Merge branch 'master' into for-next

28 files changed, 1348 insertions, 435 deletions

diff --git a/fs/btrfs/acl.c b/fs/btrfs/acl.c
index 6df6d6e..6ef7b26 100644
--- a/fs/btrfs/acl.c
+++ b/fs/btrfs/acl.c
@@ -22,6 +22,7 @@
 #include <linux/posix_acl_xattr.h>
 #include <linux/posix_acl.h>
 #include <linux/sched.h>
+#include <linux/slab.h>
 
 #include "ctree.h"
 #include "btrfs_inode.h"
diff --git a/fs/btrfs/async-thread.c b/fs/btrfs/async-thread.c
index c0861e7..462859a 100644
--- a/fs/btrfs/async-thread.c
+++ b/fs/btrfs/async-thread.c
@@ -17,6 +17,7 @@
  */
 
 #include <linux/kthread.h>
+#include <linux/slab.h>
 #include <linux/list.h>
 #include <linux/spinlock.h>
 #include <linux/freezer.h>
diff --git a/fs/btrfs/btrfs_inode.h b/fs/btrfs/btrfs_inode.h
index 3f1f50d..7a4dee1 100644
--- a/fs/btrfs/btrfs_inode.h
+++ b/fs/btrfs/btrfs_inode.h
@@ -153,6 +153,11 @@ struct btrfs_inode {
 	unsigned ordered_data_close:1;
 	unsigned dummy_inode:1;
 
+	/*
+	 * always compress this one file
+	 */
+	unsigned force_compress:1;
+
 	struct inode vfs_inode;
 };
 
diff --git a/fs/btrfs/compression.c b/fs/btrfs/compression.c
index a11a320..396039b 100644
--- a/fs/btrfs/compression.c
+++ b/fs/btrfs/compression.c
@@ -31,7 +31,7 @@
 #include <linux/swap.h>
 #include <linux/writeback.h>
 #include <linux/bit_spinlock.h>
-#include <linux/pagevec.h>
+#include <linux/slab.h>
 #include "compat.h"
 #include "ctree.h"
 #include "disk-io.h"
@@ -445,7 +445,6 @@ static noinline int add_ra_bio_pages(struct inode *inode,
 	unsigned long nr_pages = 0;
 	struct extent_map *em;
 	struct address_space *mapping = inode->i_mapping;
-	struct pagevec pvec;
 	struct extent_map_tree *em_tree;
 	struct extent_io_tree *tree;
 	u64 end;
@@ -461,7 +460,6 @@ static noinline int add_ra_bio_pages(struct inode *inode,
 
 	end_index = (i_size_read(inode) - 1) >> PAGE_CACHE_SHIFT;
 
-	pagevec_init(&pvec, 0);
 	while (last_offset < compressed_end) {
 		page_index = last_offset >> PAGE_CACHE_SHIFT;
 
@@ -478,26 +476,17 @@ static noinline int add_ra_bio_pages(struct inode *inode,
 			goto next;
 		}
 
-		page = alloc_page(mapping_gfp_mask(mapping) | GFP_NOFS);
+		page = __page_cache_alloc(mapping_gfp_mask(mapping) &
+								~__GFP_FS);
 		if (!page)
 			break;
 
-		page->index = page_index;
-		/*
-		 * what we want to do here is call add_to_page_cache_lru,
-		 * but that isn't exported, so we reproduce it here
-		 */
-		if (add_to_page_cache(page, mapping,
-				      page->index, GFP_NOFS)) {
+		if (add_to_page_cache_lru(page, mapping, page_index,
+								GFP_NOFS)) {
 			page_cache_release(page);
 			goto next;
 		}
 
-		/* open coding of lru_cache_add, also not exported */
-		page_cache_get(page);
-		if (!pagevec_add(&pvec, page))
-			__pagevec_lru_add_file(&pvec);
-
 		end = last_offset + PAGE_CACHE_SIZE - 1;
 		/*
 		 * at this point, we have a locked page in the page cache
@@ -551,8 +540,6 @@ static noinline int add_ra_bio_pages(struct inode *inode,
 next:
 		last_offset += PAGE_CACHE_SIZE;
 	}
-	if (pagevec_count(&pvec))
-		__pagevec_lru_add_file(&pvec);
 	return 0;
 }
 
diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c
index c4bc570..6795a71 100644
--- a/fs/btrfs/ctree.c
+++ b/fs/btrfs/ctree.c
@@ -17,6 +17,7 @@
  */
 
 #include <linux/sched.h>
+#include <linux/slab.h>
 #include "ctree.h"
 #include "disk-io.h"
 #include "transaction.h"
@@ -3040,6 +3041,10 @@ static noinline int setup_leaf_for_split(struct btrfs_trans_handle *trans,
 	if (ret > 0 || item_size != btrfs_item_size_nr(leaf, path->slots[0]))
 		goto err;
 
+	/* the leaf has  changed, it now has room.  return now */
+	if (btrfs_leaf_free_space(root, path->nodes[0]) >= ins_len)
+		goto err;
+
 	if (key.type == BTRFS_EXTENT_DATA_KEY) {
 		fi = btrfs_item_ptr(leaf, path->slots[0],
 				    struct btrfs_file_extent_item);
diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index 8b5cfdd..746a724 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -26,6 +26,7 @@
 #include <linux/completion.h>
 #include <linux/backing-dev.h>
 #include <linux/wait.h>
+#include <linux/slab.h>
 #include <asm/kmap_types.h>
 #include "extent_io.h"
 #include "extent_map.h"
@@ -373,11 +374,13 @@ struct btrfs_super_block {
  * ones specified below then we will fail to mount
  */
 #define BTRFS_FEATURE_INCOMPAT_MIXED_BACKREF	(1ULL << 0)
+#define BTRFS_FEATURE_INCOMPAT_DEFAULT_SUBVOL	(2ULL << 0)
 
 #define BTRFS_FEATURE_COMPAT_SUPP		0ULL
 #define BTRFS_FEATURE_COMPAT_RO_SUPP		0ULL
 #define BTRFS_FEATURE_INCOMPAT_SUPP		\
-	BTRFS_FEATURE_INCOMPAT_MIXED_BACKREF
+	(BTRFS_FEATURE_INCOMPAT_MIXED_BACKREF |	\
+	 BTRFS_FEATURE_INCOMPAT_DEFAULT_SUBVOL)
 
 /*
  * A leaf is full of items. offset and size tell us where to find
@@ -832,7 +835,6 @@ struct btrfs_fs_info {
 	u64 last_trans_log_full_commit;
 	u64 open_ioctl_trans;
 	unsigned long mount_opt;
-	u64 max_extent;
 	u64 max_inline;
 	u64 alloc_start;
 	struct btrfs_transaction *running_transaction;
@@ -1182,7 +1184,6 @@ struct btrfs_root {
 #define BTRFS_INODE_NOATIME		(1 << 9)
 #define BTRFS_INODE_DIRSYNC		(1 << 10)
 
-
 /* some macros to generate set/get funcs for the struct fields.  This
  * assumes there is a lefoo_to_cpu for every type, so lets make a simple
  * one for u8:
@@ -1842,7 +1843,7 @@ BTRFS_SETGET_STACK_FUNCS(super_num_devices, struct btrfs_super_block,
 BTRFS_SETGET_STACK_FUNCS(super_compat_flags, struct btrfs_super_block,
 			 compat_flags, 64);
 BTRFS_SETGET_STACK_FUNCS(super_compat_ro_flags, struct btrfs_super_block,
-			 compat_flags, 64);
+			 compat_ro_flags, 64);
 BTRFS_SETGET_STACK_FUNCS(super_incompat_flags, struct btrfs_super_block,
 			 incompat_flags, 64);
 BTRFS_SETGET_STACK_FUNCS(super_csum_type, struct btrfs_super_block,
@@ -2310,7 +2311,8 @@ int btrfs_truncate_inode_items(struct btrfs_trans_handle *trans,
 			       u32 min_type);
 
 int btrfs_start_delalloc_inodes(struct btrfs_root *root, int delay_iput);
-int btrfs_set_extent_delalloc(struct inode *inode, u64 start, u64 end);
+int btrfs_set_extent_delalloc(struct inode *inode, u64 start, u64 end,
+			      struct extent_state **cached_state);
 int btrfs_writepages(struct address_space *mapping,
 		     struct writeback_control *wbc);
 int btrfs_create_subvol_root(struct btrfs_trans_handle *trans,
@@ -2335,7 +2337,7 @@ int btrfs_init_cachep(void);
 void btrfs_destroy_cachep(void);
 long btrfs_ioctl_trans_end(struct file *file);
 struct inode *btrfs_iget(struct super_block *s, struct btrfs_key *location,
-			 struct btrfs_root *root);
+			 struct btrfs_root *root, int *was_new);
 int btrfs_commit_write(struct file *file, struct page *page,
 		       unsigned from, unsigned to);
 struct extent_map *btrfs_get_extent(struct inode *inode, struct page *page,
@@ -2386,7 +2388,6 @@ void btrfs_sysfs_del_super(struct btrfs_fs_info *root);
 ssize_t btrfs_listxattr(struct dentry *dentry, char *buffer, size_t size);
 
 /* super.c */
-u64 btrfs_parse_size(char *str);
 int btrfs_parse_options(struct btrfs_root *root, char *options);
 int btrfs_sync_fs(struct super_block *sb, int wait);
 
diff --git a/fs/btrfs/delayed-ref.c b/fs/btrfs/delayed-ref.c
index 84e6781..902ce50 100644
--- a/fs/btrfs/delayed-ref.c
+++ b/fs/btrfs/delayed-ref.c
@@ -17,6 +17,7 @@
  */
 
 #include <linux/sched.h>
+#include <linux/slab.h>
 #include <linux/sort.h>
 #include "ctree.h"
 #include "delayed-ref.h"
diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
index 0427183..e7b8f2c 100644
--- a/fs/btrfs/disk-io.c
+++ b/fs/btrfs/disk-io.c
@@ -27,6 +27,7 @@
 #include <linux/kthread.h>
 #include <linux/freezer.h>
 #include <linux/crc32c.h>
+#include <linux/slab.h>
 #include "compat.h"
 #include "ctree.h"
 #include "disk-io.h"
@@ -263,13 +264,15 @@ static int csum_tree_block(struct btrfs_root *root, struct extent_buffer *buf,
 static int verify_parent_transid(struct extent_io_tree *io_tree,
 				 struct extent_buffer *eb, u64 parent_transid)
 {
+	struct extent_state *cached_state = NULL;
 	int ret;
 
 	if (!parent_transid || btrfs_header_generation(eb) == parent_transid)
 		return 0;
 
