-	insert_extent(root, offset, 4096, 16384, 4096, disk_bytenr, 16384,
-		      BTRFS_FILE_EXTENT_REG, 0, slot);
+	offset += sectorsize;
+	insert_extent(root, offset, sectorsize, 4 * sectorsize, sectorsize,
+		      disk_bytenr, 4 * sectorsize, BTRFS_FILE_EXTENT_REG, 0,
+		      slot);
 	slot++;
-	offset += 4096;
-	insert_extent(root, offset, 8192, 16384, 8192, disk_bytenr, 16384,
+	offset += sectorsize;
+	insert_extent(root, offset, 2 * sectorsize, 4 * sectorsize,
+		      2 * sectorsize, disk_bytenr, 4 * sectorsize,
 		      BTRFS_FILE_EXTENT_PREALLOC, 0, slot);
 	slot++;
-	offset += 8192;
-	disk_bytenr += 16384;
+	offset += 2 * sectorsize;
+	disk_bytenr += 4 * sectorsize;
 
 	/* Now a normal compressed extent */
-	insert_extent(root, offset, 8192, 8192, 0, disk_bytenr, 4096,
-		      BTRFS_FILE_EXTENT_REG, BTRFS_COMPRESS_ZLIB, slot);
+	insert_extent(root, offset, 2 * sectorsize, 2 * sectorsize, 0,
+		      disk_bytenr, sectorsize, BTRFS_FILE_EXTENT_REG,
+		      BTRFS_COMPRESS_ZLIB, slot);
 	slot++;
-	offset += 8192;
+	offset += 2 * sectorsize;
 	/* No merges */
-	disk_bytenr += 8192;
+	disk_bytenr += 2 * sectorsize;
 
 	/* Now a split compressed extent */
-	insert_extent(root, offset, 4096, 16384, 0, disk_bytenr, 4096,
-		      BTRFS_FILE_EXTENT_REG, BTRFS_COMPRESS_ZLIB, slot);
+	insert_extent(root, offset, sectorsize, 4 * sectorsize, 0, disk_bytenr,
+		      sectorsize, BTRFS_FILE_EXTENT_REG,
+		      BTRFS_COMPRESS_ZLIB, slot);
 	slot++;
-	offset += 4096;
-	insert_extent(root, offset, 4096, 4096, 0, disk_bytenr + 4096, 4096,
+	offset += sectorsize;
+	insert_extent(root, offset, sectorsize, sectorsize, 0,
+		      disk_bytenr + sectorsize, sectorsize,
 		      BTRFS_FILE_EXTENT_REG, 0, slot);
 	slot++;
-	offset += 4096;
-	insert_extent(root, offset, 8192, 16384, 8192, disk_bytenr, 4096,
+	offset += sectorsize;
+	insert_extent(root, offset, 2 * sectorsize, 4 * sectorsize,
+		      2 * sectorsize, disk_bytenr, sectorsize,
 		      BTRFS_FILE_EXTENT_REG, BTRFS_COMPRESS_ZLIB, slot);
 	slot++;
-	offset += 8192;
-	disk_bytenr += 8192;
+	offset += 2 * sectorsize;
+	disk_bytenr += 2 * sectorsize;
 
 	/* Now extents that have a hole but no hole extent */
-	insert_extent(root, offset, 4096, 4096, 0, disk_bytenr, 4096,
-		      BTRFS_FILE_EXTENT_REG, 0, slot);
+	insert_extent(root, offset, sectorsize, sectorsize, 0, disk_bytenr,
+		      sectorsize, BTRFS_FILE_EXTENT_REG, 0, slot);
 	slot++;
-	offset += 16384;
-	disk_bytenr += 4096;
-	insert_extent(root, offset, 4096, 4096, 0, disk_bytenr, 4096,
-		      BTRFS_FILE_EXTENT_REG, 0, slot);
+	offset += 4 * sectorsize;
+	disk_bytenr += sectorsize;
+	insert_extent(root, offset, sectorsize, sectorsize, 0, disk_bytenr,
+		      sectorsize, BTRFS_FILE_EXTENT_REG, 0, slot);
 }
 
 static unsigned long prealloc_only = 0;
 static unsigned long compressed_only = 0;
 static unsigned long vacancy_only = 0;
 
-static noinline int test_btrfs_get_extent(void)
+static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
 {
+	struct btrfs_fs_info *fs_info = NULL;
 	struct inode *inode = NULL;
 	struct btrfs_root *root = NULL;
 	struct extent_map *em = NULL;
@@ -240,23 +249,19 @@ static noinline int test_btrfs_get_extent(void)
 	BTRFS_I(inode)->location.objectid = BTRFS_FIRST_FREE_OBJECTID;
 	BTRFS_I(inode)->location.offset = 0;
 
-	root = btrfs_alloc_dummy_root();
-	if (IS_ERR(root)) {
-		test_msg("Couldn't allocate root\n");
+	fs_info = btrfs_alloc_dummy_fs_info();
+	if (!fs_info) {
+		test_msg("Couldn't allocate dummy fs info\n");
 		goto out;
 	}
 
-	/*
-	 * We do this since btrfs_get_extent wants to assign em->bdev to
-	 * root->fs_info->fs_devices->latest_bdev.
-	 */
-	root->fs_info = btrfs_alloc_dummy_fs_info();
-	if (!root->fs_info) {
-		test_msg("Couldn't allocate dummy fs info\n");
+	root = btrfs_alloc_dummy_root(fs_info, sectorsize, nodesize);
+	if (IS_ERR(root)) {
+		test_msg("Couldn't allocate root\n");
 		goto out;
 	}
 
-	root->node = alloc_dummy_extent_buffer(NULL, 4096);
+	root->node = alloc_dummy_extent_buffer(NULL, nodesize, nodesize);
 	if (!root->node) {
 		test_msg("Couldn't allocate dummy buffer\n");
 		goto out;
@@ -273,7 +278,7 @@ static noinline int test_btrfs_get_extent(void)
 
 	/* First with no extents */
 	BTRFS_I(inode)->root = root;
-	em = btrfs_get_extent(inode, NULL, 0, 0, 4096, 0);
+	em = btrfs_get_extent(inode, NULL, 0, 0, sectorsize, 0);
 	if (IS_ERR(em)) {
 		em = NULL;
 		test_msg("Got an error when we shouldn't have\n");
@@ -295,7 +300,7 @@ static noinline int test_btrfs_get_extent(void)
 	 * setup_file_extents, so if you change anything there you need to
 	 * update the comment and update the expected values below.
 	 */
-	setup_file_extents(root);
+	setup_file_extents(root, sectorsize);
 
 	em = btrfs_get_extent(inode, NULL, 0, 0, (u64)-1, 0);
 	if (IS_ERR(em)) {
@@ -318,7 +323,7 @@ static noinline int test_btrfs_get_extent(void)
 	offset = em->start + em->len;
 	free_extent_map(em);
 
-	em = btrfs_get_extent(inode, NULL, 0, offset, 4096, 0);
+	em = btrfs_get_extent(inode, NULL, 0, offset, sectorsize, 0);
 	if (IS_ERR(em)) {
 		test_msg("Got an error when we shouldn't have\n");
 		goto out;
@@ -327,7 +332,8 @@ static noinline int test_btrfs_get_extent(void)
 		test_msg("Expected an inline, got %llu\n", em->block_start);
 		goto out;
 	}
-	if (em->start != offset || em->len != 4091) {
+
+	if (em->start != offset || em->len != (sectorsize - 5)) {
 		test_msg("Unexpected extent wanted start %llu len 1, got start "
 			 "%llu len %llu\n", offset, em->start, em->len);
 		goto out;
@@ -344,7 +350,7 @@ static noinline int test_btrfs_get_extent(void)
 	offset = em->start + em->len;
 	free_extent_map(em);
 
-	em = btrfs_get_extent(inode, NULL, 0, offset, 4096, 0);
+	em = btrfs_get_extent(inode, NULL, 0, offset, sectorsize, 0);
 	if (IS_ERR(em)) {
 		test_msg("Got an error when we shouldn't have\n");
 		goto out;
@@ -366,7 +372,7 @@ static noinline int test_btrfs_get_extent(void)
 	free_extent_map(em);
 
 	/* Regular extent */
-	em = btrfs_get_extent(inode, NULL, 0, offset, 4096, 0);
+	em = btrfs_get_extent(inode, NULL, 0, offset, sectorsize, 0);
 	if (IS_ERR(em)) {
 		test_msg("Got an error when we shouldn't have\n");
 		goto out;
@@ -375,7 +381,7 @@ static noinline int test_btrfs_get_extent(void)
 		test_msg("Expected a real extent, got %llu\n", em->block_start);
 		goto out;
 	}
-	if (em->start != offset || em->len != 4095) {
+	if (em->start != offset || em->len != sectorsize - 1) {
 		test_msg("Unexpected extent wanted start %llu len 4095, got "
 			 "start %llu len %llu\n", offset, em->start, em->len);
 		goto out;
@@ -393,7 +399,7 @@ static noinline int test_btrfs_get_extent(void)
 	free_extent_map(em);
 
 	/* The next 3 are split extents */
-	em = btrfs_get_extent(inode, NULL, 0, offset, 4096, 0);
+	em = btrfs_get_extent(inode, NULL, 0, offset, sectorsize, 0);
 	if (IS_ERR(em)) {
 		test_msg("Got an error when we shouldn't have\n");
 		goto out;
@@ -402,9 +408,10 @@ static noinline int test_btrfs_get_extent(void)
 		test_msg("Expected a real extent, got %llu\n", em->block_start);
 		goto out;
 	}
-	if (em->start != offset || em->len != 4096) {
-		test_msg("Unexpected extent wanted start %llu len 4096, got "
-			 "start %llu len %llu\n", offset, em->start, em->len);
+	if (em->start != offset || em->len != sectorsize) {
+		test_msg("Unexpected extent start %llu len %u, "
+			"got start %llu len %llu\n",
+			offset, sectorsize, em->start, em->len);
 		goto out;
 	}
 	if (em->flags != 0) {
@@ -421,7 +428,7 @@ static noinline int test_btrfs_get_extent(void)
 	offset = em->start + em->len;
 	free_extent_map(em);
 
-	em = btrfs_get_extent(inode, NULL, 0, offset, 4096, 0);
+	em = btrfs_get_extent(inode, NULL, 0, offset, sectorsize, 0);
 	if (IS_ERR(em)) {
 		test_msg("Got an error when we shouldn't have\n");
 		goto out;
@@ -430,9 +437,10 @@ static noinline int test_btrfs_get_extent(void)
 		test_msg("Expected a hole, got %llu\n", em->block_start);
 		goto out;
 	}
-	if (em->start != offset || em->len != 4096) {
-		test_msg("Unexpected extent wanted start %llu len 4096, got "
-			 "start %llu len %llu\n", offset, em->start, em->len);
+	if (em->start != offset || em->len != sectorsize) {
+		test_msg("Unexpected extent wanted start %llu len %u, "
+			"got start %llu len %llu\n",
+			offset, sectorsize, em->start, em->len);
 		goto out;
 	}
 	if (em->flags != 0) {
@@ -442,7 +450,7 @@ static noinline int test_btrfs_get_extent(void)
 	offset = em->start + em->len;
 	free_extent_map(em);
 
-	em = btrfs_get_extent(inode, NULL, 0, offset, 4096, 0);
+	em = btrfs_get_extent(inode, NULL, 0, offset, sectorsize, 0);
 	if (IS_ERR(em)) {
 		test_msg("Got an error when we shouldn't have\n");
 		goto out;
@@ -451,9 +459,10 @@ static noinline int test_btrfs_get_extent(void)
 		test_msg("Expected a real extent, got %llu\n", em->block_start);
 		goto out;
 	}
-	if (em->start != offset || em->len != 8192) {
-		test_msg("Unexpected extent wanted start %llu len 8192, got "
-			 "start %llu len %llu\n", offset, em->start, em->len);
+	if (em->start != offset || em->len != 2 * sectorsize) {
+		test_msg("Unexpected extent wanted start %llu len %u, "
+			"got start %llu len %llu\n",
+			offset, 2 * sectorsize, em->start, em->len);
 		goto out;
 	}
 	if (em->flags != 0) {
@@ -475,7 +484,7 @@ static noinline int test_btrfs_get_extent(void)
 	free_extent_map(em);
 
 	/* Prealloc extent */
-	em = btrfs_get_extent(inode, NULL, 0, offset, 4096, 0);
+	em = btrfs_get_extent(inode, NULL, 0, offset, sectorsize, 0);
 	if (IS_ERR(em)) {
 		test_msg("Got an error when we shouldn't have\n");
 		goto out;
@@ -484,9 +493,10 @@ static noinline int test_btrfs_get_extent(void)
 		test_msg("Expected a real extent, got %llu\n", em->block_start);
 		goto out;
 	}
-	if (em->start != offset || em->len != 4096) {
-		test_msg("Unexpected extent wanted start %llu len 4096, got "
-			 "start %llu len %llu\n", offset, em->start, em->len);
+	if (em->start != offset || em->len != sectorsize) {
+		test_msg("Unexpected extent wanted start %llu len %u, "
+			"got start %llu len %llu\n",
+			offset, sectorsize, em->start, em->len);
 		goto out;
 	}
 	if (em->flags != prealloc_only) {
@@ -503,7 +513,7 @@ static noinline int test_btrfs_get_extent(void)
 	free_extent_map(em);
 
 	/* The next 3 are a half written prealloc extent */
-	em = btrfs_get_extent(inode, NULL, 0, offset, 4096, 0);
+	em = btrfs_get_extent(inode, NULL, 0, offset, sectorsize, 0);
 	if (IS_ERR(em)) {
 		test_msg("Got an error when we shouldn't have\n");
 		goto out;
@@ -512,9 +522,10 @@ static noinline int test_btrfs_get_extent(void)
 		test_msg("Expected a real extent, got %llu\n", em->block_start);
 		goto out;
 	}
-	if (em->start != offset || em->len != 4096) {
-		test_msg("Unexpected extent wanted start %llu len 4096, got "
-			 "start %llu len %llu\n", offset, em->start, em->len);
+	if (em->start != offset || em->len != sectorsize) {
+		test_msg("Unexpected extent wanted start %llu len %u, "
+			"got start %llu len %llu\n",
+			offset, sectorsize, em->start, em->len);
 		goto out;
 	}
 	if (em->flags != prealloc_only) {
@@ -532,7 +543,7 @@ static noinline int test_btrfs_get_extent(void)
 	offset = em->start + em->len;
 	free_extent_map(em);
 
-	em = btrfs_get_extent(inode, NULL, 0, offset, 4096, 0);
+	em = btrfs_get_extent(inode, NULL, 0, offset, sectorsize, 0);
 	if (IS_ERR(em)) {
 		test_msg("Got an error when we shouldn't have\n");
 		goto out;
@@ -541,9 +552,10 @@ static noinline int test_btrfs_get_extent(void)
 		test_msg("Expected a real extent, got %llu\n", em->block_start);
 		goto out;
 	}
-	if (em->start != offset || em->len != 4096) {
-		test_msg("Unexpected extent wanted start %llu len 4096, got "
-			 "start %llu len %llu\n", offset, em->start, em->len);
+	if (em->start != offset || em->len != sectorsize) {
+		test_msg("Unexpected extent wanted start %llu len %u, "
+			"got start %llu len %llu\n",
+			offset, sectorsize, em->start, em->len);
 		goto out;
 	}
 	if (em->flags != 0) {
@@ -564,7 +576,7 @@ static noinline int test_btrfs_get_extent(void)
 	offset = em->start + em->len;
 	free_extent_map(em);
 
-	em = btrfs_get_extent(inode, NULL, 0, offset, 4096, 0);
+	em = btrfs_get_extent(inode, NULL, 0, offset, sectorsize, 0);
 	if (IS_ERR(em)) {
 		test_msg("Got an error when we shouldn't have\n");
 		goto out;
@@ -573,9 +585,10 @@ static noinline int test_btrfs_get_extent(void)
 		test_msg("Expected a real extent, got %llu\n", em->block_start);
 		goto out;
 	}
-	if (em->start != offset || em->len != 8192) {
-		test_msg("Unexpected extent wanted start %llu len 8192, got "
-			 "start %llu len %llu\n", offset, em->start, em->len);
+	if (em->start != offset || em->len != 2 * sectorsize) {
+		test_msg("Unexpected extent wanted start %llu len %u, "
+			"got start %llu len %llu\n",
+			offset, 2 * sectorsize, em->start, em->len);
 		goto out;
 	}
 	if (em->flags != prealloc_only) {
@@ -598,7 +611,7 @@ static noinline int test_btrfs_get_extent(void)
 	free_extent_map(em);
 
 	/* Now for the compressed extent */
-	em = btrfs_get_extent(inode, NULL, 0, offset, 4096, 0);
+	em = btrfs_get_extent(inode, NULL, 0, offset, sectorsize, 0);
 	if (IS_ERR(em)) {
 		test_msg("Got an error when we shouldn't have\n");
 		goto out;
@@ -607,9 +620,10 @@ static noinline int test_btrfs_get_extent(void)
 		test_msg("Expected a real extent, got %llu\n", em->block_start);
 		goto out;
 	}
-	if (em->start != offset || em->len != 8192) {
-		test_msg("Unexpected extent wanted start %llu len 8192, got "
-			 "start %llu len %llu\n", offset, em->start, em->len);
+	if (em->start != offset || em->len != 2 * sectorsize) {
+		test_msg("Unexpected extent wanted start %llu len %u,"
+			"got start %llu len %llu\n",
+			offset, 2 * sectorsize, em->start, em->len);
 		goto out;
 	}
 	if (em->flags != compressed_only) {
@@ -631,7 +645,7 @@ static noinline int test_btrfs_get_extent(void)
 	free_extent_map(em);
 
 	/* Split compressed extent */
-	em = btrfs_get_extent(inode, NULL, 0, offset, 4096, 0);
+	em = btrfs_get_extent(inode, NULL, 0, offset, sectorsize, 0);
 	if (IS_ERR(em)) {
 		test_msg("Got an error when we shouldn't have\n");
 		goto out;
@@ -640,9 +654,10 @@ static noinline int test_btrfs_get_extent(void)
 		test_msg("Expected a real extent, got %llu\n", em->block_start);
 		goto out;
 	}
-	if (em->start != offset || em->len != 4096) {
-		test_msg("Unexpected extent wanted start %llu len 4096, got "
-			 "start %llu len %llu\n", offset, em->start, em->len);
+	if (em->start != offset || em->len != sectorsize) {
+		test_msg("Unexpected extent wanted start %llu len %u,"
+			"got start %llu len %llu\n",
+			offset, sectorsize, em->start, em->len);
 		goto out;
 	}
 	if (em->flags != compressed_only) {
@@ -665,7 +680,7 @@ static noinline int test_btrfs_get_extent(void)
 	offset = em->start + em->len;
 	free_extent_map(em);
 