-	lock_extent(io_tree, eb->start, eb->start + eb->len - 1, GFP_NOFS);
-	if (extent_buffer_uptodate(io_tree, eb) &&
+	lock_extent_bits(io_tree, eb->start, eb->start + eb->len - 1,
+			 0, &cached_state, GFP_NOFS);
+	if (extent_buffer_uptodate(io_tree, eb, cached_state) &&
 	    btrfs_header_generation(eb) == parent_transid) {
 		ret = 0;
 		goto out;
@@ -282,10 +285,10 @@ static int verify_parent_transid(struct extent_io_tree *io_tree,
 		       (unsigned long long)btrfs_header_generation(eb));
 	}
 	ret = 1;
-	clear_extent_buffer_uptodate(io_tree, eb);
+	clear_extent_buffer_uptodate(io_tree, eb, &cached_state);
 out:
-	unlock_extent(io_tree, eb->start, eb->start + eb->len - 1,
-		      GFP_NOFS);
+	unlock_extent_cached(io_tree, eb->start, eb->start + eb->len - 1,
+			     &cached_state, GFP_NOFS);
 	return ret;
 }
 
@@ -1632,7 +1635,6 @@ struct btrfs_root *open_ctree(struct super_block *sb,
 	atomic_set(&fs_info->async_submit_draining, 0);
 	atomic_set(&fs_info->nr_async_bios, 0);
 	fs_info->sb = sb;
-	fs_info->max_extent = (u64)-1;
 	fs_info->max_inline = 8192 * 1024;
 	fs_info->metadata_ratio = 0;
 
@@ -1920,7 +1922,11 @@ struct btrfs_root *open_ctree(struct super_block *sb,
 
 	csum_root->track_dirty = 1;
 
-	btrfs_read_block_groups(extent_root);
+	ret = btrfs_read_block_groups(extent_root);
+	if (ret) {
+		printk(KERN_ERR "Failed to read block groups: %d\n", ret);
+		goto fail_block_groups;
+	}
 
 	fs_info->generation = generation;
 	fs_info->last_trans_committed = generation;
@@ -1930,7 +1936,7 @@ struct btrfs_root *open_ctree(struct super_block *sb,
 	fs_info->cleaner_kthread = kthread_run(cleaner_kthread, tree_root,
 					       "btrfs-cleaner");
 	if (IS_ERR(fs_info->cleaner_kthread))
-		goto fail_csum_root;
+		goto fail_block_groups;
 
 	fs_info->transaction_kthread = kthread_run(transaction_kthread,
 						   tree_root,
@@ -2018,7 +2024,8 @@ fail_cleaner:
 	filemap_write_and_wait(fs_info->btree_inode->i_mapping);
 	invalidate_inode_pages2(fs_info->btree_inode->i_mapping);
 
-fail_csum_root:
+fail_block_groups:
+	btrfs_free_block_groups(fs_info);
 	free_extent_buffer(csum_root->node);
 	free_extent_buffer(csum_root->commit_root);
 fail_dev_root:
@@ -2497,7 +2504,8 @@ int btrfs_buffer_uptodate(struct extent_buffer *buf, u64 parent_transid)
 	int ret;
 	struct inode *btree_inode = buf->first_page->mapping->host;
 
-	ret = extent_buffer_uptodate(&BTRFS_I(btree_inode)->io_tree, buf);
+	ret = extent_buffer_uptodate(&BTRFS_I(btree_inode)->io_tree, buf,
+				     NULL);
 	if (!ret)
 		return ret;
 
diff --git a/fs/btrfs/export.c b/fs/btrfs/export.c
index ba5c3fd..951ef09 100644
--- a/fs/btrfs/export.c
+++ b/fs/btrfs/export.c
@@ -95,7 +95,7 @@ static struct dentry *btrfs_get_dentry(struct super_block *sb, u64 objectid,
 	btrfs_set_key_type(&key, BTRFS_INODE_ITEM_KEY);
 	key.offset = 0;
 
-	inode = btrfs_iget(sb, &key, root);
+	inode = btrfs_iget(sb, &key, root, NULL);
 	if (IS_ERR(inode)) {
 		err = PTR_ERR(inode);
 		goto fail;
@@ -223,7 +223,7 @@ static struct dentry *btrfs_get_parent(struct dentry *child)
 
 	key.type = BTRFS_INODE_ITEM_KEY;
 	key.offset = 0;
-	dentry = d_obtain_alias(btrfs_iget(root->fs_info->sb, &key, root));
+	dentry = d_obtain_alias(btrfs_iget(root->fs_info->sb, &key, root, NULL));
 	if (!IS_ERR(dentry))
 		dentry->d_op = &btrfs_dentry_operations;
 	return dentry;
diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index 559f724..b34d32f 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -22,6 +22,7 @@
 #include <linux/sort.h>
 #include <linux/rcupdate.h>
 #include <linux/kthread.h>
+#include <linux/slab.h>
 #include "compat.h"
 #include "hash.h"
 #include "ctree.h"
@@ -2676,6 +2677,8 @@ static int update_space_info(struct btrfs_fs_info *info, u64 flags,
 
 	INIT_LIST_HEAD(&found->block_groups);
 	init_rwsem(&found->groups_sem);
+	init_waitqueue_head(&found->flush_wait);
+	init_waitqueue_head(&found->allocate_wait);
 	spin_lock_init(&found->lock);
 	found->flags = flags;
 	found->total_bytes = total_bytes;
@@ -2846,7 +2849,7 @@ int btrfs_unreserve_metadata_for_delalloc(struct btrfs_root *root,
 	}
 	spin_unlock(&BTRFS_I(inode)->accounting_lock);
 
-	BTRFS_I(inode)->reserved_extents--;
+	BTRFS_I(inode)->reserved_extents -= num_items;
 	BUG_ON(BTRFS_I(inode)->reserved_extents < 0);
 
 	if (meta_sinfo->bytes_delalloc < num_bytes) {
@@ -2944,12 +2947,10 @@ static void flush_delalloc(struct btrfs_root *root,
 
 	spin_lock(&info->lock);
 
-	if (!info->flushing) {
+	if (!info->flushing)
 		info->flushing = 1;
-		init_waitqueue_head(&info->flush_wait);
-	} else {
+	else
 		wait = true;
-	}
 
 	spin_unlock(&info->lock);
 
@@ -3011,7 +3012,6 @@ static int maybe_allocate_chunk(struct btrfs_root *root,
 	if (!info->allocating_chunk) {
 		info->force_alloc = 1;
 		info->allocating_chunk = 1;
-		init_waitqueue_head(&info->allocate_wait);
 	} else {
 		wait = true;
 	}
@@ -3111,7 +3111,7 @@ again:
 		return -ENOSPC;
 	}
 
-	BTRFS_I(inode)->reserved_extents++;
+	BTRFS_I(inode)->reserved_extents += num_items;
 	check_force_delalloc(meta_sinfo);
 	spin_unlock(&meta_sinfo->lock);
 
@@ -3235,7 +3235,8 @@ int btrfs_check_data_free_space(struct btrfs_root *root, struct inode *inode,
 				u64 bytes)
 {
 	struct btrfs_space_info *data_sinfo;
-	int ret = 0, committed = 0;
+	u64 used;
+	int ret = 0, committed = 0, flushed = 0;
 
 	/* make sure bytes are sectorsize aligned */
 	bytes = (bytes + root->sectorsize - 1) & ~((u64)root->sectorsize - 1);
@@ -3247,12 +3248,21 @@ int btrfs_check_data_free_space(struct btrfs_root *root, struct inode *inode,
 again:
 	/* make sure we have enough space to handle the data first */
 	spin_lock(&data_sinfo->lock);
-	if (data_sinfo->total_bytes - data_sinfo->bytes_used -
-	    data_sinfo->bytes_delalloc - data_sinfo->bytes_reserved -
-	    data_sinfo->bytes_pinned - data_sinfo->bytes_readonly -
-	    data_sinfo->bytes_may_use - data_sinfo->bytes_super < bytes) {
+	used = data_sinfo->bytes_used + data_sinfo->bytes_delalloc +
+		data_sinfo->bytes_reserved + data_sinfo->bytes_pinned +
+		data_sinfo->bytes_readonly + data_sinfo->bytes_may_use +
+		data_sinfo->bytes_super;
+
+	if (used + bytes > data_sinfo->total_bytes) {
 		struct btrfs_trans_handle *trans;
 
+		if (!flushed) {
+			spin_unlock(&data_sinfo->lock);
+			flush_delalloc(root, data_sinfo);
+			flushed = 1;
+			goto again;
+		}
+
 		/*
 		 * if we don't have enough free bytes in this space then we need
 		 * to alloc a new chunk.
@@ -4170,6 +4180,10 @@ static noinline int find_free_extent(struct btrfs_trans_handle *trans,
 	ins->offset = 0;
 
 	space_info = __find_space_info(root->fs_info, data);
+	if (!space_info) {
+		printk(KERN_ERR "No space info for %d\n", data);
+		return -ENOSPC;
+	}
 
 	if (orig_root->ref_cows || empty_size)
 		allowed_chunk_alloc = 1;
@@ -5205,6 +5219,8 @@ static noinline int do_walk_down(struct btrfs_trans_handle *trans,
 	next = btrfs_find_tree_block(root, bytenr, blocksize);
 	if (!next) {
 		next = btrfs_find_create_tree_block(root, bytenr, blocksize);
+		if (!next)
+			return -ENOMEM;
 		reada = 1;
 	}
 	btrfs_tree_lock(next);
@@ -5417,7 +5433,8 @@ static noinline int walk_down_tree(struct btrfs_trans_handle *trans,
 		if (ret > 0) {
 			path->slots[level]++;
 			continue;
-		}
+		} else if (ret < 0)
+			return ret;
 		level = wc->level;
 	}
 	return 0;
@@ -6561,6 +6578,7 @@ static noinline int invalidate_extent_cache(struct btrfs_root *root,
 	struct btrfs_key key;
 	struct inode *inode = NULL;
 	struct btrfs_file_extent_item *fi;
+	struct extent_state *cached_state = NULL;
 	u64 num_bytes;
 	u64 skip_objectid = 0;
 	u32 nritems;
@@ -6589,12 +6607,14 @@ static noinline int invalidate_extent_cache(struct btrfs_root *root,
 		}
 		num_bytes = btrfs_file_extent_num_bytes(leaf, fi);
 
-		lock_extent(&BTRFS_I(inode)->io_tree, key.offset,
-			    key.offset + num_bytes - 1, GFP_NOFS);
+		lock_extent_bits(&BTRFS_I(inode)->io_tree, key.offset,
+				 key.offset + num_bytes - 1, 0, &cached_state,
+				 GFP_NOFS);
 		btrfs_drop_extent_cache(inode, key.offset,
 					key.offset + num_bytes - 1, 1);
-		unlock_extent(&BTRFS_I(inode)->io_tree, key.offset,
-			      key.offset + num_bytes - 1, GFP_NOFS);
+		unlock_extent_cached(&BTRFS_I(inode)->io_tree, key.offset,
+				     key.offset + num_bytes - 1, &cached_state,
+				     GFP_NOFS);
 		cond_resched();
 	}
 	iput(inode);
@@ -7366,7 +7386,6 @@ static int find_first_block_group(struct btrfs_root *root,
 		}
 		path->slots[0]++;
 	}
-	ret = -ENOENT;
 out:
 	return ret;
 }
diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c
index 7073cbb..d2d0368 100644
--- a/fs/btrfs/extent_io.c
+++ b/fs/btrfs/extent_io.c
@@ -2,7 +2,6 @@
 #include <linux/slab.h>
 #include <linux/bio.h>
 #include <linux/mm.h>
-#include <linux/gfp.h>
 #include <linux/pagemap.h>
 #include <linux/page-flags.h>
 #include <linux/module.h>
@@ -513,7 +512,10 @@ int clear_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
 	u64 last_end;
 	int err;
 	int set = 0;
+	int clear = 0;
 
+	if (bits & (EXTENT_IOBITS | EXTENT_BOUNDARY))
+		clear = 1;
 again:
 	if (!prealloc && (mask & __GFP_WAIT)) {
 		prealloc = alloc_extent_state(mask);
@@ -524,14 +526,20 @@ again:
 	spin_lock(&tree->lock);
 	if (cached_state) {
 		cached = *cached_state;
-		*cached_state = NULL;
-		cached_state = NULL;
+
+		if (clear) {
+			*cached_state = NULL;
+			cached_state = NULL;
+		}
+
 		if (cached && cached->tree && cached->start == start) {
-			atomic_dec(&cached->refs);
+			if (clear)
+				atomic_dec(&cached->refs);
 			state = cached;
 			goto hit_next;
 		}
-		free_extent_state(cached);
+		if (clear)
+			free_extent_state(cached);
 	}
 	/*
 	 * this search will find the extents that end after
@@ -946,11 +954,11 @@ int clear_extent_bits(struct extent_io_tree *tree, u64 start, u64 end,
 }
 
 int set_extent_delalloc(struct extent_io_tree *tree, u64 start, u64 end,
-		     gfp_t mask)
+			struct extent_state **cached_state, gfp_t mask)
 {
 	return set_extent_bit(tree, start, end,
 			      EXTENT_DELALLOC | EXTENT_DIRTY | EXTENT_UPTODATE,
-			      0, NULL, NULL, mask);
+			      0, NULL, cached_state, mask);
 }
 
 int clear_extent_dirty(struct extent_io_tree *tree, u64 start, u64 end,
@@ -984,10 +992,11 @@ int set_extent_uptodate(struct extent_io_tree *tree, u64 start, u64 end,
 }
 
 static int clear_extent_uptodate(struct extent_io_tree *tree, u64 start,
-				 u64 end, gfp_t mask)
+				 u64 end, struct extent_state **cached_state,
+				 gfp_t mask)
 {
 	return clear_extent_bit(tree, start, end, EXTENT_UPTODATE, 0, 0,
-				NULL, mask);
+				cached_state, mask);
 }
 
 int wait_on_extent_writeback(struct extent_io_tree *tree, u64 start, u64 end)
@@ -1171,7 +1180,8 @@ out:
  * 1 is returned if we find something, 0 if nothing was in the tree
  */
 static noinline u64 find_delalloc_range(struct extent_io_tree *tree,
-					u64 *start, u64 *end, u64 max_bytes)
+					u64 *start, u64 *end, u64 max_bytes,
+					struct extent_state **cached_state)
 {
 	struct rb_node *node;
 	struct extent_state *state;
@@ -1203,8 +1213,11 @@ static noinline u64 find_delalloc_range(struct extent_io_tree *tree,
 				*end = state->end;
 			goto out;
 		}
-		if (!found)
+		if (!found) {
 			*start = state->start;
+			*cached_state = state;
+			atomic_inc(&state->refs);
+		}
 		found++;
 		*end = state->end;
 		cur_start = state->end + 1;
@@ -1336,10 +1349,11 @@ again:
 	delalloc_start = *start;
 	delalloc_end = 0;
 	found = find_delalloc_range(tree, &delalloc_start, &delalloc_end,
-				    max_bytes);
+				    max_bytes, &cached_state);
 	if (!found || delalloc_end <= *start) {
 		*start = delalloc_start;
 		*end = delalloc_end;
+		free_extent_state(cached_state);
 		return found;
 	}
 
@@ -1722,7 +1736,7 @@ static void end_bio_extent_writepage(struct bio *bio, int err)
 		}
 
 		if (!uptodate) {
-			clear_extent_uptodate(tree, start, end, GFP_NOFS);
+			clear_extent_uptodate(tree, start, end, NULL, GFP_NOFS);
 			ClearPageUptodate(page);
 			SetPageError(page);
 		}
@@ -1750,7 +1764,8 @@ static void end_bio_extent_writepage(struct bio *bio, int err)
 static void end_bio_extent_readpage(struct bio *bio, int err)
 {
 	int uptodate = test_bit(BIO_UPTODATE, &bio->bi_flags);
-	struct bio_vec *bvec = bio->bi_io_vec + bio->bi_vcnt - 1;
+	struct bio_vec *bvec_end = bio->bi_io_vec + bio->bi_vcnt - 1;
+	struct bio_vec *bvec = bio->bi_io_vec;
 	struct extent_io_tree *tree;
 	u64 start;
 	u64 end;
@@ -1773,7 +1788,7 @@ static void end_bio_extent_readpage(struct bio *bio, int err)
 		else
 			whole_page = 0;
 
-		if (--bvec >= bio->bi_io_vec)
+		if (++bvec <= bvec_end)
 			prefetchw(&bvec->bv_page->flags);
 
 		if (uptodate && tree->ops && tree->ops->readpage_end_io_hook) {
@@ -1818,7 +1833,7 @@ static void end_bio_extent_readpage(struct bio *bio, int err)
 			}
 			check_page_locked(tree, page);
 		}
-	} while (bvec >= bio->bi_io_vec);
+	} while (bvec <= bvec_end);
 
 	bio_put(bio);
 }
@@ -2663,33 +2678,20 @@ int extent_readpages(struct extent_io_tree *tree,
 {
 	struct bio *bio = NULL;
 	unsigned page_idx;
-	struct pagevec pvec;
 	unsigned long bio_flags = 0;
 
-	pagevec_init(&pvec, 0);
 	for (page_idx = 0; page_idx < nr_pages; page_idx++) {
 		struct page *page = list_entry(pages->prev, struct page, lru);
 
 		prefetchw(&page->flags);
 		list_del(&page->lru);
-		/*
-		 * what we want to do here is call add_to_page_cache_lru,
-		 * but that isn't exported, so we reproduce it here
-		 */
-		if (!add_to_page_cache(page, mapping,
+		if (!add_to_page_cache_lru(page, mapping,
 					page->index, GFP_KERNEL)) {
-
-			/* open coding of lru_cache_add, also not exported */
-			page_cache_get(page);
-			if (!pagevec_add(&pvec, page))
-				__pagevec_lru_add_file(&pvec);
 			__extent_read_full_page(tree, page, get_extent,
 						&bio, 0, &bio_flags);
 		}
 		page_cache_release(page);
 	}
-	if (pagevec_count(&pvec))
-		__pagevec_lru_add_file(&pvec);
 	BUG_ON(!list_empty(pages));
 	if (bio)
 		submit_one_bio(READ, bio, 0, bio_flags);
@@ -2704,6 +2706,7 @@ int extent_readpages(struct extent_io_tree *tree,
 int extent_invalidatepage(struct extent_io_tree *tree,
 			  struct page *page, unsigned long offset)
 {
+	struct extent_state *cached_state = NULL;
 	u64 start = ((u64)page->index << PAGE_CACHE_SHIFT);
 	u64 end = start + PAGE_CACHE_SIZE - 1;
 	size_t blocksize = page->mapping->host->i_sb->s_blocksize;
@@ -2712,12 +2715,12 @@ int extent_invalidatepage(struct extent_io_tree *tree,
 	if (start > end)
 		return 0;
 
-	lock_extent(tree, start, end, GFP_NOFS);
+	lock_extent_bits(tree, start, end, 0, &cached_state, GFP_NOFS);
 	wait_on_page_writeback(page);
 	clear_extent_bit(tree, start, end,
 			 EXTENT_LOCKED | EXTENT_DIRTY | EXTENT_DELALLOC |
 			 EXTENT_DO_ACCOUNTING,
-			 1, 1, NULL, GFP_NOFS);
+			 1, 1, &cached_state, GFP_NOFS);
 	return 0;
 }
 
@@ -2920,16 +2923,17 @@ sector_t extent_bmap(struct address_space *mapping, sector_t iblock,
 		get_extent_t *get_extent)
 {
 	struct inode *inode = mapping->host;
+	struct extent_state *cached_state = NULL;
 	u64 start = iblock << inode->i_blkbits;
 	sector_t sector = 0;
 	size_t blksize = (1 << inode->i_blkbits);
 	struct extent_map *em;
 
-	lock_extent(&BTRFS_I(inode)->io_tree, start, start + blksize - 1,
-		    GFP_NOFS);
+	lock_extent_bits(&BTRFS_I(inode)->io_tree, start, start + blksize - 1,
+			 0, &cached_state, GFP_NOFS);
 	em = get_extent(inode, NULL, 0, start, blksize, 0);
-	unlock_extent(&BTRFS_I(inode)->io_tree, start, start + blksize - 1,
-		      GFP_NOFS);
+	unlock_extent_cached(&BTRFS_I(inode)->io_tree, start,
+			     start + blksize - 1, &cached_state, GFP_NOFS);
 	if (!em || IS_ERR(em))
 		return 0;
 
@@ -2951,6 +2955,7 @@ int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 	u32 flags = 0;
 	u64 disko = 0;
 	struct extent_map *em = NULL;
+	struct extent_state *cached_state = NULL;
 	int end = 0;
 	u64 em_start = 0, em_len = 0;
 	unsigned long emflags;
@@ -2959,8 +2964,8 @@ int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 	if (len == 0)
 		return -EINVAL;
 
-	lock_extent(&BTRFS_I(inode)->io_tree, start, start + len,
-		GFP_NOFS);
+	lock_extent_bits(&BTRFS_I(inode)->io_tree, start, start + len, 0,
+			 &cached_state, GFP_NOFS);
 	em = get_extent(inode, NULL, 0, off, max - off, 0);
 	if (!em)
 		goto out;
@@ -3023,8 +3028,8 @@ int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 out_free:
 	free_extent_map(em);
 out:
-	unlock_extent(&BTRFS_I(inode)->io_tree, start, start + len,
-			GFP_NOFS);
+	unlock_extent_cached(&BTRFS_I(inode)->io_tree, start, start + len,
+			     &cached_state, GFP_NOFS);
 	return ret;
 }
 