-	em = btrfs_get_extent(inode, NULL, 0, offset, 4096, 0);
+	em = btrfs_get_extent(inode, NULL, 0, offset, sectorsize, 0);
 	if (IS_ERR(em)) {
 		test_msg("Got an error when we shouldn't have\n");
 		goto out;
@@ -674,9 +689,10 @@ static noinline int test_btrfs_get_extent(void)
 		test_msg("Expected a real extent, got %llu\n", em->block_start);
 		goto out;
 	}
-	if (em->start != offset || em->len != 4096) {
-		test_msg("Unexpected extent wanted start %llu len 4096, got "
-			 "start %llu len %llu\n", offset, em->start, em->len);
+	if (em->start != offset || em->len != sectorsize) {
+		test_msg("Unexpected extent wanted start %llu len %u, "
+			"got start %llu len %llu\n",
+			offset, sectorsize, em->start, em->len);
 		goto out;
 	}
 	if (em->flags != 0) {
@@ -691,7 +707,7 @@ static noinline int test_btrfs_get_extent(void)
 	offset = em->start + em->len;
 	free_extent_map(em);
 
-	em = btrfs_get_extent(inode, NULL, 0, offset, 4096, 0);
+	em = btrfs_get_extent(inode, NULL, 0, offset, sectorsize, 0);
 	if (IS_ERR(em)) {
 		test_msg("Got an error when we shouldn't have\n");
 		goto out;
@@ -701,9 +717,10 @@ static noinline int test_btrfs_get_extent(void)
 			 disk_bytenr, em->block_start);
 		goto out;
 	}
-	if (em->start != offset || em->len != 8192) {
-		test_msg("Unexpected extent wanted start %llu len 8192, got "
-			 "start %llu len %llu\n", offset, em->start, em->len);
+	if (em->start != offset || em->len != 2 * sectorsize) {
+		test_msg("Unexpected extent wanted start %llu len %u, "
+			"got start %llu len %llu\n",
+			offset, 2 * sectorsize, em->start, em->len);
 		goto out;
 	}
 	if (em->flags != compressed_only) {
@@ -725,7 +742,7 @@ static noinline int test_btrfs_get_extent(void)
 	free_extent_map(em);
 
 	/* A hole between regular extents but no hole extent */
-	em = btrfs_get_extent(inode, NULL, 0, offset + 6, 4096, 0);
+	em = btrfs_get_extent(inode, NULL, 0, offset + 6, sectorsize, 0);
 	if (IS_ERR(em)) {
 		test_msg("Got an error when we shouldn't have\n");
 		goto out;
@@ -734,9 +751,10 @@ static noinline int test_btrfs_get_extent(void)
 		test_msg("Expected a real extent, got %llu\n", em->block_start);
 		goto out;
 	}
-	if (em->start != offset || em->len != 4096) {
-		test_msg("Unexpected extent wanted start %llu len 4096, got "
-			 "start %llu len %llu\n", offset, em->start, em->len);
+	if (em->start != offset || em->len != sectorsize) {
+		test_msg("Unexpected extent wanted start %llu len %u, "
+			"got start %llu len %llu\n",
+			offset, sectorsize, em->start, em->len);
 		goto out;
 	}
 	if (em->flags != 0) {
@@ -765,9 +783,10 @@ static noinline int test_btrfs_get_extent(void)
 	 * length of the actual hole, if this changes we'll have to change this
 	 * test.
 	 */
-	if (em->start != offset || em->len != 12288) {
-		test_msg("Unexpected extent wanted start %llu len 12288, got "
-			 "start %llu len %llu\n", offset, em->start, em->len);
+	if (em->start != offset || em->len != 3 * sectorsize) {
+		test_msg("Unexpected extent wanted start %llu len %u, "
+			"got start %llu len %llu\n",
+			offset, 3 * sectorsize, em->start, em->len);
 		goto out;
 	}
 	if (em->flags != vacancy_only) {
@@ -783,7 +802,7 @@ static noinline int test_btrfs_get_extent(void)
 	offset = em->start + em->len;
 	free_extent_map(em);
 
-	em = btrfs_get_extent(inode, NULL, 0, offset, 4096, 0);
+	em = btrfs_get_extent(inode, NULL, 0, offset, sectorsize, 0);
 	if (IS_ERR(em)) {
 		test_msg("Got an error when we shouldn't have\n");
 		goto out;
@@ -792,9 +811,10 @@ static noinline int test_btrfs_get_extent(void)
 		test_msg("Expected a real extent, got %llu\n", em->block_start);
 		goto out;
 	}
-	if (em->start != offset || em->len != 4096) {
-		test_msg("Unexpected extent wanted start %llu len 4096, got "
-			 "start %llu len %llu\n", offset, em->start, em->len);
+	if (em->start != offset || em->len != sectorsize) {
+		test_msg("Unexpected extent wanted start %llu len %u,"
+			"got start %llu len %llu\n",
+			offset, sectorsize, em->start, em->len);
 		goto out;
 	}
 	if (em->flags != 0) {
@@ -812,11 +832,13 @@ out:
 		free_extent_map(em);
 	iput(inode);
 	btrfs_free_dummy_root(root);
+	btrfs_free_dummy_fs_info(fs_info);
 	return ret;
 }
 
-static int test_hole_first(void)
+static int test_hole_first(u32 sectorsize, u32 nodesize)
 {
+	struct btrfs_fs_info *fs_info = NULL;
 	struct inode *inode = NULL;
 	struct btrfs_root *root = NULL;
 	struct extent_map *em = NULL;
@@ -832,19 +854,19 @@ static int test_hole_first(void)
 	BTRFS_I(inode)->location.objectid = BTRFS_FIRST_FREE_OBJECTID;
 	BTRFS_I(inode)->location.offset = 0;
 
-	root = btrfs_alloc_dummy_root();
-	if (IS_ERR(root)) {
-		test_msg("Couldn't allocate root\n");
+	fs_info = btrfs_alloc_dummy_fs_info();
+	if (!fs_info) {
+		test_msg("Couldn't allocate dummy fs info\n");
 		goto out;
 	}
 
-	root->fs_info = btrfs_alloc_dummy_fs_info();
-	if (!root->fs_info) {
-		test_msg("Couldn't allocate dummy fs info\n");
+	root = btrfs_alloc_dummy_root(fs_info, sectorsize, nodesize);
+	if (IS_ERR(root)) {
+		test_msg("Couldn't allocate root\n");
 		goto out;
 	}
 
-	root->node = alloc_dummy_extent_buffer(NULL, 4096);
+	root->node = alloc_dummy_extent_buffer(NULL, nodesize, nodesize);
 	if (!root->node) {
 		test_msg("Couldn't allocate dummy buffer\n");
 		goto out;
@@ -861,9 +883,9 @@ static int test_hole_first(void)
 	 * btrfs_get_extent.
 	 */
 	insert_inode_item_key(root);
-	insert_extent(root, 4096, 4096, 4096, 0, 4096, 4096,
-		      BTRFS_FILE_EXTENT_REG, 0, 1);
-	em = btrfs_get_extent(inode, NULL, 0, 0, 8192, 0);
+	insert_extent(root, sectorsize, sectorsize, sectorsize, 0, sectorsize,
+		      sectorsize, BTRFS_FILE_EXTENT_REG, 0, 1);
+	em = btrfs_get_extent(inode, NULL, 0, 0, 2 * sectorsize, 0);
 	if (IS_ERR(em)) {
 		test_msg("Got an error when we shouldn't have\n");
 		goto out;
@@ -872,9 +894,10 @@ static int test_hole_first(void)
 		test_msg("Expected a hole, got %llu\n", em->block_start);
 		goto out;
 	}
-	if (em->start != 0 || em->len != 4096) {
-		test_msg("Unexpected extent wanted start 0 len 4096, got start "
-			 "%llu len %llu\n", em->start, em->len);
+	if (em->start != 0 || em->len != sectorsize) {
+		test_msg("Unexpected extent wanted start 0 len %u, "
+			"got start %llu len %llu\n",
+			sectorsize, em->start, em->len);
 		goto out;
 	}
 	if (em->flags != vacancy_only) {
@@ -884,18 +907,19 @@ static int test_hole_first(void)
 	}
 	free_extent_map(em);
 
-	em = btrfs_get_extent(inode, NULL, 0, 4096, 8192, 0);
+	em = btrfs_get_extent(inode, NULL, 0, sectorsize, 2 * sectorsize, 0);
 	if (IS_ERR(em)) {
 		test_msg("Got an error when we shouldn't have\n");
 		goto out;
 	}
-	if (em->block_start != 4096) {
+	if (em->block_start != sectorsize) {
 		test_msg("Expected a real extent, got %llu\n", em->block_start);
 		goto out;
 	}
-	if (em->start != 4096 || em->len != 4096) {
-		test_msg("Unexpected extent wanted start 4096 len 4096, got "
-			 "start %llu len %llu\n", em->start, em->len);
+	if (em->start != sectorsize || em->len != sectorsize) {
+		test_msg("Unexpected extent wanted start %u len %u, "
+			"got start %llu len %llu\n",
+			sectorsize, sectorsize, em->start, em->len);
 		goto out;
 	}
 	if (em->flags != 0) {
@@ -909,11 +933,13 @@ out:
 		free_extent_map(em);
 	iput(inode);
 	btrfs_free_dummy_root(root);
+	btrfs_free_dummy_fs_info(fs_info);
 	return ret;
 }
 
-static int test_extent_accounting(void)
+static int test_extent_accounting(u32 sectorsize, u32 nodesize)
 {
+	struct btrfs_fs_info *fs_info = NULL;
 	struct inode *inode = NULL;
 	struct btrfs_root *root = NULL;
 	int ret = -ENOMEM;
@@ -924,15 +950,15 @@ static int test_extent_accounting(void)
 		return ret;
 	}
 
-	root = btrfs_alloc_dummy_root();
-	if (IS_ERR(root)) {
-		test_msg("Couldn't allocate root\n");
+	fs_info = btrfs_alloc_dummy_fs_info();
+	if (!fs_info) {
+		test_msg("Couldn't allocate dummy fs info\n");
 		goto out;
 	}
 
-	root->fs_info = btrfs_alloc_dummy_fs_info();
-	if (!root->fs_info) {
-		test_msg("Couldn't allocate dummy fs info\n");
+	root = btrfs_alloc_dummy_root(fs_info, sectorsize, nodesize);
+	if (IS_ERR(root)) {
+		test_msg("Couldn't allocate root\n");
 		goto out;
 	}
 
@@ -942,7 +968,7 @@ static int test_extent_accounting(void)
 	/* [BTRFS_MAX_EXTENT_SIZE] */
 	BTRFS_I(inode)->outstanding_extents++;
 	ret = btrfs_set_extent_delalloc(inode, 0, BTRFS_MAX_EXTENT_SIZE - 1,
-					NULL);
+					NULL, 0);
 	if (ret) {
 		test_msg("btrfs_set_extent_delalloc returned %d\n", ret);
 		goto out;
@@ -954,10 +980,11 @@ static int test_extent_accounting(void)
 		goto out;
 	}
 
-	/* [BTRFS_MAX_EXTENT_SIZE][4k] */
+	/* [BTRFS_MAX_EXTENT_SIZE][sectorsize] */
 	BTRFS_I(inode)->outstanding_extents++;
 	ret = btrfs_set_extent_delalloc(inode, BTRFS_MAX_EXTENT_SIZE,
-					BTRFS_MAX_EXTENT_SIZE + 4095, NULL);
+					BTRFS_MAX_EXTENT_SIZE + sectorsize - 1,
+					NULL, 0);
 	if (ret) {
 		test_msg("btrfs_set_extent_delalloc returned %d\n", ret);
 		goto out;
@@ -969,10 +996,10 @@ static int test_extent_accounting(void)
 		goto out;
 	}
 
-	/* [BTRFS_MAX_EXTENT_SIZE/2][4K HOLE][the rest] */
+	/* [BTRFS_MAX_EXTENT_SIZE/2][sectorsize HOLE][the rest] */
 	ret = clear_extent_bit(&BTRFS_I(inode)->io_tree,
 			       BTRFS_MAX_EXTENT_SIZE >> 1,
-			       (BTRFS_MAX_EXTENT_SIZE >> 1) + 4095,
+			       (BTRFS_MAX_EXTENT_SIZE >> 1) + sectorsize - 1,
 			       EXTENT_DELALLOC | EXTENT_DIRTY |
 			       EXTENT_UPTODATE | EXTENT_DO_ACCOUNTING, 0, 0,
 			       NULL, GFP_KERNEL);
@@ -987,11 +1014,12 @@ static int test_extent_accounting(void)
 		goto out;
 	}
 
-	/* [BTRFS_MAX_EXTENT_SIZE][4K] */
+	/* [BTRFS_MAX_EXTENT_SIZE][sectorsize] */
 	BTRFS_I(inode)->outstanding_extents++;
 	ret = btrfs_set_extent_delalloc(inode, BTRFS_MAX_EXTENT_SIZE >> 1,
-					(BTRFS_MAX_EXTENT_SIZE >> 1) + 4095,
-					NULL);
+					(BTRFS_MAX_EXTENT_SIZE >> 1)
+					+ sectorsize - 1,
+					NULL, 0);
 	if (ret) {
 		test_msg("btrfs_set_extent_delalloc returned %d\n", ret);
 		goto out;
@@ -1004,16 +1032,17 @@ static int test_extent_accounting(void)
 	}
 
 	/*
-	 * [BTRFS_MAX_EXTENT_SIZE+4K][4K HOLE][BTRFS_MAX_EXTENT_SIZE+4K]
+	 * [BTRFS_MAX_EXTENT_SIZE+sectorsize][sectorsize HOLE][BTRFS_MAX_EXTENT_SIZE+sectorsize]
 	 *
 	 * I'm artificially adding 2 to outstanding_extents because in the
 	 * buffered IO case we'd add things up as we go, but I don't feel like
 	 * doing that here, this isn't the interesting case we want to test.
 	 */
 	BTRFS_I(inode)->outstanding_extents += 2;
-	ret = btrfs_set_extent_delalloc(inode, BTRFS_MAX_EXTENT_SIZE + 8192,
-					(BTRFS_MAX_EXTENT_SIZE << 1) + 12287,
-					NULL);
+	ret = btrfs_set_extent_delalloc(inode,
+			BTRFS_MAX_EXTENT_SIZE + 2 * sectorsize,
+			(BTRFS_MAX_EXTENT_SIZE << 1) + 3 * sectorsize - 1,
+			NULL, 0);
 	if (ret) {
 		test_msg("btrfs_set_extent_delalloc returned %d\n", ret);
 		goto out;
@@ -1025,10 +1054,13 @@ static int test_extent_accounting(void)
 		goto out;
 	}
 
-	/* [BTRFS_MAX_EXTENT_SIZE+4k][4k][BTRFS_MAX_EXTENT_SIZE+4k] */
+	/*
+	* [BTRFS_MAX_EXTENT_SIZE+sectorsize][sectorsize][BTRFS_MAX_EXTENT_SIZE+sectorsize]
+	*/
 	BTRFS_I(inode)->outstanding_extents++;
-	ret = btrfs_set_extent_delalloc(inode, BTRFS_MAX_EXTENT_SIZE+4096,
-					BTRFS_MAX_EXTENT_SIZE+8191, NULL);
+	ret = btrfs_set_extent_delalloc(inode,
+			BTRFS_MAX_EXTENT_SIZE + sectorsize,
+			BTRFS_MAX_EXTENT_SIZE + 2 * sectorsize - 1, NULL, 0);
 	if (ret) {
 		test_msg("btrfs_set_extent_delalloc returned %d\n", ret);
 		goto out;
@@ -1042,8 +1074,8 @@ static int test_extent_accounting(void)
 
 	/* [BTRFS_MAX_EXTENT_SIZE+4k][4K HOLE][BTRFS_MAX_EXTENT_SIZE+4k] */
 	ret = clear_extent_bit(&BTRFS_I(inode)->io_tree,
-			       BTRFS_MAX_EXTENT_SIZE+4096,
-			       BTRFS_MAX_EXTENT_SIZE+8191,
+			       BTRFS_MAX_EXTENT_SIZE + sectorsize,
+			       BTRFS_MAX_EXTENT_SIZE + 2 * sectorsize - 1,
 			       EXTENT_DIRTY | EXTENT_DELALLOC |
 			       EXTENT_DO_ACCOUNTING | EXTENT_UPTODATE, 0, 0,
 			       NULL, GFP_KERNEL);
@@ -1063,8 +1095,9 @@ static int test_extent_accounting(void)
 	 * might fail and I'd rather satisfy my paranoia at this point.
 	 */
 	BTRFS_I(inode)->outstanding_extents++;
-	ret = btrfs_set_extent_delalloc(inode, BTRFS_MAX_EXTENT_SIZE+4096,
-					BTRFS_MAX_EXTENT_SIZE+8191, NULL);
+	ret = btrfs_set_extent_delalloc(inode,
+			BTRFS_MAX_EXTENT_SIZE + sectorsize,
+			BTRFS_MAX_EXTENT_SIZE + 2 * sectorsize - 1, NULL, 0);
 	if (ret) {
 		test_msg("btrfs_set_extent_delalloc returned %d\n", ret);
 		goto out;
@@ -1100,10 +1133,11 @@ out:
 				 NULL, GFP_KERNEL);
 	iput(inode);
 	btrfs_free_dummy_root(root);
+	btrfs_free_dummy_fs_info(fs_info);
 	return ret;
 }
 
-int btrfs_test_inodes(void)
+int btrfs_test_inodes(u32 sectorsize, u32 nodesize)
 {
 	int ret;
 
@@ -1112,13 +1146,13 @@ int btrfs_test_inodes(void)
 	set_bit(EXTENT_FLAG_PREALLOC, &prealloc_only);
 
 	test_msg("Running btrfs_get_extent tests\n");
-	ret = test_btrfs_get_extent();
+	ret = test_btrfs_get_extent(sectorsize, nodesize);
 	if (ret)
 		return ret;
 	test_msg("Running hole first btrfs_get_extent test\n");
-	ret = test_hole_first();
+	ret = test_hole_first(sectorsize, nodesize);
 	if (ret)
 		return ret;
 	test_msg("Running outstanding_extents tests\n");
-	return test_extent_accounting();
+	return test_extent_accounting(sectorsize, nodesize);
 }
diff --git a/fs/btrfs/tests/qgroup-tests.c b/fs/btrfs/tests/qgroup-tests.c
index 8aa4ded31326..ca7cb5e6d385 100644
--- a/fs/btrfs/tests/qgroup-tests.c
+++ b/fs/btrfs/tests/qgroup-tests.c
@@ -16,6 +16,7 @@
  * Boston, MA 021110-1307, USA.
  */
 
+#include <linux/types.h>
 #include "btrfs-tests.h"
 #include "../ctree.h"
 #include "../transaction.h"
@@ -216,7 +217,8 @@ static int remove_extent_ref(struct btrfs_root *root, u64 bytenr,
 	return ret;
 }
 