@@ -3264,7 +3269,8 @@ int set_extent_buffer_dirty(struct extent_io_tree *tree,
 }
 
 int clear_extent_buffer_uptodate(struct extent_io_tree *tree,
-				struct extent_buffer *eb)
+				struct extent_buffer *eb,
+				struct extent_state **cached_state)
 {
 	unsigned long i;
 	struct page *page;
@@ -3274,7 +3280,7 @@ int clear_extent_buffer_uptodate(struct extent_io_tree *tree,
 	clear_bit(EXTENT_BUFFER_UPTODATE, &eb->bflags);
 
 	clear_extent_uptodate(tree, eb->start, eb->start + eb->len - 1,
-			      GFP_NOFS);
+			      cached_state, GFP_NOFS);
 	for (i = 0; i < num_pages; i++) {
 		page = extent_buffer_page(eb, i);
 		if (page)
@@ -3334,7 +3340,8 @@ int extent_range_uptodate(struct extent_io_tree *tree,
 }
 
 int extent_buffer_uptodate(struct extent_io_tree *tree,
-			   struct extent_buffer *eb)
+			   struct extent_buffer *eb,
+			   struct extent_state *cached_state)
 {
 	int ret = 0;
 	unsigned long num_pages;
@@ -3346,7 +3353,7 @@ int extent_buffer_uptodate(struct extent_io_tree *tree,
 		return 1;
 
 	ret = test_range_bit(tree, eb->start, eb->start + eb->len - 1,
-			   EXTENT_UPTODATE, 1, NULL);
+			   EXTENT_UPTODATE, 1, cached_state);
 	if (ret)
 		return ret;
 
diff --git a/fs/btrfs/extent_io.h b/fs/btrfs/extent_io.h
index 36de250..bbab481 100644
--- a/fs/btrfs/extent_io.h
+++ b/fs/btrfs/extent_io.h
@@ -163,6 +163,8 @@ int lock_extent(struct extent_io_tree *tree, u64 start, u64 end, gfp_t mask);
 int lock_extent_bits(struct extent_io_tree *tree, u64 start, u64 end,
 		     int bits, struct extent_state **cached, gfp_t mask);
 int unlock_extent(struct extent_io_tree *tree, u64 start, u64 end, gfp_t mask);
+int unlock_extent_cached(struct extent_io_tree *tree, u64 start, u64 end,
+			 struct extent_state **cached, gfp_t mask);
 int try_lock_extent(struct extent_io_tree *tree, u64 start, u64 end,
 		    gfp_t mask);
 int extent_read_full_page(struct extent_io_tree *tree, struct page *page,
@@ -196,7 +198,7 @@ int clear_extent_ordered(struct extent_io_tree *tree, u64 start, u64 end,
 int clear_extent_ordered_metadata(struct extent_io_tree *tree, u64 start,
 				  u64 end, gfp_t mask);
 int set_extent_delalloc(struct extent_io_tree *tree, u64 start, u64 end,
-		     gfp_t mask);
+			struct extent_state **cached_state, gfp_t mask);
 int set_extent_ordered(struct extent_io_tree *tree, u64 start, u64 end,
 		     gfp_t mask);
 int find_first_extent_bit(struct extent_io_tree *tree, u64 start,
@@ -281,9 +283,11 @@ int test_extent_buffer_dirty(struct extent_io_tree *tree,
 int set_extent_buffer_uptodate(struct extent_io_tree *tree,
 			       struct extent_buffer *eb);
 int clear_extent_buffer_uptodate(struct extent_io_tree *tree,
-				struct extent_buffer *eb);
+				struct extent_buffer *eb,
+				struct extent_state **cached_state);
 int extent_buffer_uptodate(struct extent_io_tree *tree,
-			   struct extent_buffer *eb);
+			   struct extent_buffer *eb,
+			   struct extent_state *cached_state);
 int map_extent_buffer(struct extent_buffer *eb, unsigned long offset,
 		      unsigned long min_len, char **token, char **map,
 		      unsigned long *map_start,
diff --git a/fs/btrfs/extent_map.c b/fs/btrfs/extent_map.c
index 28d87ba..454ca52 100644
--- a/fs/btrfs/extent_map.c
+++ b/fs/btrfs/extent_map.c
@@ -1,5 +1,4 @@
 #include <linux/err.h>
-#include <linux/gfp.h>
 #include <linux/slab.h>
 #include <linux/module.h>
 #include <linux/spinlock.h>
diff --git a/fs/btrfs/file-item.c b/fs/btrfs/file-item.c
index 9b99886..54a2550 100644
--- a/fs/btrfs/file-item.c
+++ b/fs/btrfs/file-item.c
@@ -17,6 +17,7 @@
  */
 
 #include <linux/bio.h>
+#include <linux/slab.h>
 #include <linux/pagemap.h>
 #include <linux/highmem.h>
 #include "ctree.h"
diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index 6ed434a..29ff749 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -28,6 +28,7 @@
 #include <linux/writeback.h>
 #include <linux/statfs.h>
 #include <linux/compat.h>
+#include <linux/slab.h>
 #include "ctree.h"
 #include "disk-io.h"
 #include "transaction.h"
@@ -123,7 +124,8 @@ static noinline int dirty_and_release_pages(struct btrfs_trans_handle *trans,
 		    root->sectorsize - 1) & ~((u64)root->sectorsize - 1);
 
 	end_of_last_block = start_pos + num_bytes - 1;
-	err = btrfs_set_extent_delalloc(inode, start_pos, end_of_last_block);
+	err = btrfs_set_extent_delalloc(inode, start_pos, end_of_last_block,
+					NULL);
 	if (err)
 		return err;
 
@@ -753,6 +755,7 @@ static noinline int prepare_pages(struct btrfs_root *root, struct file *file,
 			 loff_t pos, unsigned long first_index,
 			 unsigned long last_index, size_t write_bytes)
 {
+	struct extent_state *cached_state = NULL;
 	int i;
 	unsigned long index = pos >> PAGE_CACHE_SHIFT;
 	struct inode *inode = fdentry(file)->d_inode;
@@ -781,16 +784,18 @@ again:
 	}
 	if (start_pos < inode->i_size) {
 		struct btrfs_ordered_extent *ordered;
-		lock_extent(&BTRFS_I(inode)->io_tree,
-			    start_pos, last_pos - 1, GFP_NOFS);
+		lock_extent_bits(&BTRFS_I(inode)->io_tree,
+				 start_pos, last_pos - 1, 0, &cached_state,
+				 GFP_NOFS);
 		ordered = btrfs_lookup_first_ordered_extent(inode,
 							    last_pos - 1);
 		if (ordered &&
 		    ordered->file_offset + ordered->len > start_pos &&
 		    ordered->file_offset < last_pos) {
 			btrfs_put_ordered_extent(ordered);
-			unlock_extent(&BTRFS_I(inode)->io_tree,
-				      start_pos, last_pos - 1, GFP_NOFS);
+			unlock_extent_cached(&BTRFS_I(inode)->io_tree,
+					     start_pos, last_pos - 1,
+					     &cached_state, GFP_NOFS);
 			for (i = 0; i < num_pages; i++) {
 				unlock_page(pages[i]);
 				page_cache_release(pages[i]);
@@ -802,12 +807,13 @@ again:
 		if (ordered)
 			btrfs_put_ordered_extent(ordered);
 
-		clear_extent_bits(&BTRFS_I(inode)->io_tree, start_pos,
+		clear_extent_bit(&BTRFS_I(inode)->io_tree, start_pos,
 				  last_pos - 1, EXTENT_DIRTY | EXTENT_DELALLOC |
-				  EXTENT_DO_ACCOUNTING,
+				  EXTENT_DO_ACCOUNTING, 0, 0, &cached_state,
 				  GFP_NOFS);
-		unlock_extent(&BTRFS_I(inode)->io_tree,
-			      start_pos, last_pos - 1, GFP_NOFS);
+		unlock_extent_cached(&BTRFS_I(inode)->io_tree,
+				     start_pos, last_pos - 1, &cached_state,
+				     GFP_NOFS);
 	}
 	for (i = 0; i < num_pages; i++) {
 		clear_page_dirty_for_io(pages[i]);
diff --git a/fs/btrfs/free-space-cache.c b/fs/btrfs/free-space-cache.c
index dd831ed..f488fac 100644
--- a/fs/btrfs/free-space-cache.c
+++ b/fs/btrfs/free-space-cache.c
@@ -18,6 +18,7 @@
 
 #include <linux/pagemap.h>
 #include <linux/sched.h>
+#include <linux/slab.h>
 #include <linux/math64.h>
 #include "ctree.h"
 #include "free-space-cache.h"
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index c41db6d..2bfdc64 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -36,6 +36,7 @@
 #include <linux/xattr.h>
 #include <linux/posix_acl.h>
 #include <linux/falloc.h>
+#include <linux/slab.h>
 #include "compat.h"
 #include "ctree.h"
 #include "disk-io.h"
@@ -379,7 +380,8 @@ again:
 	 * change at any time if we discover bad compression ratios.
 	 */
 	if (!(BTRFS_I(inode)->flags & BTRFS_INODE_NOCOMPRESS) &&
-	    btrfs_test_opt(root, COMPRESS)) {
+	    (btrfs_test_opt(root, COMPRESS) ||
+	     (BTRFS_I(inode)->force_compress))) {
 		WARN_ON(pages);
 		pages = kzalloc(sizeof(struct page *) * nr_pages, GFP_NOFS);
 
@@ -483,8 +485,10 @@ again:
 		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, FORCE_COMPRESS) &&
+		    !(BTRFS_I(inode)->force_compress)) {
 			BTRFS_I(inode)->flags |= BTRFS_INODE_NOCOMPRESS;
+		}
 	}
 	if (will_compress) {
 		*num_added += 1;
@@ -570,8 +574,8 @@ retry:
 			unsigned long nr_written = 0;
 
 			lock_extent(io_tree, async_extent->start,
-				    async_extent->start +
-				    async_extent->ram_size - 1, GFP_NOFS);
+					 async_extent->start +
+					 async_extent->ram_size - 1, GFP_NOFS);
 
 			/* allocate blocks */
 			ret = cow_file_range(inode, async_cow->locked_page,
@@ -793,7 +797,7 @@ static noinline int cow_file_range(struct inode *inode,
 	while (disk_num_bytes > 0) {
 		unsigned long op;
 
-		cur_alloc_size = min(disk_num_bytes, root->fs_info->max_extent);
+		cur_alloc_size = disk_num_bytes;
 		ret = btrfs_reserve_extent(trans, root, cur_alloc_size,
 					   root->sectorsize, 0, alloc_hint,
 					   (u64)-1, &ins, 1);
@@ -1211,7 +1215,8 @@ static int run_delalloc_range(struct inode *inode, struct page *locked_page,
 	else if (BTRFS_I(inode)->flags & BTRFS_INODE_PREALLOC)
 		ret = run_delalloc_nocow(inode, locked_page, start, end,
 					 page_started, 0, nr_written);
-	else if (!btrfs_test_opt(root, COMPRESS))
+	else if (!btrfs_test_opt(root, COMPRESS) &&
+		 !(BTRFS_I(inode)->force_compress))
 		ret = cow_file_range(inode, locked_page, start, end,
 				      page_started, nr_written, 1);
 	else
@@ -1223,30 +1228,9 @@ static int run_delalloc_range(struct inode *inode, struct page *locked_page,
 static int btrfs_split_extent_hook(struct inode *inode,
 				    struct extent_state *orig, u64 split)
 {
-	struct btrfs_root *root = BTRFS_I(inode)->root;
-	u64 size;
-
 	if (!(orig->state & EXTENT_DELALLOC))
 		return 0;
 
-	size = orig->end - orig->start + 1;
-	if (size > root->fs_info->max_extent) {
-		u64 num_extents;
-		u64 new_size;
-
-		new_size = orig->end - split + 1;
-		num_extents = div64_u64(size + root->fs_info->max_extent - 1,
-					root->fs_info->max_extent);
-
-		/*
-		 * if we break a large extent up then leave oustanding_extents
-		 * be, since we've already accounted for the large extent.
-		 */
-		if (div64_u64(new_size + root->fs_info->max_extent - 1,
-			      root->fs_info->max_extent) < num_extents)
-			return 0;
-	}
-
 	spin_lock(&BTRFS_I(inode)->accounting_lock);
 	BTRFS_I(inode)->outstanding_extents++;
 	spin_unlock(&BTRFS_I(inode)->accounting_lock);
@@ -1264,38 +1248,10 @@ static int btrfs_merge_extent_hook(struct inode *inode,
 				   struct extent_state *new,
 				   struct extent_state *other)
 {
-	struct btrfs_root *root = BTRFS_I(inode)->root;
-	u64 new_size, old_size;
-	u64 num_extents;
-
 	/* not delalloc, ignore it */
 	if (!(other->state & EXTENT_DELALLOC))
 		return 0;
 
-	old_size = other->end - other->start + 1;
-	if (new->start < other->start)
-		new_size = other->end - new->start + 1;
-	else
-		new_size = new->end - other->start + 1;
-
-	/* we're not bigger than the max, unreserve the space and go */
-	if (new_size <= root->fs_info->max_extent) {
-		spin_lock(&BTRFS_I(inode)->accounting_lock);
-		BTRFS_I(inode)->outstanding_extents--;
-		spin_unlock(&BTRFS_I(inode)->accounting_lock);
-		return 0;
-	}
-
-	/*
-	 * If we grew by another max_extent, just return, we want to keep that
-	 * reserved amount.
-	 */
-	num_extents = div64_u64(old_size + root->fs_info->max_extent - 1,
-				root->fs_info->max_extent);
-	if (div64_u64(new_size + root->fs_info->max_extent - 1,
-		      root->fs_info->max_extent) > num_extents)
-		return 0;
-
 	spin_lock(&BTRFS_I(inode)->accounting_lock);
 	BTRFS_I(inode)->outstanding_extents--;
 	spin_unlock(&BTRFS_I(inode)->accounting_lock);
@@ -1324,6 +1280,7 @@ static int btrfs_set_bit_hook(struct inode *inode, u64 start, u64 end,
 		BTRFS_I(inode)->outstanding_extents++;
 		spin_unlock(&BTRFS_I(inode)->accounting_lock);
 		btrfs_delalloc_reserve_space(root, inode, end - start + 1);
+
 		spin_lock(&root->fs_info->delalloc_lock);
 		BTRFS_I(inode)->delalloc_bytes += end - start + 1;
 		root->fs_info->delalloc_bytes += end - start + 1;
@@ -1352,6 +1309,7 @@ static int btrfs_clear_bit_hook(struct inode *inode,
 
 		if (bits & EXTENT_DO_ACCOUNTING) {
 			spin_lock(&BTRFS_I(inode)->accounting_lock);
+			WARN_ON(!BTRFS_I(inode)->outstanding_extents);
 			BTRFS_I(inode)->outstanding_extents--;
 			spin_unlock(&BTRFS_I(inode)->accounting_lock);
 			btrfs_unreserve_metadata_for_delalloc(root, inode, 1);
@@ -1508,12 +1466,13 @@ static noinline int add_pending_csums(struct btrfs_trans_handle *trans,
 	return 0;
 }
 
-int btrfs_set_extent_delalloc(struct inode *inode, u64 start, u64 end)
+int btrfs_set_extent_delalloc(struct inode *inode, u64 start, u64 end,
+			      struct extent_state **cached_state)
 {
 	if ((end & (PAGE_CACHE_SIZE - 1)) == 0)
 		WARN_ON(1);
 	return set_extent_delalloc(&BTRFS_I(inode)->io_tree, start, end,
-				   GFP_NOFS);
+				   cached_state, GFP_NOFS);
 }
 
 /* see btrfs_writepage_start_hook for details on why this is required */
@@ -1526,6 +1485,7 @@ static void btrfs_writepage_fixup_worker(struct btrfs_work *work)
 {
 	struct btrfs_writepage_fixup *fixup;
 	struct btrfs_ordered_extent *ordered;
+	struct extent_state *cached_state = NULL;
 	struct page *page;
 	struct inode *inode;
 	u64 page_start;
@@ -1544,7 +1504,8 @@ again:
 	page_start = page_offset(page);
 	page_end = page_offset(page) + PAGE_CACHE_SIZE - 1;
 
-	lock_extent(&BTRFS_I(inode)->io_tree, page_start, page_end, GFP_NOFS);
+	lock_extent_bits(&BTRFS_I(inode)->io_tree, page_start, page_end, 0,
+			 &cached_state, GFP_NOFS);
 
 	/* already ordered? We're done */
 	if (PagePrivate2(page))
@@ -1552,17 +1513,18 @@ again:
 
 	ordered = btrfs_lookup_ordered_extent(inode, page_start);
 	if (ordered) {
-		unlock_extent(&BTRFS_I(inode)->io_tree, page_start,
-			      page_end, GFP_NOFS);
+		unlock_extent_cached(&BTRFS_I(inode)->io_tree, page_start,
+				     page_end, &cached_state, GFP_NOFS);
 		unlock_page(page);
 		btrfs_start_ordered_extent(inode, ordered, 1);
 		goto again;
 	}
 
-	btrfs_set_extent_delalloc(inode, page_start, page_end);
+	btrfs_set_extent_delalloc(inode, page_start, page_end, &cached_state);
 	ClearPageChecked(page);
 out:
-	unlock_extent(&BTRFS_I(inode)->io_tree, page_start, page_end, GFP_NOFS);
+	unlock_extent_cached(&BTRFS_I(inode)->io_tree, page_start, page_end,
+			     &cached_state, GFP_NOFS);
 out_page:
 	unlock_page(page);
 	page_cache_release(page);
@@ -1691,14 +1653,14 @@ static int btrfs_finish_ordered_io(struct inode *inode, u64 start, u64 end)
 	struct btrfs_trans_handle *trans;
 	struct btrfs_ordered_extent *ordered_extent = NULL;
 	struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree;
+	struct extent_state *cached_state = NULL;
 	int compressed = 0;
 	int ret;
 
-	ret = btrfs_dec_test_ordered_pending(inode, start, end - start + 1);
+	ret = btrfs_dec_test_ordered_pending(inode, &ordered_extent, start,
+					     end - start + 1);
 	if (!ret)
 		return 0;
-
-	ordered_extent = btrfs_lookup_ordered_extent(inode, start);
 	BUG_ON(!ordered_extent);
 
 	if (test_bit(BTRFS_ORDERED_NOCOW, &ordered_extent->flags)) {
@@ -1713,9 +1675,9 @@ static int btrfs_finish_ordered_io(struct inode *inode, u64 start, u64 end)
 		goto out;
 	}
 
-	lock_extent(io_tree, ordered_extent->file_offset,
-		    ordered_extent->file_offset + ordered_extent->len - 1,
-		    GFP_NOFS);
+	lock_extent_bits(io_tree, ordered_extent->file_offset,
+			 ordered_extent->file_offset + ordered_extent->len - 1,
+			 0, &cached_state, GFP_NOFS);
 
 	trans = btrfs_join_transaction(root, 1);
 
@@ -1742,9 +1704,10 @@ static int btrfs_finish_ordered_io(struct inode *inode, u64 start, u64 end)
 				   ordered_extent->len);
 		BUG_ON(ret);
 	}
-	unlock_extent(io_tree, ordered_extent->file_offset,
-		    ordered_extent->file_offset + ordered_extent->len - 1,
-		    GFP_NOFS);
+	unlock_extent_cached(io_tree, ordered_extent->file_offset,
+			     ordered_extent->file_offset +
+			     ordered_extent->len - 1, &cached_state, GFP_NOFS);
+
 	add_pending_csums(trans, inode, ordered_extent->file_offset,
 			  &ordered_extent->list);
 
@@ -2153,7 +2116,7 @@ void btrfs_orphan_cleanup(struct btrfs_root *root)
 		found_key.objectid = found_key.offset;
 		found_key.type = BTRFS_INODE_ITEM_KEY;
 		found_key.offset = 0;
-		inode = btrfs_iget(root->fs_info->sb, &found_key, root);
+		inode = btrfs_iget(root->fs_info->sb, &found_key, root, NULL);
 		if (IS_ERR(inode))
 			break;
 
@@ -3081,6 +3044,7 @@ static int btrfs_truncate_page(struct address_space *mapping, loff_t from)
 	struct btrfs_root *root = BTRFS_I(inode)->root;
 	struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree;
 	struct btrfs_ordered_extent *ordered;
+	struct extent_state *cached_state = NULL;
 	char *kaddr;
 	u32 blocksize = root->sectorsize;
 	pgoff_t index = from >> PAGE_CACHE_SHIFT;
@@ -3127,12 +3091,14 @@ again:
 	}
 	wait_on_page_writeback(page);
 
-	lock_extent(io_tree, page_start, page_end, GFP_NOFS);
+	lock_extent_bits(io_tree, page_start, page_end, 0, &cached_state,
+			 GFP_NOFS);
 	set_page_extent_mapped(page);
 
 	ordered = btrfs_lookup_ordered_extent(inode, page_start);
 	if (ordered) {
-		unlock_extent(io_tree, page_start, page_end, GFP_NOFS);
+		unlock_extent_cached(io_tree, page_start, page_end,
+				     &cached_state, GFP_NOFS);
 		unlock_page(page);
 		page_cache_release(page);
 		btrfs_start_ordered_extent(inode, ordered, 1);
@@ -3140,13 +3106,15 @@ again:
 		goto again;
 	}
 
-	clear_extent_bits(&BTRFS_I(inode)->io_tree, page_start, page_end,
+	clear_extent_bit(&BTRFS_I(inode)->io_tree, page_start, page_end,
 			  EXTENT_DIRTY | EXTENT_DELALLOC | EXTENT_DO_ACCOUNTING,
-			  GFP_NOFS);
+			  0, 0, &cached_state, GFP_NOFS);
 