-static int test_no_shared_qgroup(struct btrfs_root *root)
+static int test_no_shared_qgroup(struct btrfs_root *root,
+		u32 sectorsize, u32 nodesize)
 {
 	struct btrfs_trans_handle trans;
 	struct btrfs_fs_info *fs_info = root->fs_info;
@@ -227,7 +229,7 @@ static int test_no_shared_qgroup(struct btrfs_root *root)
 	btrfs_init_dummy_trans(&trans);
 
 	test_msg("Qgroup basic add\n");
-	ret = btrfs_create_qgroup(NULL, fs_info, 5);
+	ret = btrfs_create_qgroup(NULL, fs_info, BTRFS_FS_TREE_OBJECTID);
 	if (ret) {
 		test_msg("Couldn't create a qgroup %d\n", ret);
 		return ret;
@@ -238,18 +240,19 @@ static int test_no_shared_qgroup(struct btrfs_root *root)
 	 * we can only call btrfs_qgroup_account_extent() directly to test
 	 * quota.
 	 */
-	ret = btrfs_find_all_roots(&trans, fs_info, 4096, 0, &old_roots);
+	ret = btrfs_find_all_roots(&trans, fs_info, nodesize, 0, &old_roots);
 	if (ret) {
 		ulist_free(old_roots);
 		test_msg("Couldn't find old roots: %d\n", ret);
 		return ret;
 	}
 
-	ret = insert_normal_tree_ref(root, 4096, 4096, 0, 5);
+	ret = insert_normal_tree_ref(root, nodesize, nodesize, 0,
+				BTRFS_FS_TREE_OBJECTID);
 	if (ret)
 		return ret;
 
-	ret = btrfs_find_all_roots(&trans, fs_info, 4096, 0, &new_roots);
+	ret = btrfs_find_all_roots(&trans, fs_info, nodesize, 0, &new_roots);
 	if (ret) {
 		ulist_free(old_roots);
 		ulist_free(new_roots);
@@ -257,32 +260,33 @@ static int test_no_shared_qgroup(struct btrfs_root *root)
 		return ret;
 	}
 
-	ret = btrfs_qgroup_account_extent(&trans, fs_info, 4096, 4096,
-					  old_roots, new_roots);
+	ret = btrfs_qgroup_account_extent(&trans, fs_info, nodesize,
+					  nodesize, old_roots, new_roots);
 	if (ret) {
 		test_msg("Couldn't account space for a qgroup %d\n", ret);
 		return ret;
 	}
 
-	if (btrfs_verify_qgroup_counts(fs_info, 5, 4096, 4096)) {
+	if (btrfs_verify_qgroup_counts(fs_info, BTRFS_FS_TREE_OBJECTID,
+				nodesize, nodesize)) {
 		test_msg("Qgroup counts didn't match expected values\n");
 		return -EINVAL;
 	}
 	old_roots = NULL;
 	new_roots = NULL;
 
-	ret = btrfs_find_all_roots(&trans, fs_info, 4096, 0, &old_roots);
+	ret = btrfs_find_all_roots(&trans, fs_info, nodesize, 0, &old_roots);
 	if (ret) {
 		ulist_free(old_roots);
 		test_msg("Couldn't find old roots: %d\n", ret);
 		return ret;
 	}
 
-	ret = remove_extent_item(root, 4096, 4096);
+	ret = remove_extent_item(root, nodesize, nodesize);
 	if (ret)
 		return -EINVAL;
 
-	ret = btrfs_find_all_roots(&trans, fs_info, 4096, 0, &new_roots);
+	ret = btrfs_find_all_roots(&trans, fs_info, nodesize, 0, &new_roots);
 	if (ret) {
 		ulist_free(old_roots);
 		ulist_free(new_roots);
@@ -290,14 +294,14 @@ static int test_no_shared_qgroup(struct btrfs_root *root)
 		return ret;
 	}
 
-	ret = btrfs_qgroup_account_extent(&trans, fs_info, 4096, 4096,
-					  old_roots, new_roots);
+	ret = btrfs_qgroup_account_extent(&trans, fs_info, nodesize,
+					  nodesize, old_roots, new_roots);
 	if (ret) {
 		test_msg("Couldn't account space for a qgroup %d\n", ret);
 		return -EINVAL;
 	}
 
-	if (btrfs_verify_qgroup_counts(fs_info, 5, 0, 0)) {
+	if (btrfs_verify_qgroup_counts(fs_info, BTRFS_FS_TREE_OBJECTID, 0, 0)) {
 		test_msg("Qgroup counts didn't match expected values\n");
 		return -EINVAL;
 	}
@@ -310,7 +314,8 @@ static int test_no_shared_qgroup(struct btrfs_root *root)
  * right, also remove one of the roots and make sure the exclusive count is
  * adjusted properly.
  */
-static int test_multiple_refs(struct btrfs_root *root)
+static int test_multiple_refs(struct btrfs_root *root,
+		u32 sectorsize, u32 nodesize)
 {
 	struct btrfs_trans_handle trans;
 	struct btrfs_fs_info *fs_info = root->fs_info;
@@ -322,25 +327,29 @@ static int test_multiple_refs(struct btrfs_root *root)
 
 	test_msg("Qgroup multiple refs test\n");
 
-	/* We have 5 created already from the previous test */
-	ret = btrfs_create_qgroup(NULL, fs_info, 256);
+	/*
+	 * We have BTRFS_FS_TREE_OBJECTID created already from the
+	 * previous test.
+	 */
+	ret = btrfs_create_qgroup(NULL, fs_info, BTRFS_FIRST_FREE_OBJECTID);
 	if (ret) {
 		test_msg("Couldn't create a qgroup %d\n", ret);
 		return ret;
 	}
 
-	ret = btrfs_find_all_roots(&trans, fs_info, 4096, 0, &old_roots);
+	ret = btrfs_find_all_roots(&trans, fs_info, nodesize, 0, &old_roots);
 	if (ret) {
 		ulist_free(old_roots);
 		test_msg("Couldn't find old roots: %d\n", ret);
 		return ret;
 	}
 
-	ret = insert_normal_tree_ref(root, 4096, 4096, 0, 5);
+	ret = insert_normal_tree_ref(root, nodesize, nodesize, 0,
+				BTRFS_FS_TREE_OBJECTID);
 	if (ret)
 		return ret;
 
-	ret = btrfs_find_all_roots(&trans, fs_info, 4096, 0, &new_roots);
+	ret = btrfs_find_all_roots(&trans, fs_info, nodesize, 0, &new_roots);
 	if (ret) {
 		ulist_free(old_roots);
 		ulist_free(new_roots);
@@ -348,30 +357,32 @@ static int test_multiple_refs(struct btrfs_root *root)
 		return ret;
 	}
 
-	ret = btrfs_qgroup_account_extent(&trans, fs_info, 4096, 4096,
-					  old_roots, new_roots);
+	ret = btrfs_qgroup_account_extent(&trans, fs_info, nodesize,
+					  nodesize, old_roots, new_roots);
 	if (ret) {
 		test_msg("Couldn't account space for a qgroup %d\n", ret);
 		return ret;
 	}
 
-	if (btrfs_verify_qgroup_counts(fs_info, 5, 4096, 4096)) {
+	if (btrfs_verify_qgroup_counts(fs_info, BTRFS_FS_TREE_OBJECTID,
+				       nodesize, nodesize)) {
 		test_msg("Qgroup counts didn't match expected values\n");
 		return -EINVAL;
 	}
 
-	ret = btrfs_find_all_roots(&trans, fs_info, 4096, 0, &old_roots);
+	ret = btrfs_find_all_roots(&trans, fs_info, nodesize, 0, &old_roots);
 	if (ret) {
 		ulist_free(old_roots);
 		test_msg("Couldn't find old roots: %d\n", ret);
 		return ret;
 	}
 
-	ret = add_tree_ref(root, 4096, 4096, 0, 256);
+	ret = add_tree_ref(root, nodesize, nodesize, 0,
+			BTRFS_FIRST_FREE_OBJECTID);
 	if (ret)
 		return ret;
 
-	ret = btrfs_find_all_roots(&trans, fs_info, 4096, 0, &new_roots);
+	ret = btrfs_find_all_roots(&trans, fs_info, nodesize, 0, &new_roots);
 	if (ret) {
 		ulist_free(old_roots);
 		ulist_free(new_roots);
@@ -379,35 +390,38 @@ static int test_multiple_refs(struct btrfs_root *root)
 		return ret;
 	}
 
-	ret = btrfs_qgroup_account_extent(&trans, fs_info, 4096, 4096,
-					  old_roots, new_roots);
+	ret = btrfs_qgroup_account_extent(&trans, fs_info, nodesize,
+					  nodesize, old_roots, new_roots);
 	if (ret) {
 		test_msg("Couldn't account space for a qgroup %d\n", ret);
 		return ret;
 	}
 
-	if (btrfs_verify_qgroup_counts(fs_info, 5, 4096, 0)) {
+	if (btrfs_verify_qgroup_counts(fs_info, BTRFS_FS_TREE_OBJECTID,
+					nodesize, 0)) {
 		test_msg("Qgroup counts didn't match expected values\n");
 		return -EINVAL;
 	}
 
-	if (btrfs_verify_qgroup_counts(fs_info, 256, 4096, 0)) {
+	if (btrfs_verify_qgroup_counts(fs_info, BTRFS_FIRST_FREE_OBJECTID,
+					nodesize, 0)) {
 		test_msg("Qgroup counts didn't match expected values\n");
 		return -EINVAL;
 	}
 
-	ret = btrfs_find_all_roots(&trans, fs_info, 4096, 0, &old_roots);
+	ret = btrfs_find_all_roots(&trans, fs_info, nodesize, 0, &old_roots);
 	if (ret) {
 		ulist_free(old_roots);
 		test_msg("Couldn't find old roots: %d\n", ret);
 		return ret;
 	}
 
-	ret = remove_extent_ref(root, 4096, 4096, 0, 256);
+	ret = remove_extent_ref(root, nodesize, nodesize, 0,
+				BTRFS_FIRST_FREE_OBJECTID);
 	if (ret)
 		return ret;
 
-	ret = btrfs_find_all_roots(&trans, fs_info, 4096, 0, &new_roots);
+	ret = btrfs_find_all_roots(&trans, fs_info, nodesize, 0, &new_roots);
 	if (ret) {
 		ulist_free(old_roots);
 		ulist_free(new_roots);
@@ -415,19 +429,21 @@ static int test_multiple_refs(struct btrfs_root *root)
 		return ret;
 	}
 
-	ret = btrfs_qgroup_account_extent(&trans, fs_info, 4096, 4096,
-					  old_roots, new_roots);
+	ret = btrfs_qgroup_account_extent(&trans, fs_info, nodesize,
+					  nodesize, old_roots, new_roots);
 	if (ret) {
 		test_msg("Couldn't account space for a qgroup %d\n", ret);
 		return ret;
 	}
 
-	if (btrfs_verify_qgroup_counts(fs_info, 256, 0, 0)) {
+	if (btrfs_verify_qgroup_counts(fs_info, BTRFS_FIRST_FREE_OBJECTID,
+					0, 0)) {
 		test_msg("Qgroup counts didn't match expected values\n");
 		return -EINVAL;
 	}
 
-	if (btrfs_verify_qgroup_counts(fs_info, 5, 4096, 4096)) {
+	if (btrfs_verify_qgroup_counts(fs_info, BTRFS_FS_TREE_OBJECTID,
+					nodesize, nodesize)) {
 		test_msg("Qgroup counts didn't match expected values\n");
 		return -EINVAL;
 	}
@@ -435,24 +451,26 @@ static int test_multiple_refs(struct btrfs_root *root)
 	return 0;
 }
 
-int btrfs_test_qgroups(void)
+int btrfs_test_qgroups(u32 sectorsize, u32 nodesize)
 {
+	struct btrfs_fs_info *fs_info = NULL;
 	struct btrfs_root *root;
 	struct btrfs_root *tmp_root;
 	int ret = 0;
 
-	root = btrfs_alloc_dummy_root();
-	if (IS_ERR(root)) {
-		test_msg("Couldn't allocate root\n");
-		return PTR_ERR(root);
+	fs_info = btrfs_alloc_dummy_fs_info();
+	if (!fs_info) {
+		test_msg("Couldn't allocate dummy fs info\n");
+		return -ENOMEM;
 	}
 
-	root->fs_info = btrfs_alloc_dummy_fs_info();
-	if (!root->fs_info) {
-		test_msg("Couldn't allocate dummy fs info\n");
-		ret = -ENOMEM;
+	root = btrfs_alloc_dummy_root(fs_info, sectorsize, nodesize);
+	if (IS_ERR(root)) {
+		test_msg("Couldn't allocate root\n");
+		ret = PTR_ERR(root);
 		goto out;
 	}
+
 	/* We are using this root as our extent root */
 	root->fs_info->extent_root = root;
 
@@ -462,13 +480,14 @@ int btrfs_test_qgroups(void)
 	 */
 	root->fs_info->tree_root = root;
 	root->fs_info->quota_root = root;
-	root->fs_info->quota_enabled = 1;
+	set_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags);
 
 	/*
 	 * Can't use bytenr 0, some things freak out
 	 * *cough*backref walking code*cough*
 	 */
-	root->node = alloc_test_extent_buffer(root->fs_info, 4096);
+	root->node = alloc_test_extent_buffer(root->fs_info, nodesize,
+					nodesize);
 	if (!root->node) {
 		test_msg("Couldn't allocate dummy buffer\n");
 		ret = -ENOMEM;
@@ -476,16 +495,16 @@ int btrfs_test_qgroups(void)
 	}
 	btrfs_set_header_level(root->node, 0);
 	btrfs_set_header_nritems(root->node, 0);
-	root->alloc_bytenr += 8192;
+	root->alloc_bytenr += 2 * nodesize;
 
-	tmp_root = btrfs_alloc_dummy_root();
+	tmp_root = btrfs_alloc_dummy_root(fs_info, sectorsize, nodesize);
 	if (IS_ERR(tmp_root)) {
 		test_msg("Couldn't allocate a fs root\n");
 		ret = PTR_ERR(tmp_root);
 		goto out;
 	}
 
-	tmp_root->root_key.objectid = 5;
+	tmp_root->root_key.objectid = BTRFS_FS_TREE_OBJECTID;
 	root->fs_info->fs_root = tmp_root;
 	ret = btrfs_insert_fs_root(root->fs_info, tmp_root);
 	if (ret) {
@@ -493,14 +512,14 @@ int btrfs_test_qgroups(void)
 		goto out;
 	}
 
-	tmp_root = btrfs_alloc_dummy_root();
+	tmp_root = btrfs_alloc_dummy_root(fs_info, sectorsize, nodesize);
 	if (IS_ERR(tmp_root)) {
 		test_msg("Couldn't allocate a fs root\n");
 		ret = PTR_ERR(tmp_root);
 		goto out;
 	}
 
-	tmp_root->root_key.objectid = 256;
+	tmp_root->root_key.objectid = BTRFS_FIRST_FREE_OBJECTID;
 	ret = btrfs_insert_fs_root(root->fs_info, tmp_root);
 	if (ret) {
 		test_msg("Couldn't insert fs root %d\n", ret);
@@ -508,11 +527,12 @@ int btrfs_test_qgroups(void)
 	}
 
 	test_msg("Running qgroup tests\n");
-	ret = test_no_shared_qgroup(root);
+	ret = test_no_shared_qgroup(root, sectorsize, nodesize);
 	if (ret)
 		goto out;
-	ret = test_multiple_refs(root);
+	ret = test_multiple_refs(root, sectorsize, nodesize);
 out:
 	btrfs_free_dummy_root(root);
+	btrfs_free_dummy_fs_info(fs_info);
 	return ret;
 }
diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c
index f6e24cb423ae..9517de0e668c 100644
--- a/fs/btrfs/transaction.c
+++ b/fs/btrfs/transaction.c
@@ -65,8 +65,9 @@ void btrfs_put_transaction(struct btrfs_transaction *transaction)
 		BUG_ON(!list_empty(&transaction->list));
 		WARN_ON(!RB_EMPTY_ROOT(&transaction->delayed_refs.href_root));
 		if (transaction->delayed_refs.pending_csums)
-			printk(KERN_ERR "pending csums is %llu\n",
-			       transaction->delayed_refs.pending_csums);
+			btrfs_err(transaction->fs_info,
+				  "pending csums is %llu",
+				  transaction->delayed_refs.pending_csums);
 		while (!list_empty(&transaction->pending_chunks)) {
 			struct extent_map *em;
 
@@ -245,6 +246,7 @@ loop:
 		return -EROFS;
 	}
 
+	cur_trans->fs_info = fs_info;
 	atomic_set(&cur_trans->num_writers, 1);
 	extwriter_counter_init(cur_trans, type);
 	init_waitqueue_head(&cur_trans->writer_wait);
@@ -272,11 +274,9 @@ loop:
 	 */
 	smp_mb();
 	if (!list_empty(&fs_info->tree_mod_seq_list))
-		WARN(1, KERN_ERR "BTRFS: tree_mod_seq_list not empty when "
-			"creating a fresh transaction\n");
+		WARN(1, KERN_ERR "BTRFS: tree_mod_seq_list not empty when creating a fresh transaction\n");
 	if (!RB_EMPTY_ROOT(&fs_info->tree_mod_log))
-		WARN(1, KERN_ERR "BTRFS: tree_mod_log rb tree not empty when "
-			"creating a fresh transaction\n");
+		WARN(1, KERN_ERR "BTRFS: tree_mod_log rb tree not empty when creating a fresh transaction\n");
 	atomic64_set(&fs_info->tree_mod_seq, 0);
 
 	spin_lock_init(&cur_trans->delayed_refs.lock);
@@ -441,7 +441,7 @@ static void wait_current_trans(struct btrfs_root *root)
 
 static int may_wait_transaction(struct btrfs_root *root, int type)
 {
-	if (root->fs_info->log_root_recovering)
+	if (test_bit(BTRFS_FS_LOG_RECOVERING, &root->fs_info->flags))
 		return 0;
 
 	if (type == TRANS_USERSPACE)
@@ -549,11 +549,8 @@ again:
 		}
 	} while (ret == -EBUSY);
 
-	if (ret < 0) {
-		/* We must get the transaction if we are JOIN_NOLOCK. */
-		BUG_ON(type == TRANS_JOIN_NOLOCK);
+	if (ret < 0)
 		goto join_fail;
-	}
 
 	cur_trans = root->fs_info->running_transaction;
 