-	ret = btrfs_set_extent_delalloc(inode, page_start, page_end);
+	ret = btrfs_set_extent_delalloc(inode, page_start, page_end,
+					&cached_state);
 	if (ret) {
-		unlock_extent(io_tree, page_start, page_end, GFP_NOFS);
+		unlock_extent_cached(io_tree, page_start, page_end,
+				     &cached_state, GFP_NOFS);
 		goto out_unlock;
 	}
 
@@ -3159,7 +3127,8 @@ again:
 	}
 	ClearPageChecked(page);
 	set_page_dirty(page);
-	unlock_extent(io_tree, page_start, page_end, GFP_NOFS);
+	unlock_extent_cached(io_tree, page_start, page_end, &cached_state,
+			     GFP_NOFS);
 
 out_unlock:
 	if (ret)
@@ -3177,6 +3146,7 @@ int btrfs_cont_expand(struct inode *inode, loff_t size)
 	struct btrfs_root *root = BTRFS_I(inode)->root;
 	struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree;
 	struct extent_map *em;
+	struct extent_state *cached_state = NULL;
 	u64 mask = root->sectorsize - 1;
 	u64 hole_start = (inode->i_size + mask) & ~mask;
 	u64 block_end = (size + mask) & ~mask;
@@ -3192,11 +3162,13 @@ int btrfs_cont_expand(struct inode *inode, loff_t size)
 		struct btrfs_ordered_extent *ordered;
 		btrfs_wait_ordered_range(inode, hole_start,
 					 block_end - hole_start);
-		lock_extent(io_tree, hole_start, block_end - 1, GFP_NOFS);
+		lock_extent_bits(io_tree, hole_start, block_end - 1, 0,
+				 &cached_state, GFP_NOFS);
 		ordered = btrfs_lookup_ordered_extent(inode, hole_start);
 		if (!ordered)
 			break;
-		unlock_extent(io_tree, hole_start, block_end - 1, GFP_NOFS);
+		unlock_extent_cached(io_tree, hole_start, block_end - 1,
+				     &cached_state, GFP_NOFS);
 		btrfs_put_ordered_extent(ordered);
 	}
 
@@ -3241,7 +3213,8 @@ int btrfs_cont_expand(struct inode *inode, loff_t size)
 			break;
 	}
 
-	unlock_extent(io_tree, hole_start, block_end - 1, GFP_NOFS);
+	unlock_extent_cached(io_tree, hole_start, block_end - 1, &cached_state,
+			     GFP_NOFS);
 	return err;
 }
 
@@ -3639,6 +3612,7 @@ static noinline void init_btrfs_i(struct inode *inode)
 	bi->index_cnt = (u64)-1;
 	bi->last_unlink_trans = 0;
 	bi->ordered_data_close = 0;
+	bi->force_compress = 0;
 	extent_map_tree_init(&BTRFS_I(inode)->extent_tree, GFP_NOFS);
 	extent_io_tree_init(&BTRFS_I(inode)->io_tree,
 			     inode->i_mapping, GFP_NOFS);
@@ -3687,7 +3661,7 @@ static struct inode *btrfs_iget_locked(struct super_block *s,
  * Returns in *is_new if the inode was read from disk
  */
 struct inode *btrfs_iget(struct super_block *s, struct btrfs_key *location,
-			 struct btrfs_root *root)
+			 struct btrfs_root *root, int *new)
 {
 	struct inode *inode;
 
@@ -3702,6 +3676,8 @@ struct inode *btrfs_iget(struct super_block *s, struct btrfs_key *location,
 
 		inode_tree_add(inode);
 		unlock_new_inode(inode);
+		if (new)
+			*new = 1;
 	}
 
 	return inode;
@@ -3754,7 +3730,7 @@ struct inode *btrfs_lookup_dentry(struct inode *dir, struct dentry *dentry)
 		return NULL;
 
 	if (location.type == BTRFS_INODE_ITEM_KEY) {
-		inode = btrfs_iget(dir->i_sb, &location, root);
+		inode = btrfs_iget(dir->i_sb, &location, root, NULL);
 		return inode;
 	}
 
@@ -3769,7 +3745,7 @@ struct inode *btrfs_lookup_dentry(struct inode *dir, struct dentry *dentry)
 		else
 			inode = new_simple_dir(dir->i_sb, &location, sub_root);
 	} else {
-		inode = btrfs_iget(dir->i_sb, &location, sub_root);
+		inode = btrfs_iget(dir->i_sb, &location, sub_root, NULL);
 	}
 	srcu_read_unlock(&root->fs_info->subvol_srcu, index);
 
@@ -4501,7 +4477,7 @@ static int btrfs_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 	err = btrfs_find_free_objectid(trans, root, dir->i_ino, &objectid);
 	if (err) {
 		err = -ENOSPC;
-		goto out_unlock;
+		goto out_fail;
 	}
 
 	inode = btrfs_new_inode(trans, root, dir, dentry->d_name.name,
@@ -4979,6 +4955,7 @@ static void btrfs_invalidatepage(struct page *page, unsigned long offset)
 {
 	struct extent_io_tree *tree;
 	struct btrfs_ordered_extent *ordered;
+	struct extent_state *cached_state = NULL;
 	u64 page_start = page_offset(page);
 	u64 page_end = page_start + PAGE_CACHE_SIZE - 1;
 
@@ -4997,7 +4974,8 @@ static void btrfs_invalidatepage(struct page *page, unsigned long offset)
 		btrfs_releasepage(page, GFP_NOFS);
 		return;
 	}
-	lock_extent(tree, page_start, page_end, GFP_NOFS);
+	lock_extent_bits(tree, page_start, page_end, 0, &cached_state,
+			 GFP_NOFS);
 	ordered = btrfs_lookup_ordered_extent(page->mapping->host,
 					   page_offset(page));
 	if (ordered) {
@@ -5008,7 +4986,7 @@ static void btrfs_invalidatepage(struct page *page, unsigned long offset)
 		clear_extent_bit(tree, page_start, page_end,
 				 EXTENT_DIRTY | EXTENT_DELALLOC |
 				 EXTENT_LOCKED | EXTENT_DO_ACCOUNTING, 1, 0,
-				 NULL, GFP_NOFS);
+				 &cached_state, GFP_NOFS);
 		/*
 		 * whoever cleared the private bit is responsible
 		 * for the finish_ordered_io
@@ -5018,11 +4996,13 @@ static void btrfs_invalidatepage(struct page *page, unsigned long offset)
 						page_start, page_end);
 		}
 		btrfs_put_ordered_extent(ordered);
-		lock_extent(tree, page_start, page_end, GFP_NOFS);
+		cached_state = NULL;
+		lock_extent_bits(tree, page_start, page_end, 0, &cached_state,
+				 GFP_NOFS);
 	}
 	clear_extent_bit(tree, page_start, page_end,
 		 EXTENT_LOCKED | EXTENT_DIRTY | EXTENT_DELALLOC |
-		 EXTENT_DO_ACCOUNTING, 1, 1, NULL, GFP_NOFS);
+		 EXTENT_DO_ACCOUNTING, 1, 1, &cached_state, GFP_NOFS);
 	__btrfs_releasepage(page, GFP_NOFS);
 
 	ClearPageChecked(page);
@@ -5055,6 +5035,7 @@ int btrfs_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf)
 	struct btrfs_root *root = BTRFS_I(inode)->root;
 	struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree;
 	struct btrfs_ordered_extent *ordered;
+	struct extent_state *cached_state = NULL;
 	char *kaddr;
 	unsigned long zero_start;
 	loff_t size;
@@ -5093,7 +5074,8 @@ again:
 	}
 	wait_on_page_writeback(page);
 
-	lock_extent(io_tree, page_start, page_end, GFP_NOFS);
+	lock_extent_bits(io_tree, page_start, page_end, 0, &cached_state,
+			 GFP_NOFS);
 	set_page_extent_mapped(page);
 
 	/*
@@ -5102,7 +5084,8 @@ again:
 	 */
 	ordered = btrfs_lookup_ordered_extent(inode, page_start);
 	if (ordered) {
-		unlock_extent(io_tree, page_start, page_end, GFP_NOFS);
+		unlock_extent_cached(io_tree, page_start, page_end,
+				     &cached_state, GFP_NOFS);
 		unlock_page(page);
 		btrfs_start_ordered_extent(inode, ordered, 1);
 		btrfs_put_ordered_extent(ordered);
@@ -5116,13 +5099,15 @@ again:
 	 * is probably a better way to do this, but for now keep consistent with
 	 * prepare_pages in the normal write path.
 	 */
-	clear_extent_bits(&BTRFS_I(inode)->io_tree, page_start, page_end,
+	clear_extent_bit(&BTRFS_I(inode)->io_tree, page_start, page_end,
 			  EXTENT_DIRTY | EXTENT_DELALLOC | EXTENT_DO_ACCOUNTING,
-			  GFP_NOFS);
+			  0, 0, &cached_state, GFP_NOFS);
 
-	ret = btrfs_set_extent_delalloc(inode, page_start, page_end);
+	ret = btrfs_set_extent_delalloc(inode, page_start, page_end,
+					&cached_state);
 	if (ret) {
-		unlock_extent(io_tree, page_start, page_end, GFP_NOFS);
+		unlock_extent_cached(io_tree, page_start, page_end,
+				     &cached_state, GFP_NOFS);
 		ret = VM_FAULT_SIGBUS;
 		btrfs_free_reserved_data_space(root, inode, PAGE_CACHE_SIZE);
 		goto out_unlock;
@@ -5148,7 +5133,7 @@ again:
 	BTRFS_I(inode)->last_trans = root->fs_info->generation;
 	BTRFS_I(inode)->last_sub_trans = BTRFS_I(inode)->root->log_transid;
 
-	unlock_extent(io_tree, page_start, page_end, GFP_NOFS);
+	unlock_extent_cached(io_tree, page_start, page_end, &cached_state, GFP_NOFS);
 
 out_unlock:
 	btrfs_unreserve_metadata_for_delalloc(root, inode, 1);
@@ -5353,7 +5338,6 @@ free:
 void btrfs_drop_inode(struct inode *inode)
 {
 	struct btrfs_root *root = BTRFS_I(inode)->root;
-
 	if (inode->i_nlink > 0 && btrfs_root_refs(&root->root_item) == 0)
 		generic_delete_inode(inode);
 	else
@@ -5757,18 +5741,15 @@ static int prealloc_file_range(struct inode *inode, u64 start, u64 end,
 	struct btrfs_trans_handle *trans;
 	struct btrfs_root *root = BTRFS_I(inode)->root;
 	struct btrfs_key ins;
-	u64 alloc_size;
 	u64 cur_offset = start;
 	u64 num_bytes = end - start;
 	int ret = 0;
 	u64 i_size;
 
 	while (num_bytes > 0) {
-		alloc_size = min(num_bytes, root->fs_info->max_extent);
-
 		trans = btrfs_start_transaction(root, 1);
 