@@ -561,6 +558,7 @@ again:
 	h->transaction = cur_trans;
 	h->root = root;
 	h->use_count = 1;
+	h->fs_info = root->fs_info;
 
 	h->type = type;
 	h->can_flush_pending_bgs = true;
@@ -818,6 +816,7 @@ static int __btrfs_end_transaction(struct btrfs_trans_handle *trans,
 {
 	struct btrfs_transaction *cur_trans = trans->transaction;
 	struct btrfs_fs_info *info = root->fs_info;
+	u64 transid = trans->transid;
 	unsigned long cur = trans->delayed_ref_updates;
 	int lock = (trans->type != TRANS_JOIN_NOLOCK);
 	int err = 0;
@@ -905,7 +904,7 @@ static int __btrfs_end_transaction(struct btrfs_trans_handle *trans,
 
 	kmem_cache_free(btrfs_trans_handle_cachep, trans);
 	if (must_run_delayed_refs) {
-		btrfs_async_run_delayed_refs(root, cur,
+		btrfs_async_run_delayed_refs(root, cur, transid,
 					     must_run_delayed_refs == 1);
 	}
 	return err;
@@ -991,7 +990,6 @@ int btrfs_wait_marked_extents(struct btrfs_root *root,
 	struct extent_state *cached_state = NULL;
 	u64 start = 0;
 	u64 end;
-	struct btrfs_inode *btree_ino = BTRFS_I(root->fs_info->btree_inode);
 	bool errors = false;
 
 	while (!find_first_extent_bit(dirty_pages, start, &start, &end,
@@ -1023,17 +1021,17 @@ int btrfs_wait_marked_extents(struct btrfs_root *root,
 
 	if (root->root_key.objectid == BTRFS_TREE_LOG_OBJECTID) {
 		if ((mark & EXTENT_DIRTY) &&
-		    test_and_clear_bit(BTRFS_INODE_BTREE_LOG1_ERR,
-				       &btree_ino->runtime_flags))
+		    test_and_clear_bit(BTRFS_FS_LOG1_ERR,
+				       &root->fs_info->flags))
 			errors = true;
 
 		if ((mark & EXTENT_NEW) &&
-		    test_and_clear_bit(BTRFS_INODE_BTREE_LOG2_ERR,
-				       &btree_ino->runtime_flags))
+		    test_and_clear_bit(BTRFS_FS_LOG2_ERR,
+				       &root->fs_info->flags))
 			errors = true;
 	} else {
-		if (test_and_clear_bit(BTRFS_INODE_BTREE_ERR,
-				       &btree_ino->runtime_flags))
+		if (test_and_clear_bit(BTRFS_FS_BTREE_ERR,
+				       &root->fs_info->flags))
 			errors = true;
 	}
 
@@ -1298,11 +1296,11 @@ int btrfs_defrag_root(struct btrfs_root *root)
 		btrfs_btree_balance_dirty(info->tree_root);
 		cond_resched();
 
-		if (btrfs_fs_closing(root->fs_info) || ret != -EAGAIN)
+		if (btrfs_fs_closing(info) || ret != -EAGAIN)
 			break;
 
-		if (btrfs_defrag_cancelled(root->fs_info)) {
-			pr_debug("BTRFS: defrag_root cancelled\n");
+		if (btrfs_defrag_cancelled(info)) {
+			btrfs_debug(info, "defrag_root cancelled");
 			ret = -EAGAIN;
 			break;
 		}
@@ -1311,11 +1309,6 @@ int btrfs_defrag_root(struct btrfs_root *root)
 	return ret;
 }
 
-/* Bisesctability fixup, remove in 4.8 */
-#ifndef btrfs_std_error
-#define btrfs_std_error btrfs_handle_fs_error
-#endif
-
 /*
  * Do all special snapshot related qgroup dirty hack.
  *
@@ -1338,7 +1331,7 @@ static int qgroup_account_snapshot(struct btrfs_trans_handle *trans,
 	 * kick in anyway.
 	 */
 	mutex_lock(&fs_info->qgroup_ioctl_lock);
-	if (!fs_info->quota_enabled) {
+	if (!test_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags)) {
 		mutex_unlock(&fs_info->qgroup_ioctl_lock);
 		return 0;
 	}
@@ -1385,7 +1378,7 @@ static int qgroup_account_snapshot(struct btrfs_trans_handle *trans,
 	switch_commit_roots(trans->transaction, fs_info);
 	ret = btrfs_write_and_wait_transaction(trans, src);
 	if (ret)
-		btrfs_std_error(fs_info, ret,
+		btrfs_handle_fs_error(fs_info, ret,
 			"Error while writing out transaction for qgroup");
 
 out:
@@ -1477,7 +1470,7 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
 	parent_root = BTRFS_I(parent_inode)->root;
 	record_root_in_trans(trans, parent_root, 0);
 
-	cur_time = current_fs_time(parent_inode->i_sb);
+	cur_time = current_time(parent_inode);
 
 	/*
 	 * insert the directory item
@@ -1495,7 +1488,7 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
 		goto dir_item_existed;
 	} else if (IS_ERR(dir_item)) {
 		ret = PTR_ERR(dir_item);
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 		goto fail;
 	}
 	btrfs_release_path(path);
@@ -1508,7 +1501,7 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
 	 */
 	ret = btrfs_run_delayed_items(trans, root);
 	if (ret) {	/* Transaction aborted */
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 		goto fail;
 	}
 
@@ -1547,7 +1540,7 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
 	if (ret) {
 		btrfs_tree_unlock(old);
 		free_extent_buffer(old);
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 		goto fail;
 	}
 
@@ -1558,7 +1551,7 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
 	btrfs_tree_unlock(old);
 	free_extent_buffer(old);
 	if (ret) {
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 		goto fail;
 	}
 	/* see comments in should_cow_block() */
@@ -1572,7 +1565,7 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
 	btrfs_tree_unlock(tmp);
 	free_extent_buffer(tmp);
 	if (ret) {
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 		goto fail;
 	}
 
@@ -1584,7 +1577,7 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
 				 btrfs_ino(parent_inode), index,
 				 dentry->d_name.name, dentry->d_name.len);
 	if (ret) {
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 		goto fail;
 	}
 
@@ -1592,19 +1585,19 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
 	pending->snap = btrfs_read_fs_root_no_name(root->fs_info, &key);
 	if (IS_ERR(pending->snap)) {
 		ret = PTR_ERR(pending->snap);
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 		goto fail;
 	}
 
 	ret = btrfs_reloc_post_snapshot(trans, pending);
 	if (ret) {
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 		goto fail;
 	}
 
 	ret = btrfs_run_delayed_refs(trans, root, (unsigned long)-1);
 	if (ret) {
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 		goto fail;
 	}
 
@@ -1626,23 +1619,23 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
 	/* We have check then name at the beginning, so it is impossible. */
 	BUG_ON(ret == -EEXIST || ret == -EOVERFLOW);
 	if (ret) {
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 		goto fail;
 	}
 
 	btrfs_i_size_write(parent_inode, parent_inode->i_size +
 					 dentry->d_name.len * 2);
 	parent_inode->i_mtime = parent_inode->i_ctime =
-		current_fs_time(parent_inode->i_sb);
+		current_time(parent_inode);
 	ret = btrfs_update_inode_fallback(trans, parent_root, parent_inode);
 	if (ret) {
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 		goto fail;
 	}
 	ret = btrfs_uuid_tree_add(trans, fs_info->uuid_root, new_uuid.b,
 				  BTRFS_UUID_KEY_SUBVOL, objectid);
 	if (ret) {
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 		goto fail;
 	}
 	if (!btrfs_is_empty_uuid(new_root_item->received_uuid)) {
@@ -1651,14 +1644,14 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
 					  BTRFS_UUID_KEY_RECEIVED_SUBVOL,
 					  objectid);
 		if (ret && ret != -EEXIST) {
-			btrfs_abort_transaction(trans, root, ret);
+			btrfs_abort_transaction(trans, ret);
 			goto fail;
 		}
 	}
 
 	ret = btrfs_run_delayed_refs(trans, root, (unsigned long)-1);
 	if (ret) {
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 		goto fail;
 	}
 
@@ -1713,9 +1706,9 @@ static void update_super_roots(struct btrfs_root *root)
 	super->root = root_item->bytenr;
 	super->generation = root_item->generation;
 	super->root_level = root_item->level;
-	if (btrfs_test_opt(root, SPACE_CACHE))
+	if (btrfs_test_opt(root->fs_info, SPACE_CACHE))
 		super->cache_generation = root_item->generation;
-	if (root->fs_info->update_uuid_tree_gen)
+	if (test_bit(BTRFS_FS_UPDATE_UUID_TREE_GEN, &root->fs_info->flags))
 		super->uuid_tree_generation = root_item->generation;
 }
 
@@ -1854,7 +1847,7 @@ static void cleanup_transaction(struct btrfs_trans_handle *trans,
 
 	WARN_ON(trans->use_count > 1);
 
-	btrfs_abort_transaction(trans, root, err);
+	btrfs_abort_transaction(trans, err);
 
 	spin_lock(&root->fs_info->trans_lock);
 
@@ -1899,14 +1892,14 @@ static void cleanup_transaction(struct btrfs_trans_handle *trans,
 
 static inline int btrfs_start_delalloc_flush(struct btrfs_fs_info *fs_info)
 {
-	if (btrfs_test_opt(fs_info->tree_root, FLUSHONCOMMIT))
+	if (btrfs_test_opt(fs_info, FLUSHONCOMMIT))
 		return btrfs_start_delalloc_roots(fs_info, 1, -1);
 	return 0;
 }
 
 static inline void btrfs_wait_delalloc_flush(struct btrfs_fs_info *fs_info)
 {
-	if (btrfs_test_opt(fs_info->tree_root, FLUSHONCOMMIT))
+	if (btrfs_test_opt(fs_info, FLUSHONCOMMIT))
 		btrfs_wait_ordered_roots(fs_info, -1, 0, (u64)-1);
 }
 
@@ -1922,7 +1915,6 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans,
 {
 	struct btrfs_transaction *cur_trans = trans->transaction;
 	struct btrfs_transaction *prev_trans = NULL;
-	struct btrfs_inode *btree_ino = BTRFS_I(root->fs_info->btree_inode);
 	int ret;
 
 	/* Stop the commit early if ->aborted is set */
@@ -2216,8 +2208,8 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans,
 	btrfs_update_commit_device_size(root->fs_info);
 	btrfs_update_commit_device_bytes_used(root, cur_trans);
 
-	clear_bit(BTRFS_INODE_BTREE_LOG1_ERR, &btree_ino->runtime_flags);
-	clear_bit(BTRFS_INODE_BTREE_LOG2_ERR, &btree_ino->runtime_flags);
+	clear_bit(BTRFS_FS_LOG1_ERR, &root->fs_info->flags);
+	clear_bit(BTRFS_FS_LOG2_ERR, &root->fs_info->flags);
 
 	btrfs_trans_release_chunk_metadata(trans);
 
@@ -2281,8 +2273,13 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans,
 
 	kmem_cache_free(btrfs_trans_handle_cachep, trans);
 
+	/*
+	 * If fs has been frozen, we can not handle delayed iputs, otherwise
+	 * it'll result in deadlock about SB_FREEZE_FS.
+	 */
 	if (current != root->fs_info->transaction_kthread &&
-	    current != root->fs_info->cleaner_kthread)
+	    current != root->fs_info->cleaner_kthread &&
+	    !root->fs_info->fs_frozen)
 		btrfs_run_delayed_iputs(root);
 
 	return ret;
@@ -2326,7 +2323,7 @@ int btrfs_clean_one_deleted_snapshot(struct btrfs_root *root)
 	list_del_init(&root->root_list);
 	spin_unlock(&fs_info->trans_lock);
 
-	pr_debug("BTRFS: cleaner removing %llu\n", root->objectid);
+	btrfs_debug(fs_info, "cleaner removing %llu", root->objectid);
 
 	btrfs_kill_all_delayed_nodes(root);
 
diff --git a/fs/btrfs/transaction.h b/fs/btrfs/transaction.h
index 9fe0ec2bf0fe..6cf0d37d4f76 100644
--- a/fs/btrfs/transaction.h
+++ b/fs/btrfs/transaction.h
@@ -82,6 +82,7 @@ struct btrfs_transaction {
 	spinlock_t dropped_roots_lock;
 	struct btrfs_delayed_ref_root delayed_refs;
 	int aborted;
+	struct btrfs_fs_info *fs_info;
 };
 
 #define __TRANS_FREEZABLE	(1U << 0)
@@ -110,7 +111,6 @@ struct btrfs_trans_handle {
 	u64 chunk_bytes_reserved;
 	unsigned long use_count;
 	unsigned long blocks_reserved;
-	unsigned long blocks_used;
 	unsigned long delayed_ref_updates;
 	struct btrfs_transaction *transaction;
 	struct btrfs_block_rsv *block_rsv;
@@ -121,6 +121,7 @@ struct btrfs_trans_handle {
 	bool can_flush_pending_bgs;
 	bool reloc_reserved;
 	bool sync;
+	bool dirty;
 	unsigned int type;
 	/*
 	 * this root is only needed to validate that the root passed to
@@ -128,6 +129,7 @@ struct btrfs_trans_handle {
 	 * Subvolume quota depends on this
 	 */
 	struct btrfs_root *root;
+	struct btrfs_fs_info *fs_info;
 	struct seq_list delayed_ref_elem;
 	struct list_head qgroup_ref_list;
 	struct list_head new_bgs;
diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c
index b7665af471d8..528cae123dc9 100644
--- a/fs/btrfs/tree-log.c
+++ b/fs/btrfs/tree-log.c
@@ -27,6 +27,7 @@
 #include "backref.h"
 #include "hash.h"
 #include "compression.h"
+#include "qgroup.h"
 
 /* magic values for the inode_only field in btrfs_log_inode:
  *
@@ -680,6 +681,21 @@ static noinline int replay_one_extent(struct btrfs_trans_handle *trans,
 		ins.type = BTRFS_EXTENT_ITEM_KEY;
 		offset = key->offset - btrfs_file_extent_offset(eb, item);
 
+		/*
+		 * Manually record dirty extent, as here we did a shallow
+		 * file extent item copy and skip normal backref update,
+		 * but modifying extent tree all by ourselves.
+		 * So need to manually record dirty extent for qgroup,
+		 * as the owner of the file extent changed from log tree
+		 * (doesn't affect qgroup) to fs/file tree(affects qgroup)
+		 */
+		ret = btrfs_qgroup_insert_dirty_extent(trans, root->fs_info,
+				btrfs_file_extent_disk_bytenr(eb, item),
+				btrfs_file_extent_disk_num_bytes(eb, item),
+				GFP_NOFS);
+		if (ret < 0)
+			goto out;
+
 		if (ins.objectid > 0) {
 			u64 csum_start;
 			u64 csum_end;
@@ -2422,8 +2438,8 @@ static noinline int walk_down_log_tree(struct btrfs_trans_handle *trans,
 		root_owner = btrfs_header_owner(parent);
 
 		next = btrfs_find_create_tree_block(root, bytenr);
-		if (!next)
-			return -ENOMEM;
+		if (IS_ERR(next))
+			return PTR_ERR(next);
 
 		if (*level == 1) {
 			ret = wc->process_func(root, next, wc, ptr_gen);
@@ -2757,7 +2773,7 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans,
 	while (1) {
 		int batch = atomic_read(&root->log_batch);
 		/* when we're on an ssd, just kick the log commit out */
-		if (!btrfs_test_opt(root, SSD) &&
+		if (!btrfs_test_opt(root->fs_info, SSD) &&
 		    test_bit(BTRFS_ROOT_MULTI_LOG_TASKS, &root->state)) {
 			mutex_unlock(&root->log_mutex);
 			schedule_timeout_uninterruptible(1);
@@ -2788,7 +2804,7 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans,
 	ret = btrfs_write_marked_extents(log, &log->dirty_log_pages, mark);
 	if (ret) {
 		blk_finish_plug(&plug);
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 		btrfs_free_logged_extents(log, log_transid);
 		btrfs_set_log_full_commit(root->fs_info, trans);
 		mutex_unlock(&root->log_mutex);
@@ -2807,7 +2823,7 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans,
 	 */
 	mutex_unlock(&root->log_mutex);
 
-	btrfs_init_log_ctx(&root_log_ctx);
+	btrfs_init_log_ctx(&root_log_ctx, NULL);
 
 	mutex_lock(&log_root_tree->log_mutex);
 	atomic_inc(&log_root_tree->log_batch);
@@ -2838,7 +2854,7 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans,
 		btrfs_set_log_full_commit(root->fs_info, trans);
 
 		if (ret != -ENOSPC) {
-			btrfs_abort_transaction(trans, root, ret);
+			btrfs_abort_transaction(trans, ret);
 			mutex_unlock(&log_root_tree->log_mutex);
 			goto out;
 		}
@@ -2851,6 +2867,7 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans,
 
 	if (log_root_tree->log_transid_committed >= root_log_ctx.log_transid) {
 		blk_finish_plug(&plug);
+		list_del_init(&root_log_ctx.list);
 		mutex_unlock(&log_root_tree->log_mutex);
 		ret = root_log_ctx.log_ret;
 		goto out;
@@ -2898,7 +2915,7 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans,
 	blk_finish_plug(&plug);
 	if (ret) {
 		btrfs_set_log_full_commit(root->fs_info, trans);
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 		btrfs_free_logged_extents(log, log_transid);
 		mutex_unlock(&log_root_tree->log_mutex);
 		goto out_wake_log_root;
@@ -2934,7 +2951,7 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans,
 	ret = write_ctree_super(trans, root->fs_info->tree_root, 1);
 	if (ret) {
 		btrfs_set_log_full_commit(root->fs_info, trans);
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 		goto out_wake_log_root;
 	}
 
@@ -2991,7 +3008,7 @@ static void free_log_tree(struct btrfs_trans_handle *trans,
 	ret = walk_log_tree(trans, log, &wc);
 	/* I don't think this can happen but just in case */
 	if (ret)
-		btrfs_abort_transaction(trans, log, ret);
+		btrfs_abort_transaction(trans, ret);
 
 	while (1) {
 		ret = find_first_extent_bit(&log->dirty_log_pages,
@@ -3160,7 +3177,7 @@ out_unlock:
 		btrfs_set_log_full_commit(root->fs_info, trans);
 		ret = 0;
 	} else if (ret < 0)
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 
 	btrfs_end_log_trans(root);
 
@@ -3193,7 +3210,7 @@ int btrfs_del_inode_ref_in_log(struct btrfs_trans_handle *trans,
 		btrfs_set_log_full_commit(root->fs_info, trans);
 		ret = 0;
 	} else if (ret < 0 && ret != -ENOENT)
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 	btrfs_end_log_trans(root);
 
 	return ret;
@@ -3944,7 +3961,7 @@ static int wait_ordered_extents(struct btrfs_trans_handle *trans,
 			 * i_mapping flags, so that the next fsync won't get
 			 * an outdated io error too.
 			 */
-			btrfs_inode_check_errors(inode);
+			filemap_check_errors(inode->i_mapping);
 			*ordered_io_error = true;
 			break;
 		}
@@ -4181,7 +4198,7 @@ static int btrfs_log_changed_extents(struct btrfs_trans_handle *trans,
 	 * without writing to the log tree and the fsync must report the
 	 * file data write error and not commit the current transaction.
 	 */
-	ret = btrfs_inode_check_errors(inode);
+	ret = filemap_check_errors(inode->i_mapping);
 	if (ret)
 		ctx->io_err = ret;
 process:
@@ -4469,7 +4486,8 @@ static int btrfs_log_trailing_hole(struct btrfs_trans_handle *trans,
 static int btrfs_check_ref_name_override(struct extent_buffer *eb,
 					 const int slot,
 					 const struct btrfs_key *key,
-					 struct inode *inode)
+					 struct inode *inode,
+					 u64 *other_ino)
 {
 	int ret;
 	struct btrfs_path *search_path;
@@ -4528,7 +4546,16 @@ static int btrfs_check_ref_name_override(struct extent_buffer *eb,
 					   search_path, parent,
 					   name, this_name_len, 0);
 		if (di && !IS_ERR(di)) {
-			ret = 1;
+			struct btrfs_key di_key;
+
+			btrfs_dir_item_key_to_cpu(search_path->nodes[0],
+						  di, &di_key);
+			if (di_key.type == BTRFS_INODE_ITEM_KEY) {
+				ret = 1;
+				*other_ino = di_key.objectid;
+			} else {
+				ret = -EAGAIN;
+			}
 			goto out;
 		} else if (IS_ERR(di)) {
 			ret = PTR_ERR(di);
@@ -4703,6 +4730,10 @@ static int btrfs_log_inode(struct btrfs_trans_handle *trans,
 		ins_nr = 0;
 		ret = btrfs_search_forward(root, &min_key,
 					   path, trans->transid);
+		if (ret < 0) {
+			err = ret;
+			goto out_unlock;
+		}
 		if (ret != 0)
 			break;
 again:
@@ -4718,16 +4749,72 @@ again:
 		if ((min_key.type == BTRFS_INODE_REF_KEY ||
 		     min_key.type == BTRFS_INODE_EXTREF_KEY) &&
 		    BTRFS_I(inode)->generation == trans->transid) {
+			u64 other_ino = 0;
+
 			ret = btrfs_check_ref_name_override(path->nodes[0],
 							    path->slots[0],
-							    &min_key, inode);
+							    &min_key, inode,
+							    &other_ino);
 			if (ret < 0) {
 				err = ret;
 				goto out_unlock;
-			} else if (ret > 0) {
-				err = 1;
-				btrfs_set_log_full_commit(root->fs_info, trans);
-				goto out_unlock;
+			} else if (ret > 0 && ctx &&
+				   other_ino != btrfs_ino(ctx->inode)) {
+				struct btrfs_key inode_key;
+				struct inode *other_inode;
+
+				if (ins_nr > 0) {
+					ins_nr++;
+				} else {
+					ins_nr = 1;
+					ins_start_slot = path->slots[0];
+				}
+				ret = copy_items(trans, inode, dst_path, path,
+						 &last_extent, ins_start_slot,
+						 ins_nr, inode_only,
+						 logged_isize);
+				if (ret < 0) {
+					err = ret;
+					goto out_unlock;
+				}
+				ins_nr = 0;
+				btrfs_release_path(path);
+				inode_key.objectid = other_ino;
+				inode_key.type = BTRFS_INODE_ITEM_KEY;
+				inode_key.offset = 0;
+				other_inode = btrfs_iget(root->fs_info->sb,
+							 &inode_key, root,
+							 NULL);
+				/*
+				 * If the other inode that had a conflicting dir
+				 * entry was deleted in the current transaction,
+				 * we don't need to do more work nor fallback to
+				 * a transaction commit.
+				 */
+				if (IS_ERR(other_inode) &&
+				    PTR_ERR(other_inode) == -ENOENT) {
+					goto next_key;
+				} else if (IS_ERR(other_inode)) {
+					err = PTR_ERR(other_inode);
+					goto out_unlock;
+				}
+				/*
+				 * We are safe logging the other inode without
+				 * acquiring its i_mutex as long as we log with
+				 * the LOG_INODE_EXISTS mode. We're safe against
+				 * concurrent renames of the other inode as well
+				 * because during a rename we pin the log and
+				 * update the log with the new name before we
+				 * unpin it.
+				 */
+				err = btrfs_log_inode(trans, root, other_inode,
+						      LOG_INODE_EXISTS,
+						      0, LLONG_MAX, ctx);
+				iput(other_inode);
+				if (err)
+					goto out_unlock;
+				else
+					goto next_key;
 			}
 		}
 
@@ -4795,7 +4882,7 @@ next_slot:
 			ins_nr = 0;
 		}
 		btrfs_release_path(path);
-
+next_key:
 		if (min_key.offset < (u64)-1) {
 			min_key.offset++;
 		} else if (min_key.type < max_key.type) {
@@ -4989,8 +5076,12 @@ static noinline int check_parent_dirs_for_sync(struct btrfs_trans_handle *trans,
 		if (!parent || d_really_is_negative(parent) || sb != parent->d_sb)
 			break;
 
-		if (IS_ROOT(parent))
+		if (IS_ROOT(parent)) {
+			inode = d_inode(parent);
+			if (btrfs_must_commit_transaction(trans, inode))
+				ret = 1;
 			break;
+		}
 
 		parent = dget_parent(parent);
 		dput(old_parent);
@@ -5301,7 +5392,7 @@ static int btrfs_log_inode_parent(struct btrfs_trans_handle *trans,
 
 	sb = inode->i_sb;
 
-	if (btrfs_test_opt(root, NOTREELOG)) {
+	if (btrfs_test_opt(root->fs_info, NOTREELOG)) {
 		ret = 1;
 		goto end_no_trans;
 	}
@@ -5488,7 +5579,7 @@ int btrfs_recover_log_trees(struct btrfs_root *log_root_tree)
 	if (!path)
 		return -ENOMEM;
 
-	fs_info->log_root_recovering = 1;
+	set_bit(BTRFS_FS_LOG_RECOVERING, &fs_info->flags);
 
 	trans = btrfs_start_transaction(fs_info->tree_root, 0);
 	if (IS_ERR(trans)) {
@@ -5501,8 +5592,8 @@ int btrfs_recover_log_trees(struct btrfs_root *log_root_tree)
 
 	ret = walk_log_tree(trans, log_root_tree, &wc);
 	if (ret) {
-		btrfs_handle_fs_error(fs_info, ret, "Failed to pin buffers while "
-			    "recovering log root tree.");
+		btrfs_handle_fs_error(fs_info, ret,
+			"Failed to pin buffers while recovering log root tree.");
 		goto error;
 	}
 
@@ -5548,8 +5639,8 @@ again:
 			free_extent_buffer(log->node);
 			free_extent_buffer(log->commit_root);
 			kfree(log);
-			btrfs_handle_fs_error(fs_info, ret, "Couldn't read target root "
-				    "for tree log recovery.");
+			btrfs_handle_fs_error(fs_info, ret,
+				"Couldn't read target root for tree log recovery.");
 			goto error;
 		}
 
@@ -5598,7 +5689,7 @@ again:
 
 	free_extent_buffer(log_root_tree->node);
 	log_root_tree->log_root = NULL;
-	fs_info->log_root_recovering = 0;
+	clear_bit(BTRFS_FS_LOG_RECOVERING, &fs_info->flags);
 	kfree(log_root_tree);
 
 	return 0;
diff --git a/fs/btrfs/tree-log.h b/fs/btrfs/tree-log.h
index a9f1b75d080d..ab858e31ccbc 100644
--- a/fs/btrfs/tree-log.h
+++ b/fs/btrfs/tree-log.h
@@ -30,15 +30,18 @@ struct btrfs_log_ctx {
 	int log_transid;
 	int io_err;
 	bool log_new_dentries;
+	struct inode *inode;
 	struct list_head list;
 };
 
-static inline void btrfs_init_log_ctx(struct btrfs_log_ctx *ctx)
+static inline void btrfs_init_log_ctx(struct btrfs_log_ctx *ctx,
+				      struct inode *inode)
 {
 	ctx->log_ret = 0;
 	ctx->log_transid = 0;
 	ctx->io_err = 0;
 	ctx->log_new_dentries = false;
+	ctx->inode = inode;
 	INIT_LIST_HEAD(&ctx->list);
 }
 
diff --git a/fs/btrfs/uuid-tree.c b/fs/btrfs/uuid-tree.c
index 778282944530..7fc89e4adb41 100644
--- a/fs/btrfs/uuid-tree.c
+++ b/fs/btrfs/uuid-tree.c
@@ -69,8 +69,9 @@ static int btrfs_uuid_tree_lookup(struct btrfs_root *uuid_root, u8 *uuid,
 	ret = -ENOENT;
 
 	if (!IS_ALIGNED(item_size, sizeof(u64))) {
-		btrfs_warn(uuid_root->fs_info, "uuid item with illegal size %lu!",
-			(unsigned long)item_size);
+		btrfs_warn(uuid_root->fs_info,
+			   "uuid item with illegal size %lu!",
+			   (unsigned long)item_size);
 		goto out;
 	}
 	while (item_size) {
@@ -137,10 +138,10 @@ int btrfs_uuid_tree_add(struct btrfs_trans_handle *trans,
 		offset = btrfs_item_ptr_offset(eb, slot);
 		offset += btrfs_item_size_nr(eb, slot) - sizeof(subid_le);
 	} else if (ret < 0) {
-		btrfs_warn(uuid_root->fs_info, "insert uuid item failed %d "
-			"(0x%016llx, 0x%016llx) type %u!",
-			ret, (unsigned long long)key.objectid,
-			(unsigned long long)key.offset, type);
+		btrfs_warn(uuid_root->fs_info,
+			   "insert uuid item failed %d (0x%016llx, 0x%016llx) type %u!",
+			   ret, (unsigned long long)key.objectid,
+			   (unsigned long long)key.offset, type);
 		goto out;
 	}
 
@@ -184,8 +185,8 @@ int btrfs_uuid_tree_rem(struct btrfs_trans_handle *trans,
 
 	ret = btrfs_search_slot(trans, uuid_root, &key, path, -1, 1);
 	if (ret < 0) {
-		btrfs_warn(uuid_root->fs_info, "error %d while searching for uuid item!",
-			ret);
+		btrfs_warn(uuid_root->fs_info,
+			   "error %d while searching for uuid item!", ret);
 		goto out;
 	}
 	if (ret > 0) {
@@ -198,8 +199,9 @@ int btrfs_uuid_tree_rem(struct btrfs_trans_handle *trans,
 	offset = btrfs_item_ptr_offset(eb, slot);
 	item_size = btrfs_item_size_nr(eb, slot);
 	if (!IS_ALIGNED(item_size, sizeof(u64))) {
-		btrfs_warn(uuid_root->fs_info, "uuid item with illegal size %lu!",
-			(unsigned long)item_size);
+		btrfs_warn(uuid_root->fs_info,
+			   "uuid item with illegal size %lu!",
+			   (unsigned long)item_size);
 		ret = -ENOENT;
 		goto out;
 	}
@@ -299,8 +301,9 @@ again_search_slot:
 		offset = btrfs_item_ptr_offset(leaf, slot);
 		item_size = btrfs_item_size_nr(leaf, slot);
 		if (!IS_ALIGNED(item_size, sizeof(u64))) {
-			btrfs_warn(fs_info, "uuid item with illegal size %lu!",
-				(unsigned long)item_size);
+			btrfs_warn(fs_info,
+				   "uuid item with illegal size %lu!",
+				   (unsigned long)item_size);
 			goto skip;
 		}
 		while (item_size) {
diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
index bdc62561ede8..71a60cc01451 100644
--- a/fs/btrfs/volumes.c
+++ b/fs/btrfs/volumes.c
@@ -140,7 +140,6 @@ static int btrfs_relocate_sys_chunks(struct btrfs_root *root);
 static void __btrfs_reset_dev_stats(struct btrfs_device *dev);
 static void btrfs_dev_stat_print_on_error(struct btrfs_device *dev);
 static void btrfs_dev_stat_print_on_load(struct btrfs_device *device);
-static void btrfs_close_one_device(struct btrfs_device *device);
 
 DEFINE_MUTEX(uuid_mutex);
 static LIST_HEAD(fs_uuids);
@@ -462,7 +461,7 @@ loop_lock:
 			sync_pending = 0;
 		}
 
-		btrfsic_submit_bio(cur->bi_rw, cur);
+		btrfsic_submit_bio(cur);
 		num_run++;
 		batch_run++;
 
@@ -835,10 +834,6 @@ static void __free_device(struct work_struct *work)
 	struct btrfs_device *device;
 
 	device = container_of(work, struct btrfs_device, rcu_work);
-
-	if (device->bdev)
-		blkdev_put(device->bdev, device->mode);
-
 	rcu_string_free(device->name);
 	kfree(device);
 }
@@ -853,19 +848,81 @@ static void free_device(struct rcu_head *head)
 	schedule_work(&device->rcu_work);
 }
 
+static void btrfs_close_bdev(struct btrfs_device *device)
+{
+	if (device->bdev && device->writeable) {
+		sync_blockdev(device->bdev);
+		invalidate_bdev(device->bdev);
+	}
+
+	if (device->bdev)
+		blkdev_put(device->bdev, device->mode);
+}
+
+static void btrfs_prepare_close_one_device(struct btrfs_device *device)
+{
+	struct btrfs_fs_devices *fs_devices = device->fs_devices;
+	struct btrfs_device *new_device;
+	struct rcu_string *name;
+
+	if (device->bdev)
+		fs_devices->open_devices--;
+
+	if (device->writeable &&
+	    device->devid != BTRFS_DEV_REPLACE_DEVID) {
+		list_del_init(&device->dev_alloc_list);
+		fs_devices->rw_devices--;
+	}
+
+	if (device->missing)
+		fs_devices->missing_devices--;
+
+	new_device = btrfs_alloc_device(NULL, &device->devid,
+					device->uuid);
+	BUG_ON(IS_ERR(new_device)); /* -ENOMEM */
+
+	/* Safe because we are under uuid_mutex */
+	if (device->name) {
+		name = rcu_string_strdup(device->name->str, GFP_NOFS);
+		BUG_ON(!name); /* -ENOMEM */
+		rcu_assign_pointer(new_device->name, name);
+	}
+
+	list_replace_rcu(&device->dev_list, &new_device->dev_list);
+	new_device->fs_devices = device->fs_devices;
+}
+
 static int __btrfs_close_devices(struct btrfs_fs_devices *fs_devices)
 {
 	struct btrfs_device *device, *tmp;
+	struct list_head pending_put;
+
+	INIT_LIST_HEAD(&pending_put);
 
 	if (--fs_devices->opened > 0)
 		return 0;
 
 	mutex_lock(&fs_devices->device_list_mutex);
 	list_for_each_entry_safe(device, tmp, &fs_devices->devices, dev_list) {
-		btrfs_close_one_device(device);
+		btrfs_prepare_close_one_device(device);
+		list_add(&device->dev_list, &pending_put);
 	}
 	mutex_unlock(&fs_devices->device_list_mutex);
 
+	/*
+	 * btrfs_show_devname() is using the device_list_mutex,
+	 * sometimes call to blkdev_put() leads vfs calling
+	 * into this func. So do put outside of device_list_mutex,
+	 * as of now.
+	 */
+	while (!list_empty(&pending_put)) {
+		device = list_first_entry(&pending_put,
+				struct btrfs_device, dev_list);
+		list_del(&device->dev_list);
+		btrfs_close_bdev(device);
+		call_rcu(&device->rcu, free_device);
+	}
+
 	WARN_ON(fs_devices->open_devices);
 	WARN_ON(fs_devices->rw_devices);
 	fs_devices->opened = 0;
@@ -1097,12 +1154,12 @@ int btrfs_scan_one_device(const char *path, fmode_t flags, void *holder,
 	ret = device_list_add(path, disk_super, devid, fs_devices_ret);
 	if (ret > 0) {
 		if (disk_super->label[0]) {
-			printk(KERN_INFO "BTRFS: device label %s ", disk_super->label);
+			pr_info("BTRFS: device label %s ", disk_super->label);
 		} else {
-			printk(KERN_INFO "BTRFS: device fsid %pU ", disk_super->fsid);
+			pr_info("BTRFS: device fsid %pU ", disk_super->fsid);
 		}
 
-		printk(KERN_CONT "devid %llu transid %llu %s\n", devid, transid, path);
+		pr_cont("devid %llu transid %llu %s\n", devid, transid, path);
 		ret = 0;
 	}
 	if (!ret && fs_devices_ret)
@@ -1803,7 +1860,6 @@ int btrfs_rm_device(struct btrfs_root *root, char *device_path, u64 devid)
 	u64 num_devices;
 	int ret = 0;
 	bool clear_super = false;
-	char *dev_name = NULL;
 
 	mutex_lock(&uuid_mutex);
 
@@ -1839,11 +1895,6 @@ int btrfs_rm_device(struct btrfs_root *root, char *device_path, u64 devid)
 		list_del_init(&device->dev_alloc_list);
 		device->fs_devices->rw_devices--;
 		unlock_chunks(root);
-		dev_name = kstrdup(device->name->str, GFP_KERNEL);
-		if (!dev_name) {
-			ret = -ENOMEM;
-			goto error_undo;
-		}
 		clear_super = true;
 	}
 
@@ -1893,12 +1944,21 @@ int btrfs_rm_device(struct btrfs_root *root, char *device_path, u64 devid)
 		btrfs_sysfs_rm_device_link(root->fs_info->fs_devices, device);
 	}
 
-	call_rcu(&device->rcu, free_device);
-
 	num_devices = btrfs_super_num_devices(root->fs_info->super_copy) - 1;
 	btrfs_set_super_num_devices(root->fs_info->super_copy, num_devices);
 	mutex_unlock(&root->fs_info->fs_devices->device_list_mutex);
 
+	/*
+	 * at this point, the device is zero sized and detached from
+	 * the devices list.  All that's left is to zero out the old
+	 * supers and free the device.
+	 */
+	if (device->writeable)
+		btrfs_scratch_superblocks(device->bdev, device->name->str);
+
+	btrfs_close_bdev(device);
+	call_rcu(&device->rcu, free_device);
+
 	if (cur_devices->open_devices == 0) {
 		struct btrfs_fs_devices *fs_devices;
 		fs_devices = root->fs_info->fs_devices;
@@ -1917,24 +1977,7 @@ int btrfs_rm_device(struct btrfs_root *root, char *device_path, u64 devid)
 	root->fs_info->num_tolerated_disk_barrier_failures =
 		btrfs_calc_num_tolerated_disk_barrier_failures(root->fs_info);
 
-	/*
-	 * at this point, the device is zero sized.  We want to
-	 * remove it from the devices list and zero out the old super
-	 */
-	if (clear_super) {
-		struct block_device *bdev;
-
-		bdev = blkdev_get_by_path(dev_name, FMODE_READ | FMODE_EXCL,
-						root->fs_info->bdev_holder);
-		if (!IS_ERR(bdev)) {
-			btrfs_scratch_superblocks(bdev, dev_name);
-			blkdev_put(bdev, FMODE_READ | FMODE_EXCL);
-		}
-	}
-
 out:
-	kfree(dev_name);
-
 	mutex_unlock(&uuid_mutex);
 	return ret;
 