-		ret = btrfs_reserve_extent(trans, root, alloc_size,
+		ret = btrfs_reserve_extent(trans, root, num_bytes,
 					   root->sectorsize, 0, alloc_hint,
 					   (u64)-1, &ins, 1);
 		if (ret) {
@@ -5827,6 +5808,7 @@ stop_trans:
 static long btrfs_fallocate(struct inode *inode, int mode,
 			    loff_t offset, loff_t len)
 {
+	struct extent_state *cached_state = NULL;
 	u64 cur_offset;
 	u64 last_byte;
 	u64 alloc_start;
@@ -5865,16 +5847,17 @@ static long btrfs_fallocate(struct inode *inode, int mode,
 		/* the extent lock is ordered inside the running
 		 * transaction
 		 */
-		lock_extent(&BTRFS_I(inode)->io_tree, alloc_start, locked_end,
-			    GFP_NOFS);
+		lock_extent_bits(&BTRFS_I(inode)->io_tree, alloc_start,
+				 locked_end, 0, &cached_state, GFP_NOFS);
 		ordered = btrfs_lookup_first_ordered_extent(inode,
 							    alloc_end - 1);
 		if (ordered &&
 		    ordered->file_offset + ordered->len > alloc_start &&
 		    ordered->file_offset < alloc_end) {
 			btrfs_put_ordered_extent(ordered);
-			unlock_extent(&BTRFS_I(inode)->io_tree,
-				      alloc_start, locked_end, GFP_NOFS);
+			unlock_extent_cached(&BTRFS_I(inode)->io_tree,
+					     alloc_start, locked_end,
+					     &cached_state, GFP_NOFS);
 			/*
 			 * we can't wait on the range with the transaction
 			 * running or with the extent lock held
@@ -5916,8 +5899,8 @@ static long btrfs_fallocate(struct inode *inode, int mode,
 			break;
 		}
 	}
-	unlock_extent(&BTRFS_I(inode)->io_tree, alloc_start, locked_end,
-		      GFP_NOFS);
+	unlock_extent_cached(&BTRFS_I(inode)->io_tree, alloc_start, locked_end,
+			     &cached_state, GFP_NOFS);
 
 	btrfs_free_reserved_data_space(BTRFS_I(inode)->root, inode,
 				       alloc_end - alloc_start);
diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c
index 645a179..e84ef60 100644
--- a/fs/btrfs/ioctl.c
+++ b/fs/btrfs/ioctl.c
@@ -39,6 +39,7 @@
 #include <linux/security.h>
 #include <linux/xattr.h>
 #include <linux/vmalloc.h>
+#include <linux/slab.h>
 #include "compat.h"
 #include "ctree.h"
 #include "disk-io.h"
@@ -474,7 +475,79 @@ out_unlock:
 	return error;
 }
 
-static int btrfs_defrag_file(struct file *file)
+static int should_defrag_range(struct inode *inode, u64 start, u64 len,
+			       int thresh, u64 *last_len, u64 *skip,
+			       u64 *defrag_end)
+{
+	struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree;
+	struct extent_map *em = NULL;
+	struct extent_map_tree *em_tree = &BTRFS_I(inode)->extent_tree;
+	int ret = 1;
+
+
+	if (thresh == 0)
+		thresh = 256 * 1024;
+
+	/*
+	 * make sure that once we start defragging and extent, we keep on
+	 * defragging it
+	 */
+	if (start < *defrag_end)
+		return 1;
+
+	*skip = 0;
+
+	/*
+	 * hopefully we have this extent in the tree already, try without
+	 * the full extent lock
+	 */
+	read_lock(&em_tree->lock);
+	em = lookup_extent_mapping(em_tree, start, len);
+	read_unlock(&em_tree->lock);
+
+	if (!em) {
+		/* get the big lock and read metadata off disk */
+		lock_extent(io_tree, start, start + len - 1, GFP_NOFS);
+		em = btrfs_get_extent(inode, NULL, 0, start, len, 0);
+		unlock_extent(io_tree, start, start + len - 1, GFP_NOFS);
+
+		if (IS_ERR(em))
+			return 0;
+	}
+
+	/* this will cover holes, and inline extents */
+	if (em->block_start >= EXTENT_MAP_LAST_BYTE)
+		ret = 0;
+
+	/*
+	 * we hit a real extent, if it is big don't bother defragging it again
+	 */
+	if ((*last_len == 0 || *last_len >= thresh) && em->len >= thresh)
+		ret = 0;
+
+	/*
+	 * last_len ends up being a counter of how many bytes we've defragged.
+	 * every time we choose not to defrag an extent, we reset *last_len
+	 * so that the next tiny extent will force a defrag.
+	 *
+	 * The end result of this is that tiny extents before a single big
+	 * extent will force at least part of that big extent to be defragged.
+	 */
+	if (ret) {
+		*last_len += len;
+		*defrag_end = extent_map_end(em);
+	} else {
+		*last_len = 0;
+		*skip = extent_map_end(em);
+		*defrag_end = 0;
+	}
+
+	free_extent_map(em);
+	return ret;
+}
+
+static int btrfs_defrag_file(struct file *file,
+			     struct btrfs_ioctl_defrag_range_args *range)
 {
 	struct inode *inode = fdentry(file)->d_inode;
 	struct btrfs_root *root = BTRFS_I(inode)->root;
@@ -486,37 +559,96 @@ static int btrfs_defrag_file(struct file *file)
 	unsigned long total_read = 0;
 	u64 page_start;
 	u64 page_end;
+	u64 last_len = 0;
+	u64 skip = 0;
+	u64 defrag_end = 0;
 	unsigned long i;
 	int ret;
 
-	ret = btrfs_check_data_free_space(root, inode, inode->i_size);
-	if (ret)
-		return -ENOSPC;
+	if (inode->i_size == 0)
+		return 0;
+
+	if (range->start + range->len > range->start) {
+		last_index = min_t(u64, inode->i_size - 1,
+			 range->start + range->len - 1) >> PAGE_CACHE_SHIFT;
+	} else {
+		last_index = (inode->i_size - 1) >> PAGE_CACHE_SHIFT;
+	}
+
+	i = range->start >> PAGE_CACHE_SHIFT;
+	while (i <= last_index) {
+		if (!should_defrag_range(inode, (u64)i << PAGE_CACHE_SHIFT,
+					PAGE_CACHE_SIZE,
+					range->extent_thresh,
+					&last_len, &skip,
+					&defrag_end)) {
+			unsigned long next;
+			/*
+			 * the should_defrag function tells us how much to skip
+			 * bump our counter by the suggested amount
+			 */
+			next = (skip + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT;
+			i = max(i + 1, next);
+			continue;
+		}
 
-	mutex_lock(&inode->i_mutex);
-	last_index = inode->i_size >> PAGE_CACHE_SHIFT;
-	for (i = 0; i <= last_index; i++) {
 		if (total_read % ra_pages == 0) {
 			btrfs_force_ra(inode->i_mapping, &file->f_ra, file, i,
 				       min(last_index, i + ra_pages - 1));
 		}
 		total_read++;
+		mutex_lock(&inode->i_mutex);
+		if (range->flags & BTRFS_DEFRAG_RANGE_COMPRESS)
+			BTRFS_I(inode)->force_compress = 1;
+
+		ret = btrfs_check_data_free_space(root, inode, PAGE_CACHE_SIZE);
+		if (ret) {
+			ret = -ENOSPC;
+			break;
+		}
+
+		ret = btrfs_reserve_metadata_for_delalloc(root, inode, 1);
+		if (ret) {
+			btrfs_free_reserved_data_space(root, inode,
+						       PAGE_CACHE_SIZE);
+			ret = -ENOSPC;
+			break;
+		}
 again:
+		if (inode->i_size == 0 ||
+		    i > ((inode->i_size - 1) >> PAGE_CACHE_SHIFT)) {
+			ret = 0;
+			goto err_reservations;
+		}
+
 		page = grab_cache_page(inode->i_mapping, i);
 		if (!page)
-			goto out_unlock;
+			goto err_reservations;
+
 		if (!PageUptodate(page)) {
 			btrfs_readpage(NULL, page);
 			lock_page(page);
 			if (!PageUptodate(page)) {
 				unlock_page(page);
 				page_cache_release(page);
-				goto out_unlock;
+				goto err_reservations;
 			}
 		}
 
+		if (page->mapping != inode->i_mapping) {
+			unlock_page(page);
+			page_cache_release(page);
+			goto again;
+		}
+
 		wait_on_page_writeback(page);
 
+		if (PageDirty(page)) {
+			btrfs_free_reserved_data_space(root, inode,
+						       PAGE_CACHE_SIZE);
+			goto loop_unlock;
+		}
+
 		page_start = (u64)page->index << PAGE_CACHE_SHIFT;
 		page_end = page_start + PAGE_CACHE_SIZE - 1;
 		lock_extent(io_tree, page_start, page_end, GFP_NOFS);
@@ -537,18 +669,54 @@ again:
 		 * page if it is dirtied again later
 		 */
 		clear_page_dirty_for_io(page);
+		clear_extent_bits(&BTRFS_I(inode)->io_tree, page_start,
+				  page_end, EXTENT_DIRTY | EXTENT_DELALLOC |
+				  EXTENT_DO_ACCOUNTING, GFP_NOFS);
 
-		btrfs_set_extent_delalloc(inode, page_start, page_end);
+		btrfs_set_extent_delalloc(inode, page_start, page_end, NULL);
+		ClearPageChecked(page);
 		set_page_dirty(page);
 		unlock_extent(io_tree, page_start, page_end, GFP_NOFS);
+
+loop_unlock:
 		unlock_page(page);
 		page_cache_release(page);
+		mutex_unlock(&inode->i_mutex);
+
+		btrfs_unreserve_metadata_for_delalloc(root, inode, 1);
 		balance_dirty_pages_ratelimited_nr(inode->i_mapping, 1);
+		i++;
+	}
+
+	if ((range->flags & BTRFS_DEFRAG_RANGE_START_IO))
+		filemap_flush(inode->i_mapping);
+
+	if ((range->flags & BTRFS_DEFRAG_RANGE_COMPRESS)) {
+		/* the filemap_flush will queue IO into the worker threads, but
+		 * we have to make sure the IO is actually started and that
+		 * ordered extents get created before we return
+		 */
+		atomic_inc(&root->fs_info->async_submit_draining);
+		while (atomic_read(&root->fs_info->nr_async_submits) ||
+		      atomic_read(&root->fs_info->async_delalloc_pages)) {
+			wait_event(root->fs_info->async_submit_wait,
+			   (atomic_read(&root->fs_info->nr_async_submits) == 0 &&
+			    atomic_read(&root->fs_info->async_delalloc_pages) == 0));
+		}
+		atomic_dec(&root->fs_info->async_submit_draining);
+
+		mutex_lock(&inode->i_mutex);
+		BTRFS_I(inode)->force_compress = 0;
+		mutex_unlock(&inode->i_mutex);
 	}
 