@@ -1986,6 +2029,9 @@ void btrfs_rm_dev_replace_free_srcdev(struct btrfs_fs_info *fs_info,
 		/* zero out the old super if it is writable */
 		btrfs_scratch_superblocks(srcdev->bdev, srcdev->name->str);
 	}
+
+	btrfs_close_bdev(srcdev);
+
 	call_rcu(&srcdev->rcu, free_device);
 
 	/*
@@ -2041,6 +2087,8 @@ void btrfs_destroy_dev_replace_tgtdev(struct btrfs_fs_info *fs_info,
 	 * the device_list_mutex lock.
 	 */
 	btrfs_scratch_superblocks(tgtdev->bdev, tgtdev->name->str);
+
+	btrfs_close_bdev(tgtdev);
 	call_rcu(&tgtdev->rcu, free_device);
 }
 
@@ -2399,14 +2447,14 @@ int btrfs_init_new_device(struct btrfs_root *root, char *device_path)
 		ret = init_first_rw_device(trans, root, device);
 		unlock_chunks(root);
 		if (ret) {
-			btrfs_abort_transaction(trans, root, ret);
+			btrfs_abort_transaction(trans, ret);
 			goto error_trans;
 		}
 	}
 
 	ret = btrfs_add_device(trans, root, device);
 	if (ret) {
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 		goto error_trans;
 	}
 
@@ -2415,7 +2463,7 @@ int btrfs_init_new_device(struct btrfs_root *root, char *device_path)
 
 		ret = btrfs_finish_sprout(trans, root);
 		if (ret) {
-			btrfs_abort_transaction(trans, root, ret);
+			btrfs_abort_transaction(trans, ret);
 			goto error_trans;
 		}
 
@@ -2444,9 +2492,7 @@ int btrfs_init_new_device(struct btrfs_root *root, char *device_path)
 		ret = btrfs_relocate_sys_chunks(root);
 		if (ret < 0)
 			btrfs_handle_fs_error(root->fs_info, ret,
-				    "Failed to relocate sys chunks after "
-				    "device initialization. This can be fixed "
-				    "using the \"btrfs balance\" command.");
+				    "Failed to relocate sys chunks after device initialization. This can be fixed using the \"btrfs balance\" command.");
 		trans = btrfs_attach_transaction(root);
 		if (IS_ERR(trans)) {
 			if (PTR_ERR(trans) == -ENOENT)
@@ -2505,7 +2551,8 @@ int btrfs_init_dev_replace_tgtdev(struct btrfs_root *root, char *device_path,
 	devices = &fs_info->fs_devices->devices;
 	list_for_each_entry(device, devices, dev_list) {
 		if (device->bdev == bdev) {
-			btrfs_err(fs_info, "target device is in the filesystem!");
+			btrfs_err(fs_info,
+				  "target device is in the filesystem!");
 			ret = -EEXIST;
 			goto error;
 		}
@@ -2514,7 +2561,8 @@ int btrfs_init_dev_replace_tgtdev(struct btrfs_root *root, char *device_path,
 
 	if (i_size_read(bdev->bd_inode) <
 	    btrfs_device_get_total_bytes(srcdev)) {
-		btrfs_err(fs_info, "target device is smaller than source device!");
+		btrfs_err(fs_info,
+			  "target device is smaller than source device!");
 		ret = -EINVAL;
 		goto error;
 	}
@@ -2761,6 +2809,7 @@ int btrfs_remove_chunk(struct btrfs_trans_handle *trans,
 	u64 dev_extent_len = 0;
 	u64 chunk_objectid = BTRFS_FIRST_CHUNK_TREE_OBJECTID;
 	int i, ret = 0;
+	struct btrfs_fs_devices *fs_devices = root->fs_info->fs_devices;
 
 	/* Just in case */
 	root = root->fs_info->chunk_root;
@@ -2787,13 +2836,20 @@ int btrfs_remove_chunk(struct btrfs_trans_handle *trans,
 	check_system_chunk(trans, extent_root, map->type);
 	unlock_chunks(root->fs_info->chunk_root);
 
+	/*
+	 * Take the device list mutex to prevent races with the final phase of
+	 * a device replace operation that replaces the device object associated
+	 * with map stripes (dev-replace.c:btrfs_dev_replace_finishing()).
+	 */
+	mutex_lock(&fs_devices->device_list_mutex);
 	for (i = 0; i < map->num_stripes; i++) {
 		struct btrfs_device *device = map->stripes[i].dev;
 		ret = btrfs_free_dev_extent(trans, device,
 					    map->stripes[i].physical,
 					    &dev_extent_len);
 		if (ret) {
-			btrfs_abort_transaction(trans, root, ret);
+			mutex_unlock(&fs_devices->device_list_mutex);
+			btrfs_abort_transaction(trans, ret);
 			goto out;
 		}
 
@@ -2811,14 +2867,17 @@ int btrfs_remove_chunk(struct btrfs_trans_handle *trans,
 		if (map->stripes[i].dev) {
 			ret = btrfs_update_device(trans, map->stripes[i].dev);
 			if (ret) {
-				btrfs_abort_transaction(trans, root, ret);
+				mutex_unlock(&fs_devices->device_list_mutex);
+				btrfs_abort_transaction(trans, ret);
 				goto out;
 			}
 		}
 	}
+	mutex_unlock(&fs_devices->device_list_mutex);
+
 	ret = btrfs_free_chunk(trans, root, chunk_objectid, chunk_offset);
 	if (ret) {
-		btrfs_abort_transaction(trans, root, ret);
+		btrfs_abort_transaction(trans, ret);
 		goto out;
 	}
 
@@ -2827,14 +2886,14 @@ int btrfs_remove_chunk(struct btrfs_trans_handle *trans,
 	if (map->type & BTRFS_BLOCK_GROUP_SYSTEM) {
 		ret = btrfs_del_sys_chunk(root, chunk_objectid, chunk_offset);
 		if (ret) {
-			btrfs_abort_transaction(trans, root, ret);
+			btrfs_abort_transaction(trans, ret);
 			goto out;
 		}
 	}
 
 	ret = btrfs_remove_block_group(trans, extent_root, chunk_offset, em);
 	if (ret) {
-		btrfs_abort_transaction(trans, extent_root, ret);
+		btrfs_abort_transaction(trans, ret);
 		goto out;
 	}
 
@@ -2891,7 +2950,7 @@ static int btrfs_relocate_chunk(struct btrfs_root *root, u64 chunk_offset)
 	 * chunk tree entries
 	 */
 	ret = btrfs_remove_chunk(trans, root, chunk_offset);
-	btrfs_end_transaction(trans, root);
+	btrfs_end_transaction(trans, extent_root);
 	return ret;
 }
 
@@ -3410,7 +3469,7 @@ static int __btrfs_balance(struct btrfs_fs_info *fs_info)
 	u64 size_to_free;
 	u64 chunk_type;
 	struct btrfs_chunk *chunk;
-	struct btrfs_path *path;
+	struct btrfs_path *path = NULL;
 	struct btrfs_key key;
 	struct btrfs_key found_key;
 	struct btrfs_trans_handle *trans;
@@ -3444,13 +3503,33 @@ static int __btrfs_balance(struct btrfs_fs_info *fs_info)
 		ret = btrfs_shrink_device(device, old_size - size_to_free);
 		if (ret == -ENOSPC)
 			break;
-		BUG_ON(ret);
+		if (ret) {
+			/* btrfs_shrink_device never returns ret > 0 */
+			WARN_ON(ret > 0);
+			goto error;
+		}
 
 		trans = btrfs_start_transaction(dev_root, 0);
-		BUG_ON(IS_ERR(trans));
+		if (IS_ERR(trans)) {
+			ret = PTR_ERR(trans);
+			btrfs_info_in_rcu(fs_info,
+		 "resize: unable to start transaction after shrinking device %s (error %d), old size %llu, new size %llu",
+					  rcu_str_deref(device->name), ret,
+					  old_size, old_size - size_to_free);
+			goto error;
+		}
 
 		ret = btrfs_grow_device(trans, device, old_size);
-		BUG_ON(ret);
+		if (ret) {
+			btrfs_end_transaction(trans, dev_root);
+			/* btrfs_grow_device never returns ret > 0 */
+			WARN_ON(ret > 0);
+			btrfs_info_in_rcu(fs_info,
+		 "resize: unable to grow device after shrinking device %s (error %d), old size %llu, new size %llu",
+					  rcu_str_deref(device->name), ret,
+					  old_size, old_size - size_to_free);
+			goto error;
+		}
 
 		btrfs_end_transaction(trans, dev_root);
 	}
@@ -3617,7 +3696,7 @@ error:
 	btrfs_free_path(path);
 	if (enospc_errors) {
 		btrfs_info(fs_info, "%d enospc errors during balance",
-		       enospc_errors);
+			   enospc_errors);
 		if (!ret)
 			ret = -ENOSPC;
 	}
@@ -3711,8 +3790,8 @@ int btrfs_balance(struct btrfs_balance_control *bctl,
 		if (!(bctl->flags & BTRFS_BALANCE_DATA) ||
 		    !(bctl->flags & BTRFS_BALANCE_METADATA) ||
 		    memcmp(&bctl->data, &bctl->meta, sizeof(bctl->data))) {
-			btrfs_err(fs_info, "with mixed groups data and "
-				   "metadata balance options must be the same");
+			btrfs_err(fs_info,
+				  "with mixed groups data and metadata balance options must be the same");
 			ret = -EINVAL;
 			goto out;
 		}
@@ -3734,23 +3813,23 @@ int btrfs_balance(struct btrfs_balance_control *bctl,
 		allowed |= (BTRFS_BLOCK_GROUP_RAID10 |
 			    BTRFS_BLOCK_GROUP_RAID6);
 	if (validate_convert_profile(&bctl->data, allowed)) {
-		btrfs_err(fs_info, "unable to start balance with target "
-			   "data profile %llu",
-		       bctl->data.target);
+		btrfs_err(fs_info,
+			  "unable to start balance with target data profile %llu",
+			  bctl->data.target);
 		ret = -EINVAL;
 		goto out;
 	}
 	if (validate_convert_profile(&bctl->meta, allowed)) {
 		btrfs_err(fs_info,
-			   "unable to start balance with target metadata profile %llu",
-		       bctl->meta.target);
+			  "unable to start balance with target metadata profile %llu",
+			  bctl->meta.target);
 		ret = -EINVAL;
 		goto out;
 	}
 	if (validate_convert_profile(&bctl->sys, allowed)) {
 		btrfs_err(fs_info,
-			   "unable to start balance with target system profile %llu",
-		       bctl->sys.target);
+			  "unable to start balance with target system profile %llu",
+			  bctl->sys.target);
 		ret = -EINVAL;
 		goto out;
 	}
@@ -3770,10 +3849,11 @@ int btrfs_balance(struct btrfs_balance_control *bctl,
 		     (fs_info->avail_metadata_alloc_bits & allowed) &&
 		     !(bctl->meta.target & allowed))) {
 			if (bctl->flags & BTRFS_BALANCE_FORCE) {
-				btrfs_info(fs_info, "force reducing metadata integrity");
+				btrfs_info(fs_info,
+					   "force reducing metadata integrity");
 			} else {
-				btrfs_err(fs_info, "balance will reduce metadata "
-					   "integrity, use force if you want this");
+				btrfs_err(fs_info,
+					  "balance will reduce metadata integrity, use force if you want this");
 				ret = -EINVAL;
 				goto out;
 			}
@@ -3783,8 +3863,8 @@ int btrfs_balance(struct btrfs_balance_control *bctl,
 	if (btrfs_get_num_tolerated_disk_barrier_failures(bctl->meta.target) <
 		btrfs_get_num_tolerated_disk_barrier_failures(bctl->data.target)) {
 		btrfs_warn(fs_info,
-	"metadata profile 0x%llx has lower redundancy than data profile 0x%llx",
-			bctl->meta.target, bctl->data.target);
+			   "metadata profile 0x%llx has lower redundancy than data profile 0x%llx",
+			   bctl->meta.target, bctl->data.target);
 	}
 
 	if (bctl->sys.flags & BTRFS_BALANCE_ARGS_CONVERT) {
@@ -3874,7 +3954,7 @@ int btrfs_resume_balance_async(struct btrfs_fs_info *fs_info)
 	}
 	spin_unlock(&fs_info->balance_lock);
 
-	if (btrfs_test_opt(fs_info->tree_root, SKIP_BALANCE)) {
+	if (btrfs_test_opt(fs_info, SKIP_BALANCE)) {
 		btrfs_info(fs_info, "force skipping balance");
 		return 0;
 	}
@@ -4140,7 +4220,7 @@ out:
 	if (ret)
 		btrfs_warn(fs_info, "btrfs_uuid_scan_kthread failed %d", ret);
 	else
-		fs_info->update_uuid_tree_gen = 1;
+		set_bit(BTRFS_FS_UPDATE_UUID_TREE_GEN, &fs_info->flags);
 	up(&fs_info->uuid_tree_rescan_sem);
 	return 0;
 }
@@ -4229,7 +4309,8 @@ int btrfs_create_uuid_tree(struct btrfs_fs_info *fs_info)
 				      BTRFS_UUID_TREE_OBJECTID);
 	if (IS_ERR(uuid_root)) {
 		ret = PTR_ERR(uuid_root);
-		btrfs_abort_transaction(trans, tree_root, ret);
+		btrfs_abort_transaction(trans, ret);
+		btrfs_end_transaction(trans, tree_root);
 		return ret;
 	}
 
@@ -4502,8 +4583,7 @@ static void check_raid56_incompat_flag(struct btrfs_fs_info *info, u64 type)
 	btrfs_set_fs_incompat(info, RAID56);
 }
 
-#define BTRFS_MAX_DEVS(r) ((BTRFS_LEAF_DATA_SIZE(r)		\
-			- sizeof(struct btrfs_item)		\
+#define BTRFS_MAX_DEVS(r) ((BTRFS_MAX_ITEM_SIZE(r)		\
 			- sizeof(struct btrfs_chunk))		\
 			/ sizeof(struct btrfs_stripe) + 1)
 
@@ -4682,12 +4762,12 @@ static int __btrfs_alloc_chunk(struct btrfs_trans_handle *trans,
 
 	if (type & BTRFS_BLOCK_GROUP_RAID5) {
 		raid_stripe_len = find_raid56_stripe_len(ndevs - 1,
-				 btrfs_super_stripesize(info->super_copy));
+						extent_root->stripesize);
 		data_stripes = num_stripes - 1;
 	}
 	if (type & BTRFS_BLOCK_GROUP_RAID6) {
 		raid_stripe_len = find_raid56_stripe_len(ndevs - 2,
-				 btrfs_super_stripesize(info->super_copy));
+						extent_root->stripesize);
 		data_stripes = num_stripes - 2;
 	}
 
@@ -4832,15 +4912,16 @@ int btrfs_finish_chunk_alloc(struct btrfs_trans_handle *trans,
 	read_unlock(&em_tree->lock);
 
 	if (!em) {
-		btrfs_crit(extent_root->fs_info, "unable to find logical "
-			   "%Lu len %Lu", chunk_offset, chunk_size);
+		btrfs_crit(extent_root->fs_info,
+			   "unable to find logical %Lu len %Lu",
+			   chunk_offset, chunk_size);
 		return -EINVAL;
 	}
 
 	if (em->start != chunk_offset || em->len != chunk_size) {
-		btrfs_crit(extent_root->fs_info, "found a bad mapping, wanted"
-			  " %Lu-%Lu, found %Lu-%Lu", chunk_offset,
-			  chunk_size, em->start, em->len);
+		btrfs_crit(extent_root->fs_info,
+			   "found a bad mapping, wanted %Lu-%Lu, found %Lu-%Lu",
+			    chunk_offset, chunk_size, em->start, em->len);
 		free_extent_map(em);
 		return -EINVAL;
 	}
@@ -5073,9 +5154,9 @@ int btrfs_num_copies(struct btrfs_fs_info *fs_info, u64 logical, u64 len)
 	}
 
 	if (em->start > logical || em->start + em->len < logical) {
-		btrfs_crit(fs_info, "Invalid mapping for %Lu-%Lu, got "
-			    "%Lu-%Lu", logical, logical+len, em->start,
-			    em->start + em->len);
+		btrfs_crit(fs_info, "Invalid mapping for %Lu-%Lu, got %Lu-%Lu",
+			   logical, logical+len, em->start,
+			   em->start + em->len);
 		free_extent_map(em);
 		return 1;
 	}
@@ -5248,7 +5329,7 @@ void btrfs_put_bbio(struct btrfs_bio *bbio)
 		kfree(bbio);
 }
 
-static int __btrfs_map_block(struct btrfs_fs_info *fs_info, int rw,
+static int __btrfs_map_block(struct btrfs_fs_info *fs_info, int op,
 			     u64 logical, u64 *length,
 			     struct btrfs_bio **bbio_ret,
 			     int mirror_num, int need_raid_map)
@@ -5289,9 +5370,9 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info, int rw,
 	}
 
 	if (em->start > logical || em->start + em->len < logical) {
-		btrfs_crit(fs_info, "found a bad mapping, wanted %Lu, "
-			   "found %Lu-%Lu", logical, em->start,
-			   em->start + em->len);
+		btrfs_crit(fs_info,
+			   "found a bad mapping, wanted %Lu, found %Lu-%Lu",
+			   logical, em->start, em->start + em->len);
 		free_extent_map(em);
 		return -EINVAL;
 	}
@@ -5309,9 +5390,8 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info, int rw,
 
 	stripe_offset = stripe_nr * stripe_len;
 	if (offset < stripe_offset) {
-		btrfs_crit(fs_info, "stripe math has gone wrong, "
-			   "stripe_offset=%llu, offset=%llu, start=%llu, "
-			   "logical=%llu, stripe_len=%llu",
+		btrfs_crit(fs_info,
+			   "stripe math has gone wrong, stripe_offset=%llu, offset=%llu, start=%llu, logical=%llu, stripe_len=%llu",
 			   stripe_offset, offset, em->start, logical,
 			   stripe_len);
 		free_extent_map(em);
@@ -5334,7 +5414,7 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info, int rw,
 		raid56_full_stripe_start *= full_stripe_len;
 	}
 
-	if (rw & REQ_DISCARD) {
+	if (op == REQ_OP_DISCARD) {
 		/* we don't discard raid56 yet */
 		if (map->type & BTRFS_BLOCK_GROUP_RAID56_MASK) {
 			ret = -EOPNOTSUPP;
@@ -5347,7 +5427,7 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info, int rw,
 		   For other RAID types and for RAID[56] reads, just allow a single
 		   stripe (on a single disk). */
 		if ((map->type & BTRFS_BLOCK_GROUP_RAID56_MASK) &&
-		    (rw & REQ_WRITE)) {
+		    (op == REQ_OP_WRITE)) {
 			max_len = stripe_len * nr_data_stripes(map) -
 				(offset - raid56_full_stripe_start);
 		} else {
@@ -5372,8 +5452,8 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info, int rw,
 		btrfs_dev_replace_set_lock_blocking(dev_replace);
 
 	if (dev_replace_is_ongoing && mirror_num == map->num_stripes + 1 &&
-	    !(rw & (REQ_WRITE | REQ_DISCARD | REQ_GET_READ_MIRRORS)) &&
-	    dev_replace->tgtdev != NULL) {
+	    op != REQ_OP_WRITE && op != REQ_OP_DISCARD &&
+	    op != REQ_GET_READ_MIRRORS && dev_replace->tgtdev != NULL) {
 		/*
 		 * in dev-replace case, for repair case (that's the only
 		 * case where the mirror is selected explicitly when
@@ -5460,15 +5540,17 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info, int rw,
 			    (offset + *length);
 
 	if (map->type & BTRFS_BLOCK_GROUP_RAID0) {
-		if (rw & REQ_DISCARD)
+		if (op == REQ_OP_DISCARD)
 			num_stripes = min_t(u64, map->num_stripes,
 					    stripe_nr_end - stripe_nr_orig);
 		stripe_nr = div_u64_rem(stripe_nr, map->num_stripes,
 				&stripe_index);
-		if (!(rw & (REQ_WRITE | REQ_DISCARD | REQ_GET_READ_MIRRORS)))
+		if (op != REQ_OP_WRITE && op != REQ_OP_DISCARD &&
+		    op != REQ_GET_READ_MIRRORS)
 			mirror_num = 1;
 	} else if (map->type & BTRFS_BLOCK_GROUP_RAID1) {
-		if (rw & (REQ_WRITE | REQ_DISCARD | REQ_GET_READ_MIRRORS))
+		if (op == REQ_OP_WRITE || op == REQ_OP_DISCARD ||
+		    op == REQ_GET_READ_MIRRORS)
 			num_stripes = map->num_stripes;
 		else if (mirror_num)
 			stripe_index = mirror_num - 1;
@@ -5481,7 +5563,8 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info, int rw,
 		}
 
 	} else if (map->type & BTRFS_BLOCK_GROUP_DUP) {
-		if (rw & (REQ_WRITE | REQ_DISCARD | REQ_GET_READ_MIRRORS)) {
+		if (op == REQ_OP_WRITE || op == REQ_OP_DISCARD ||
+		    op == REQ_GET_READ_MIRRORS) {
 			num_stripes = map->num_stripes;
 		} else if (mirror_num) {
 			stripe_index = mirror_num - 1;
@@ -5495,9 +5578,9 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info, int rw,
 		stripe_nr = div_u64_rem(stripe_nr, factor, &stripe_index);
 		stripe_index *= map->sub_stripes;
 
-		if (rw & (REQ_WRITE | REQ_GET_READ_MIRRORS))
+		if (op == REQ_OP_WRITE || op == REQ_GET_READ_MIRRORS)
 			num_stripes = map->sub_stripes;
-		else if (rw & REQ_DISCARD)
+		else if (op == REQ_OP_DISCARD)
 			num_stripes = min_t(u64, map->sub_stripes *
 					    (stripe_nr_end - stripe_nr_orig),
 					    map->num_stripes);
@@ -5515,7 +5598,7 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info, int rw,
 
 	} else if (map->type & BTRFS_BLOCK_GROUP_RAID56_MASK) {
 		if (need_raid_map &&
-		    ((rw & (REQ_WRITE | REQ_GET_READ_MIRRORS)) ||
+		    (op == REQ_OP_WRITE || op == REQ_GET_READ_MIRRORS ||
 		     mirror_num > 1)) {
 			/* push stripe_nr back to the start of the full stripe */
 			stripe_nr = div_u64(raid56_full_stripe_start,
@@ -5543,8 +5626,8 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info, int rw,
 			/* We distribute the parity blocks across stripes */
 			div_u64_rem(stripe_nr + stripe_index, map->num_stripes,
 					&stripe_index);
-			if (!(rw & (REQ_WRITE | REQ_DISCARD |
-				    REQ_GET_READ_MIRRORS)) && mirror_num <= 1)
+			if ((op != REQ_OP_WRITE && op != REQ_OP_DISCARD &&
+			    op != REQ_GET_READ_MIRRORS) && mirror_num <= 1)
 				mirror_num = 1;
 		}
 	} else {
@@ -5558,8 +5641,8 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info, int rw,
 		mirror_num = stripe_index + 1;
 	}
 	if (stripe_index >= map->num_stripes) {
-		btrfs_crit(fs_info, "stripe index math went horribly wrong, "
-			   "got stripe_index=%u, num_stripes=%u",
+		btrfs_crit(fs_info,
+			   "stripe index math went horribly wrong, got stripe_index=%u, num_stripes=%u",
 			   stripe_index, map->num_stripes);
 		ret = -EINVAL;
 		goto out;
@@ -5567,9 +5650,9 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info, int rw,
 
 	num_alloc_stripes = num_stripes;
 	if (dev_replace_is_ongoing) {
-		if (rw & (REQ_WRITE | REQ_DISCARD))
+		if (op == REQ_OP_WRITE || op == REQ_OP_DISCARD)
 			num_alloc_stripes <<= 1;
-		if (rw & REQ_GET_READ_MIRRORS)
+		if (op == REQ_GET_READ_MIRRORS)
 			num_alloc_stripes++;
 		tgtdev_indexes = num_stripes;
 	}
@@ -5584,7 +5667,8 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info, int rw,
 
 	/* build raid_map */
 	if (map->type & BTRFS_BLOCK_GROUP_RAID56_MASK &&
-	    need_raid_map && ((rw & (REQ_WRITE | REQ_GET_READ_MIRRORS)) ||
+	    need_raid_map &&
+	    ((op == REQ_OP_WRITE || op == REQ_GET_READ_MIRRORS) ||
 	    mirror_num > 1)) {
 		u64 tmp;
 		unsigned rot;
@@ -5609,7 +5693,7 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info, int rw,
 				RAID6_Q_STRIPE;
 	}
 
-	if (rw & REQ_DISCARD) {
+	if (op == REQ_OP_DISCARD) {
 		u32 factor = 0;
 		u32 sub_stripes = 0;
 		u64 stripes_per_dev = 0;
@@ -5689,14 +5773,15 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info, int rw,
 		}
 	}
 
-	if (rw & (REQ_WRITE | REQ_GET_READ_MIRRORS))
+	if (op == REQ_OP_WRITE || op == REQ_GET_READ_MIRRORS)
 		max_errors = btrfs_chunk_max_errors(map);
 
 	if (bbio->raid_map)
 		sort_parity_stripes(bbio, num_stripes);
 
 	tgtdev_indexes = 0;
-	if (dev_replace_is_ongoing && (rw & (REQ_WRITE | REQ_DISCARD)) &&
+	if (dev_replace_is_ongoing &&
+	   (op == REQ_OP_WRITE || op == REQ_OP_DISCARD) &&
 	    dev_replace->tgtdev != NULL) {
 		int index_where_to_add;
 		u64 srcdev_devid = dev_replace->srcdev->devid;
@@ -5731,7 +5816,7 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info, int rw,
 			}
 		}
 		num_stripes = index_where_to_add;
-	} else if (dev_replace_is_ongoing && (rw & REQ_GET_READ_MIRRORS) &&
+	} else if (dev_replace_is_ongoing && (op == REQ_GET_READ_MIRRORS) &&
 		   dev_replace->tgtdev != NULL) {
 		u64 srcdev_devid = dev_replace->srcdev->devid;
 		int index_srcdev = 0;
@@ -5762,20 +5847,17 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info, int rw,
 			}
 		}
 		if (found) {
-			if (physical_of_found + map->stripe_len <=
-			    dev_replace->cursor_left) {
-				struct btrfs_bio_stripe *tgtdev_stripe =
-					bbio->stripes + num_stripes;
+			struct btrfs_bio_stripe *tgtdev_stripe =
+				bbio->stripes + num_stripes;
 
-				tgtdev_stripe->physical = physical_of_found;
-				tgtdev_stripe->length =
-					bbio->stripes[index_srcdev].length;
-				tgtdev_stripe->dev = dev_replace->tgtdev;
-				bbio->tgtdev_map[index_srcdev] = num_stripes;
+			tgtdev_stripe->physical = physical_of_found;
+			tgtdev_stripe->length =
+				bbio->stripes[index_srcdev].length;
+			tgtdev_stripe->dev = dev_replace->tgtdev;
+			bbio->tgtdev_map[index_srcdev] = num_stripes;
 
-				tgtdev_indexes++;
-				num_stripes++;
-			}
+			tgtdev_indexes++;
+			num_stripes++;
 		}
 	}
 
@@ -5806,28 +5888,29 @@ out:
 	return ret;
 }
 
-int btrfs_map_block(struct btrfs_fs_info *fs_info, int rw,
+int btrfs_map_block(struct btrfs_fs_info *fs_info, int op,
 		      u64 logical, u64 *length,
 		      struct btrfs_bio **bbio_ret, int mirror_num)
 {
-	return __btrfs_map_block(fs_info, rw, logical, length, bbio_ret,
+	return __btrfs_map_block(fs_info, op, logical, length, bbio_ret,
 				 mirror_num, 0);
 }
 
 /* For Scrub/replace */
-int btrfs_map_sblock(struct btrfs_fs_info *fs_info, int rw,
+int btrfs_map_sblock(struct btrfs_fs_info *fs_info, int op,
 		     u64 logical, u64 *length,
 		     struct btrfs_bio **bbio_ret, int mirror_num,
 		     int need_raid_map)
 {
-	return __btrfs_map_block(fs_info, rw, logical, length, bbio_ret,
+	return __btrfs_map_block(fs_info, op, logical, length, bbio_ret,
 				 mirror_num, need_raid_map);
 }
 
-int btrfs_rmap_block(struct btrfs_mapping_tree *map_tree,
+int btrfs_rmap_block(struct btrfs_fs_info *fs_info,
 		     u64 chunk_start, u64 physical, u64 devid,
 		     u64 **logical, int *naddrs, int *stripe_len)
 {
+	struct btrfs_mapping_tree *map_tree = &fs_info->mapping_tree;
 	struct extent_map_tree *em_tree = &map_tree->map_tree;
 	struct extent_map *em;
 	struct map_lookup *map;
@@ -5843,13 +5926,13 @@ int btrfs_rmap_block(struct btrfs_mapping_tree *map_tree,
 	read_unlock(&em_tree->lock);
 
 	if (!em) {
-		printk(KERN_ERR "BTRFS: couldn't find em for chunk %Lu\n",
-		       chunk_start);
+		btrfs_err(fs_info, "couldn't find em for chunk %Lu",
+			chunk_start);
 		return -EIO;
 	}
 
 	if (em->start != chunk_start) {
-		printk(KERN_ERR "BTRFS: bad chunk start, em=%Lu, wanted=%Lu\n",
+		btrfs_err(fs_info, "bad chunk start, em=%Lu, wanted=%Lu",
 		       em->start, chunk_start);
 		free_extent_map(em);
 		return -EIO;
@@ -5934,13 +6017,13 @@ static void btrfs_end_bio(struct bio *bio)
 			BUG_ON(stripe_index >= bbio->num_stripes);
 			dev = bbio->stripes[stripe_index].dev;
 			if (dev->bdev) {
-				if (bio->bi_rw & WRITE)
+				if (bio_op(bio) == REQ_OP_WRITE)
 					btrfs_dev_stat_inc(dev,
 						BTRFS_DEV_STAT_WRITE_ERRS);
 				else
 					btrfs_dev_stat_inc(dev,
 						BTRFS_DEV_STAT_READ_ERRS);
-				if ((bio->bi_rw & WRITE_FLUSH) == WRITE_FLUSH)
+				if ((bio->bi_opf & WRITE_FLUSH) == WRITE_FLUSH)
 					btrfs_dev_stat_inc(dev,
 						BTRFS_DEV_STAT_FLUSH_ERRS);
 				btrfs_dev_stat_print_on_error(dev);
@@ -5988,7 +6071,7 @@ static void btrfs_end_bio(struct bio *bio)
  */
 static noinline void btrfs_schedule_bio(struct btrfs_root *root,
 					struct btrfs_device *device,
-					int rw, struct bio *bio)
+					struct bio *bio)
 {
 	int should_queue = 1;
 	struct btrfs_pending_bios *pending_bios;
@@ -5999,9 +6082,9 @@ static noinline void btrfs_schedule_bio(struct btrfs_root *root,
 	}
 
 	/* don't bother with additional async steps for reads, right now */
-	if (!(rw & REQ_WRITE)) {
+	if (bio_op(bio) == REQ_OP_READ) {
 		bio_get(bio);
-		btrfsic_submit_bio(rw, bio);
+		btrfsic_submit_bio(bio);
 		bio_put(bio);
 		return;
 	}
@@ -6015,10 +6098,9 @@ static noinline void btrfs_schedule_bio(struct btrfs_root *root,
 	atomic_inc(&root->fs_info->nr_async_bios);
 	WARN_ON(bio->bi_next);
 	bio->bi_next = NULL;
-	bio->bi_rw |= rw;
 
 	spin_lock(&device->io_lock);
-	if (bio->bi_rw & REQ_SYNC)
+	if (bio->bi_opf & REQ_SYNC)
 		pending_bios = &device->pending_sync_bios;
 	else
 		pending_bios = &device->pending_bios;
@@ -6041,7 +6123,7 @@ static noinline void btrfs_schedule_bio(struct btrfs_root *root,
 
 static void submit_stripe_bio(struct btrfs_root *root, struct btrfs_bio *bbio,
 			      struct bio *bio, u64 physical, int dev_nr,
-			      int rw, int async)
+			      int async)
 {
 	struct btrfs_device *dev = bbio->stripes[dev_nr].dev;
 
@@ -6055,10 +6137,12 @@ static void submit_stripe_bio(struct btrfs_root *root, struct btrfs_bio *bbio,
 
 		rcu_read_lock();
 		name = rcu_dereference(dev->name);
-		pr_debug("btrfs_map_bio: rw %d, sector=%llu, dev=%lu "
-			 "(%s id %llu), size=%u\n", rw,
-			 (u64)bio->bi_iter.bi_sector, (u_long)dev->bdev->bd_dev,
-			 name->str, dev->devid, bio->bi_iter.bi_size);
+		btrfs_debug(fs_info,
+			"btrfs_map_bio: rw %d 0x%x, sector=%llu, dev=%lu (%s id %llu), size=%u",
+			bio_op(bio), bio->bi_opf,
+			(u64)bio->bi_iter.bi_sector,
+			(u_long)dev->bdev->bd_dev, name->str, dev->devid,
+			bio->bi_iter.bi_size);
 		rcu_read_unlock();
 	}
 #endif
@@ -6067,9 +6151,9 @@ static void submit_stripe_bio(struct btrfs_root *root, struct btrfs_bio *bbio,
 	btrfs_bio_counter_inc_noblocked(root->fs_info);
 
 	if (async)
-		btrfs_schedule_bio(root, dev, rw, bio);
+		btrfs_schedule_bio(root, dev, bio);
 	else
-		btrfsic_submit_bio(rw, bio);
+		btrfsic_submit_bio(bio);
 }
 
 static void bbio_error(struct btrfs_bio *bbio, struct bio *bio, u64 logical)
@@ -6086,7 +6170,7 @@ static void bbio_error(struct btrfs_bio *bbio, struct bio *bio, u64 logical)
 	}
 }
 
-int btrfs_map_bio(struct btrfs_root *root, int rw, struct bio *bio,
+int btrfs_map_bio(struct btrfs_root *root, struct bio *bio,
 		  int mirror_num, int async_submit)
 {
 	struct btrfs_device *dev;
@@ -6103,8 +6187,8 @@ int btrfs_map_bio(struct btrfs_root *root, int rw, struct bio *bio,
 	map_length = length;
 
 	btrfs_bio_counter_inc_blocked(root->fs_info);
-	ret = __btrfs_map_block(root->fs_info, rw, logical, &map_length, &bbio,
-			      mirror_num, 1);
+	ret = __btrfs_map_block(root->fs_info, bio_op(bio), logical,
+				&map_length, &bbio, mirror_num, 1);
 	if (ret) {
 		btrfs_bio_counter_dec(root->fs_info);
 		return ret;
@@ -6118,10 +6202,10 @@ int btrfs_map_bio(struct btrfs_root *root, int rw, struct bio *bio,
 	atomic_set(&bbio->stripes_pending, bbio->num_stripes);
 
 	if ((bbio->map_type & BTRFS_BLOCK_GROUP_RAID56_MASK) &&
-	    ((rw & WRITE) || (mirror_num > 1))) {
+	    ((bio_op(bio) == REQ_OP_WRITE) || (mirror_num > 1))) {
 		/* In this case, map_length has been set to the length of
 		   a single stripe; not the whole write */
-		if (rw & WRITE) {
+		if (bio_op(bio) == REQ_OP_WRITE) {
 			ret = raid56_parity_write(root, bio, bbio, map_length);
 		} else {
 			ret = raid56_parity_recover(root, bio, bbio, map_length,
@@ -6133,14 +6217,16 @@ int btrfs_map_bio(struct btrfs_root *root, int rw, struct bio *bio,
 	}
 
 	if (map_length < length) {
-		btrfs_crit(root->fs_info, "mapping failed logical %llu bio len %llu len %llu",
-			logical, length, map_length);
+		btrfs_crit(root->fs_info,
+			   "mapping failed logical %llu bio len %llu len %llu",
+			   logical, length, map_length);
 		BUG();
 	}
 
 	for (dev_nr = 0; dev_nr < total_devs; dev_nr++) {
 		dev = bbio->stripes[dev_nr].dev;
-		if (!dev || !dev->bdev || (rw & WRITE && !dev->writeable)) {
+		if (!dev || !dev->bdev ||
+		    (bio_op(bio) == REQ_OP_WRITE && !dev->writeable)) {
 			bbio_error(bbio, first_bio, logical);
 			continue;
 		}
@@ -6152,7 +6238,7 @@ int btrfs_map_bio(struct btrfs_root *root, int rw, struct bio *bio,
 			bio = first_bio;
 
 		submit_stripe_bio(root, bbio, bio,
-				  bbio->stripes[dev_nr].physical, dev_nr, rw,
+				  bbio->stripes[dev_nr].physical, dev_nr,
 				  async_submit);
 	}
 	btrfs_bio_counter_dec(root->fs_info);
@@ -6250,27 +6336,23 @@ struct btrfs_device *btrfs_alloc_device(struct btrfs_fs_info *fs_info,
 	return dev;
 }
 
-static int read_one_chunk(struct btrfs_root *root, struct btrfs_key *key,
-			  struct extent_buffer *leaf,
-			  struct btrfs_chunk *chunk)
+/* Return -EIO if any error, otherwise return 0. */
+static int btrfs_check_chunk_valid(struct btrfs_root *root,
+				   struct extent_buffer *leaf,
+				   struct btrfs_chunk *chunk, u64 logical)
 {
-	struct btrfs_mapping_tree *map_tree = &root->fs_info->mapping_tree;
-	struct map_lookup *map;
-	struct extent_map *em;
-	u64 logical;
 	u64 length;
 	u64 stripe_len;
-	u64 devid;
-	u8 uuid[BTRFS_UUID_SIZE];
-	int num_stripes;
-	int ret;
-	int i;
+	u16 num_stripes;
+	u16 sub_stripes;
+	u64 type;
 
-	logical = key->offset;
 	length = btrfs_chunk_length(leaf, chunk);
 	stripe_len = btrfs_chunk_stripe_len(leaf, chunk);
 	num_stripes = btrfs_chunk_num_stripes(leaf, chunk);
-	/* Validation check */
+	sub_stripes = btrfs_chunk_sub_stripes(leaf, chunk);
+	type = btrfs_chunk_type(leaf, chunk);
+
 	if (!num_stripes) {
 		btrfs_err(root->fs_info, "invalid chunk num_stripes: %u",
 			  num_stripes);
@@ -6281,6 +6363,11 @@ static int read_one_chunk(struct btrfs_root *root, struct btrfs_key *key,
 			  "invalid chunk logical %llu", logical);
 		return -EIO;
 	}
+	if (btrfs_chunk_sector_size(leaf, chunk) != root->sectorsize) {
+		btrfs_err(root->fs_info, "invalid chunk sectorsize %u",
+			  btrfs_chunk_sector_size(leaf, chunk));
+		return -EIO;
+	}
 	if (!length || !IS_ALIGNED(length, root->sectorsize)) {
 		btrfs_err(root->fs_info,
 			"invalid chunk length %llu", length);
@@ -6292,13 +6379,54 @@ static int read_one_chunk(struct btrfs_root *root, struct btrfs_key *key,
 		return -EIO;
 	}
 	if (~(BTRFS_BLOCK_GROUP_TYPE_MASK | BTRFS_BLOCK_GROUP_PROFILE_MASK) &
-	    btrfs_chunk_type(leaf, chunk)) {
+	    type) {
 		btrfs_err(root->fs_info, "unrecognized chunk type: %llu",
 			  ~(BTRFS_BLOCK_GROUP_TYPE_MASK |
 			    BTRFS_BLOCK_GROUP_PROFILE_MASK) &
 			  btrfs_chunk_type(leaf, chunk));
 		return -EIO;
 	}
+	if ((type & BTRFS_BLOCK_GROUP_RAID10 && sub_stripes != 2) ||
+	    (type & BTRFS_BLOCK_GROUP_RAID1 && num_stripes < 1) ||
+	    (type & BTRFS_BLOCK_GROUP_RAID5 && num_stripes < 2) ||
+	    (type & BTRFS_BLOCK_GROUP_RAID6 && num_stripes < 3) ||
+	    (type & BTRFS_BLOCK_GROUP_DUP && num_stripes > 2) ||
+	    ((type & BTRFS_BLOCK_GROUP_PROFILE_MASK) == 0 &&
+	     num_stripes != 1)) {
+		btrfs_err(root->fs_info,
+			"invalid num_stripes:sub_stripes %u:%u for profile %llu",
+			num_stripes, sub_stripes,
+			type & BTRFS_BLOCK_GROUP_PROFILE_MASK);
+		return -EIO;
+	}
+
+	return 0;
+}
+
+static int read_one_chunk(struct btrfs_root *root, struct btrfs_key *key,
+			  struct extent_buffer *leaf,
+			  struct btrfs_chunk *chunk)
+{
+	struct btrfs_mapping_tree *map_tree = &root->fs_info->mapping_tree;
+	struct map_lookup *map;
+	struct extent_map *em;
+	u64 logical;
+	u64 length;
+	u64 stripe_len;
+	u64 devid;
+	u8 uuid[BTRFS_UUID_SIZE];
+	int num_stripes;
+	int ret;
+	int i;
+
+	logical = key->offset;
+	length = btrfs_chunk_length(leaf, chunk);
+	stripe_len = btrfs_chunk_stripe_len(leaf, chunk);
+	num_stripes = btrfs_chunk_num_stripes(leaf, chunk);
+
+	ret = btrfs_check_chunk_valid(root, leaf, chunk, logical);
+	if (ret)
+		return ret;
 
 	read_lock(&map_tree->map_tree.lock);
 	em = lookup_extent_mapping(&map_tree->map_tree, logical, 1);
@@ -6345,7 +6473,8 @@ static int read_one_chunk(struct btrfs_root *root, struct btrfs_key *key,
 				   BTRFS_UUID_SIZE);
 		map->stripes[i].dev = btrfs_find_device(root->fs_info, devid,
 							uuid, NULL);
-		if (!map->stripes[i].dev && !btrfs_test_opt(root, DEGRADED)) {
+		if (!map->stripes[i].dev &&
+		    !btrfs_test_opt(root->fs_info, DEGRADED)) {
 			free_extent_map(em);
 			return -EIO;
 		}
@@ -6357,8 +6486,9 @@ static int read_one_chunk(struct btrfs_root *root, struct btrfs_key *key,
 				free_extent_map(em);
 				return -EIO;
 			}
-			btrfs_warn(root->fs_info, "devid %llu uuid %pU is missing",
-						devid, uuid);
+			btrfs_warn(root->fs_info,
+				   "devid %llu uuid %pU is missing",
+				   devid, uuid);
 		}
 		map->stripes[i].dev->in_fs_metadata = 1;
 	}
@@ -6413,7 +6543,7 @@ static struct btrfs_fs_devices *open_seed_devices(struct btrfs_root *root,
 
 	fs_devices = find_fsid(fsid);
 	if (!fs_devices) {
-		if (!btrfs_test_opt(root, DEGRADED))
+		if (!btrfs_test_opt(root->fs_info, DEGRADED))
 			return ERR_PTR(-ENOENT);
 
 		fs_devices = alloc_fs_devices(fsid);
@@ -6475,7 +6605,7 @@ static int read_one_dev(struct btrfs_root *root,
 
 	device = btrfs_find_device(root->fs_info, devid, dev_uuid, fs_uuid);
 	if (!device) {
-		if (!btrfs_test_opt(root, DEGRADED))
+		if (!btrfs_test_opt(root->fs_info, DEGRADED))
 			return -EIO;
 
 		device = add_missing_dev(root, fs_devices, devid, dev_uuid);
@@ -6484,7 +6614,7 @@ static int read_one_dev(struct btrfs_root *root,
 		btrfs_warn(root->fs_info, "devid %llu uuid %pU missing",
 				devid, dev_uuid);
 	} else {
-		if (!device->bdev && !btrfs_test_opt(root, DEGRADED))
+		if (!device->bdev && !btrfs_test_opt(root->fs_info, DEGRADED))
 			return -EIO;
 
 		if(!device->bdev && !device->missing) {
@@ -6535,7 +6665,8 @@ static int read_one_dev(struct btrfs_root *root,
 
 int btrfs_read_sys_array(struct btrfs_root *root)
 {
-	struct btrfs_super_block *super_copy = root->fs_info->super_copy;
+	struct btrfs_fs_info *fs_info = root->fs_info;
+	struct btrfs_super_block *super_copy = fs_info->super_copy;
 	struct extent_buffer *sb;
 	struct btrfs_disk_key *disk_key;
 	struct btrfs_chunk *chunk;
@@ -6546,6 +6677,7 @@ int btrfs_read_sys_array(struct btrfs_root *root)
 	u32 array_size;
 	u32 len = 0;
 	u32 cur_offset;
+	u64 type;
 	struct btrfs_key key;
 
 	ASSERT(BTRFS_SUPER_INFO_SIZE <= root->nodesize);
@@ -6555,8 +6687,8 @@ int btrfs_read_sys_array(struct btrfs_root *root)
 	 * overallocate but we can keep it as-is, only the first page is used.
 	 */
 	sb = btrfs_find_create_tree_block(root, BTRFS_SUPER_INFO_OFFSET);
-	if (!sb)
-		return -ENOMEM;
+	if (IS_ERR(sb))
+		return PTR_ERR(sb);
 	set_extent_buffer_uptodate(sb);
 	btrfs_set_buffer_lockdep_class(root->root_key.objectid, sb, 0);
 	/*
@@ -6605,13 +6737,22 @@ int btrfs_read_sys_array(struct btrfs_root *root)
 
 			num_stripes = btrfs_chunk_num_stripes(sb, chunk);
 			if (!num_stripes) {
-				printk(KERN_ERR
-	    "BTRFS: invalid number of stripes %u in sys_array at offset %u\n",
+				btrfs_err(fs_info,
+					"invalid number of stripes %u in sys_array at offset %u",
 					num_stripes, cur_offset);
 				ret = -EIO;
 				break;
 			}
 
+			type = btrfs_chunk_type(sb, chunk);
+			if ((type & BTRFS_BLOCK_GROUP_SYSTEM) == 0) {
+				btrfs_err(fs_info,
+			    "invalid chunk type %llu in sys_array at offset %u",
+					type, cur_offset);
+				ret = -EIO;
+				break;
+			}
+
 			len = btrfs_chunk_item_size(num_stripes);
 			if (cur_offset + len > array_size)
 				goto out_short_read;
@@ -6620,9 +6761,9 @@ int btrfs_read_sys_array(struct btrfs_root *root)
 			if (ret)
 				break;
 		} else {
-			printk(KERN_ERR
-		"BTRFS: unexpected item type %u in sys_array at offset %u\n",
-				(u32)key.type, cur_offset);
+			btrfs_err(fs_info,
+			    "unexpected item type %u in sys_array at offset %u",
+				  (u32)key.type, cur_offset);
 			ret = -EIO;
 			break;
 		}
@@ -6630,12 +6771,14 @@ int btrfs_read_sys_array(struct btrfs_root *root)
 		sb_array_offset += len;
 		cur_offset += len;
 	}
+	clear_extent_buffer_uptodate(sb);
 	free_extent_buffer_stale(sb);
 	return ret;
 
 out_short_read:
-	printk(KERN_ERR "BTRFS: sys_array too short to read %u bytes at offset %u\n",
+	btrfs_err(fs_info, "sys_array too short to read %u bytes at offset %u",
 			len, cur_offset);
+	clear_extent_buffer_uptodate(sb);
 	free_extent_buffer_stale(sb);
 	return -EIO;
 }
@@ -6648,6 +6791,7 @@ int btrfs_read_chunk_tree(struct btrfs_root *root)
 	struct btrfs_key found_key;
 	int ret;
 	int slot;
+	u64 total_dev = 0;
 
 	root = root->fs_info->chunk_root;
 
@@ -6689,6 +6833,7 @@ int btrfs_read_chunk_tree(struct btrfs_root *root)
 			ret = read_one_dev(root, leaf, dev_item);
 			if (ret)
 				goto error;
+			total_dev++;
 		} else if (found_key.type == BTRFS_CHUNK_ITEM_KEY) {
 			struct btrfs_chunk *chunk;
 			chunk = btrfs_item_ptr(leaf, slot, struct btrfs_chunk);
@@ -6698,6 +6843,28 @@ int btrfs_read_chunk_tree(struct btrfs_root *root)
 		}
 		path->slots[0]++;
 	}
+
+	/*
+	 * After loading chunk tree, we've got all device information,
+	 * do another round of validation checks.
+	 */
+	if (total_dev != root->fs_info->fs_devices->total_devices) {
+		btrfs_err(root->fs_info,
+	   "super_num_devices %llu mismatch with num_devices %llu found here",
+			  btrfs_super_num_devices(root->fs_info->super_copy),
+			  total_dev);
+		ret = -EINVAL;
+		goto error;
+	}
+	if (btrfs_super_total_bytes(root->fs_info->super_copy) <
+	    root->fs_info->fs_devices->total_rw_bytes) {
+		btrfs_err(root->fs_info,
+	"super_total_bytes %llu mismatch with fs_devices total_rw_bytes %llu",
+			  btrfs_super_total_bytes(root->fs_info->super_copy),
+			  root->fs_info->fs_devices->total_rw_bytes);
+		ret = -EINVAL;
+		goto error;
+	}
 	ret = 0;
 error:
 	unlock_chunks(root);
@@ -6933,10 +7100,12 @@ int btrfs_get_dev_stats(struct btrfs_root *root,
 	mutex_unlock(&fs_devices->device_list_mutex);
 
 	if (!dev) {
-		btrfs_warn(root->fs_info, "get dev_stats failed, device not found");
+		btrfs_warn(root->fs_info,
+			   "get dev_stats failed, device not found");
 		return -ENODEV;
 	} else if (!dev->dev_stats_valid) {
-		btrfs_warn(root->fs_info, "get dev_stats failed, not yet valid");
+		btrfs_warn(root->fs_info,
+			   "get dev_stats failed, not yet valid");
 		return -ENODEV;
 	} else if (stats->flags & BTRFS_DEV_STATS_RESET) {
 		for (i = 0; i < BTRFS_DEV_STAT_VALUES_MAX; i++) {
@@ -7051,38 +7220,3 @@ void btrfs_reset_fs_info_ptr(struct btrfs_fs_info *fs_info)
 		fs_devices = fs_devices->seed;
 	}
 }
-
-static void btrfs_close_one_device(struct btrfs_device *device)
-{
-	struct btrfs_fs_devices *fs_devices = device->fs_devices;
-	struct btrfs_device *new_device;
-	struct rcu_string *name;
-
-	if (device->bdev)
-		fs_devices->open_devices--;
-
-	if (device->writeable &&
-	    device->devid != BTRFS_DEV_REPLACE_DEVID) {
-		list_del_init(&device->dev_alloc_list);
-		fs_devices->rw_devices--;
-	}
-
-	if (device->missing)
-		fs_devices->missing_devices--;
-
-	new_device = btrfs_alloc_device(NULL, &device->devid,
-					device->uuid);
-	BUG_ON(IS_ERR(new_device)); /* -ENOMEM */
-
-	/* Safe because we are under uuid_mutex */
-	if (device->name) {
-		name = rcu_string_strdup(device->name->str, GFP_NOFS);
-		BUG_ON(!name); /* -ENOMEM */
-		rcu_assign_pointer(new_device->name, name);
-	}
-
-	list_replace_rcu(&device->dev_list, &new_device->dev_list);
-	new_device->fs_devices = device->fs_devices;
-
-	call_rcu(&device->rcu, free_device);
-}
diff --git a/fs/btrfs/volumes.h b/fs/btrfs/volumes.h
index 0ac90f8d85bd..09ed29c67848 100644
--- a/fs/btrfs/volumes.h
+++ b/fs/btrfs/volumes.h
@@ -375,14 +375,14 @@ int btrfs_account_dev_extents_size(struct btrfs_device *device, u64 start,
 				   u64 end, u64 *length);
 void btrfs_get_bbio(struct btrfs_bio *bbio);
 void btrfs_put_bbio(struct btrfs_bio *bbio);
-int btrfs_map_block(struct btrfs_fs_info *fs_info, int rw,
+int btrfs_map_block(struct btrfs_fs_info *fs_info, int op,
 		    u64 logical, u64 *length,
 		    struct btrfs_bio **bbio_ret, int mirror_num);
-int btrfs_map_sblock(struct btrfs_fs_info *fs_info, int rw,
+int btrfs_map_sblock(struct btrfs_fs_info *fs_info, int op,
 		     u64 logical, u64 *length,
 		     struct btrfs_bio **bbio_ret, int mirror_num,
 		     int need_raid_map);
-int btrfs_rmap_block(struct btrfs_mapping_tree *map_tree,
+int btrfs_rmap_block(struct btrfs_fs_info *fs_info,
 		     u64 chunk_start, u64 physical, u64 devid,
 		     u64 **logical, int *naddrs, int *stripe_len);
 int btrfs_read_sys_array(struct btrfs_root *root);
@@ -391,7 +391,7 @@ int btrfs_alloc_chunk(struct btrfs_trans_handle *trans,
 		      struct btrfs_root *extent_root, u64 type);
 void btrfs_mapping_init(struct btrfs_mapping_tree *tree);
 void btrfs_mapping_tree_free(struct btrfs_mapping_tree *tree);
-int btrfs_map_bio(struct btrfs_root *root, int rw, struct bio *bio,
+int btrfs_map_bio(struct btrfs_root *root, struct bio *bio,
 		  int mirror_num, int async_submit);
 int btrfs_open_devices(struct btrfs_fs_devices *fs_devices,
 		       fmode_t flags, void *holder);
diff --git a/fs/btrfs/xattr.c b/fs/btrfs/xattr.c
index d1a177a3dbe8..fccbf5567e78 100644
--- a/fs/btrfs/xattr.c
+++ b/fs/btrfs/xattr.c
@@ -252,7 +252,7 @@ int __btrfs_setxattr(struct btrfs_trans_handle *trans,
 		goto out;
 
 	inode_inc_iversion(inode);
-	inode->i_ctime = current_fs_time(inode->i_sb);
+	inode->i_ctime = current_time(inode);
 	set_bit(BTRFS_INODE_COPY_EVERYTHING, &BTRFS_I(inode)->runtime_flags);
 	ret = btrfs_update_inode(trans, root, inode);
 	BUG_ON(ret);
diff --git a/fs/btrfs/zlib.c b/fs/btrfs/zlib.c
index 88d274e8ecf2..441b81a3e545 100644
--- a/fs/btrfs/zlib.c
+++ b/fs/btrfs/zlib.c
@@ -95,7 +95,7 @@ static int zlib_compress_pages(struct list_head *ws,
 	*total_in = 0;
 
 	if (Z_OK != zlib_deflateInit(&workspace->strm, 3)) {
-		printk(KERN_WARNING "BTRFS: deflateInit failed\n");
+		pr_warn("BTRFS: deflateInit failed\n");
 		ret = -EIO;
 		goto out;
 	}
@@ -123,7 +123,7 @@ static int zlib_compress_pages(struct list_head *ws,
 	while (workspace->strm.total_in < len) {
 		ret = zlib_deflate(&workspace->strm, Z_SYNC_FLUSH);
 		if (ret != Z_OK) {
-			printk(KERN_DEBUG "BTRFS: deflate in loop returned %d\n",
+			pr_debug("BTRFS: deflate in loop returned %d\n",
 			       ret);
 			zlib_deflateEnd(&workspace->strm);
 			ret = -EIO;
@@ -249,7 +249,7 @@ static int zlib_decompress_biovec(struct list_head *ws, struct page **pages_in,
 	}
 
 	if (Z_OK != zlib_inflateInit2(&workspace->strm, wbits)) {
-		printk(KERN_WARNING "BTRFS: inflateInit failed\n");
+		pr_warn("BTRFS: inflateInit failed\n");
 		return -EIO;
 	}
 	while (workspace->strm.total_in < srclen) {
@@ -339,7 +339,7 @@ static int zlib_decompress(struct list_head *ws, unsigned char *data_in,
 	}
 
 	if (Z_OK != zlib_inflateInit2(&workspace->strm, wbits)) {
-		printk(KERN_WARNING "BTRFS: inflateInit failed\n");
+		pr_warn("BTRFS: inflateInit failed\n");
 		return -EIO;
 	}