-out_unlock:
-	mutex_unlock(&inode->i_mutex);
 	return 0;
+
+err_reservations:
+	mutex_unlock(&inode->i_mutex);
+	btrfs_free_reserved_data_space(root, inode, PAGE_CACHE_SIZE);
+	btrfs_unreserve_metadata_for_delalloc(root, inode, 1);
+	return ret;
 }
 
 static noinline int btrfs_ioctl_resize(struct btrfs_root *root,
@@ -608,7 +776,7 @@ static noinline int btrfs_ioctl_resize(struct btrfs_root *root,
 			mod = 1;
 			sizestr++;
 		}
-		new_size = btrfs_parse_size(sizestr);
+		new_size = memparse(sizestr, NULL);
 		if (new_size == 0) {
 			ret = -EINVAL;
 			goto out_unlock;
@@ -743,6 +911,330 @@ out:
 	return ret;
 }
 
+static noinline int key_in_sk(struct btrfs_key *key,
+			      struct btrfs_ioctl_search_key *sk)
+{
+	struct btrfs_key test;
+	int ret;
+
+	test.objectid = sk->min_objectid;
+	test.type = sk->min_type;
+	test.offset = sk->min_offset;
+
+	ret = btrfs_comp_cpu_keys(key, &test);
+	if (ret < 0)
+		return 0;
+
+	test.objectid = sk->max_objectid;
+	test.type = sk->max_type;
+	test.offset = sk->max_offset;
+
+	ret = btrfs_comp_cpu_keys(key, &test);
+	if (ret > 0)
+		return 0;
+	return 1;
+}
+
+static noinline int copy_to_sk(struct btrfs_root *root,
+			       struct btrfs_path *path,
+			       struct btrfs_key *key,
+			       struct btrfs_ioctl_search_key *sk,
+			       char *buf,
+			       unsigned long *sk_offset,
+			       int *num_found)
+{
+	u64 found_transid;
+	struct extent_buffer *leaf;
+	struct btrfs_ioctl_search_header sh;
+	unsigned long item_off;
+	unsigned long item_len;
+	int nritems;
+	int i;
+	int slot;
+	int found = 0;
+	int ret = 0;
+
+	leaf = path->nodes[0];
+	slot = path->slots[0];
+	nritems = btrfs_header_nritems(leaf);
+
+	if (btrfs_header_generation(leaf) > sk->max_transid) {
+		i = nritems;
+		goto advance_key;
+	}
+	found_transid = btrfs_header_generation(leaf);
+
+	for (i = slot; i < nritems; i++) {
+		item_off = btrfs_item_ptr_offset(leaf, i);
+		item_len = btrfs_item_size_nr(leaf, i);
+
+		if (item_len > BTRFS_SEARCH_ARGS_BUFSIZE)
+			item_len = 0;
+
+		if (sizeof(sh) + item_len + *sk_offset >
+		    BTRFS_SEARCH_ARGS_BUFSIZE) {
+			ret = 1;
+			goto overflow;
+		}
+
+		btrfs_item_key_to_cpu(leaf, key, i);
+		if (!key_in_sk(key, sk))
+			continue;
+
+		sh.objectid = key->objectid;
+		sh.offset = key->offset;
+		sh.type = key->type;
+		sh.len = item_len;
+		sh.transid = found_transid;
+
+		/* copy search result header */
+		memcpy(buf + *sk_offset, &sh, sizeof(sh));
+		*sk_offset += sizeof(sh);
+
+		if (item_len) {
+			char *p = buf + *sk_offset;
+			/* copy the item */
+			read_extent_buffer(leaf, p,
+					   item_off, item_len);
+			*sk_offset += item_len;
+		}
+		found++;
+
+		if (*num_found >= sk->nr_items)
+			break;
+	}
+advance_key:
+	ret = 0;
+	if (key->offset < (u64)-1 && key->offset < sk->max_offset)
+		key->offset++;
+	else if (key->type < (u8)-1 && key->type < sk->max_type) {
+		key->offset = 0;
+		key->type++;
+	} else if (key->objectid < (u64)-1 && key->objectid < sk->max_objectid) {
+		key->offset = 0;
+		key->type = 0;
+		key->objectid++;
+	} else
+		ret = 1;
+overflow:
+	*num_found += found;
+	return ret;
+}
+
+static noinline int search_ioctl(struct inode *inode,
+				 struct btrfs_ioctl_search_args *args)
+{
+	struct btrfs_root *root;
+	struct btrfs_key key;
+	struct btrfs_key max_key;
+	struct btrfs_path *path;
+	struct btrfs_ioctl_search_key *sk = &args->key;
+	struct btrfs_fs_info *info = BTRFS_I(inode)->root->fs_info;
+	int ret;
+	int num_found = 0;
+	unsigned long sk_offset = 0;
+
+	path = btrfs_alloc_path();
+	if (!path)
+		return -ENOMEM;
+
+	if (sk->tree_id == 0) {
+		/* search the root of the inode that was passed */
+		root = BTRFS_I(inode)->root;
+	} else {
+		key.objectid = sk->tree_id;
+		key.type = BTRFS_ROOT_ITEM_KEY;
+		key.offset = (u64)-1;
+		root = btrfs_read_fs_root_no_name(info, &key);
+		if (IS_ERR(root)) {
+			printk(KERN_ERR "could not find root %llu\n",
+			       sk->tree_id);
+			btrfs_free_path(path);
+			return -ENOENT;
+		}
+	}
+
+	key.objectid = sk->min_objectid;
+	key.type = sk->min_type;
+	key.offset = sk->min_offset;
+
+	max_key.objectid = sk->max_objectid;
+	max_key.type = sk->max_type;
+	max_key.offset = sk->max_offset;
+
+	path->keep_locks = 1;
+
+	while(1) {
+		ret = btrfs_search_forward(root, &key, &max_key, path, 0,
+					   sk->min_transid);
+		if (ret != 0) {
+			if (ret > 0)
+				ret = 0;
+			goto err;
+		}
+		ret = copy_to_sk(root, path, &key, sk, args->buf,
+				 &sk_offset, &num_found);
+		btrfs_release_path(root, path);
+		if (ret || num_found >= sk->nr_items)
+			break;
+
+	}
+	ret = 0;
+err:
+	sk->nr_items = num_found;
+	btrfs_free_path(path);
+	return ret;
+}
+
+static noinline int btrfs_ioctl_tree_search(struct file *file,
+					   void __user *argp)
+{
+	 struct btrfs_ioctl_search_args *args;
+	 struct inode *inode;
+	 int ret;
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EPERM;
+
+	args = kmalloc(sizeof(*args), GFP_KERNEL);
+	if (!args)
+		return -ENOMEM;
+
+	if (copy_from_user(args, argp, sizeof(*args))) {
+		kfree(args);
+		return -EFAULT;
+	}
+	inode = fdentry(file)->d_inode;
+	ret = search_ioctl(inode, args);
+	if (ret == 0 && copy_to_user(argp, args, sizeof(*args)))
+		ret = -EFAULT;
+	kfree(args);
+	return ret;
+}
+
+/*
+ * Search INODE_REFs to identify path name of 'dirid' directory
+ * in a 'tree_id' tree. and sets path name to 'name'.
+ */
+static noinline int btrfs_search_path_in_tree(struct btrfs_fs_info *info,
+				u64 tree_id, u64 dirid, char *name)
+{
+	struct btrfs_root *root;
+	struct btrfs_key key;
+	char *ptr;
+	int ret = -1;
+	int slot;
+	int len;
+	int total_len = 0;
+	struct btrfs_inode_ref *iref;
+	struct extent_buffer *l;
+	struct btrfs_path *path;
+
+	if (dirid == BTRFS_FIRST_FREE_OBJECTID) {
+		name[0]='\0';
+		return 0;
+	}
+
+	path = btrfs_alloc_path();
+	if (!path)
+		return -ENOMEM;
+
+	ptr = &name[BTRFS_INO_LOOKUP_PATH_MAX];
+
+	key.objectid = tree_id;
+	key.type = BTRFS_ROOT_ITEM_KEY;
+	key.offset = (u64)-1;
+	root = btrfs_read_fs_root_no_name(info, &key);
+	if (IS_ERR(root)) {
+		printk(KERN_ERR "could not find root %llu\n", tree_id);
+		ret = -ENOENT;
+		goto out;
+	}
+
+	key.objectid = dirid;
+	key.type = BTRFS_INODE_REF_KEY;
+	key.offset = (u64)-1;
+
+	while(1) {
+		ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
+		if (ret < 0)
+			goto out;
+
+		l = path->nodes[0];
+		slot = path->slots[0];
+		if (ret > 0 && slot > 0)
+			slot--;
+		btrfs_item_key_to_cpu(l, &key, slot);
+
+		if (ret > 0 && (key.objectid != dirid ||
+				key.type != BTRFS_INODE_REF_KEY)) {
+			ret = -ENOENT;
+			goto out;
+		}
+
+		iref = btrfs_item_ptr(l, slot, struct btrfs_inode_ref);
+		len = btrfs_inode_ref_name_len(l, iref);
+		ptr -= len + 1;
+		total_len += len + 1;
+		if (ptr < name)
+			goto out;
+
+		*(ptr + len) = '/';
+		read_extent_buffer(l, ptr,(unsigned long)(iref + 1), len);
+
+		if (key.offset == BTRFS_FIRST_FREE_OBJECTID)
+			break;
+
+		btrfs_release_path(root, path);
+		key.objectid = key.offset;
+		key.offset = (u64)-1;
+		dirid = key.objectid;
+
+	}
+	if (ptr < name)
+		goto out;
+	memcpy(name, ptr, total_len);
+	name[total_len]='\0';
+	ret = 0;
+out:
+	btrfs_free_path(path);
+	return ret;
+}
+
+static noinline int btrfs_ioctl_ino_lookup(struct file *file,
+					   void __user *argp)
+{
+	 struct btrfs_ioctl_ino_lookup_args *args;
+	 struct inode *inode;
+	 int ret;
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EPERM;
+
+	args = kmalloc(sizeof(*args), GFP_KERNEL);
+	if (!args)
+		return -ENOMEM;
+
+	if (copy_from_user(args, argp, sizeof(*args))) {
+		kfree(args);
+		return -EFAULT;
+	}
+	inode = fdentry(file)->d_inode;
+
+	if (args->treeid == 0)
+		args->treeid = BTRFS_I(inode)->root->root_key.objectid;
+
+	ret = btrfs_search_path_in_tree(BTRFS_I(inode)->root->fs_info,
+					args->treeid, args->objectid,
+					args->name);
+
+	if (ret == 0 && copy_to_user(argp, args, sizeof(*args)))
+		ret = -EFAULT;
+
+	kfree(args);
+	return ret;
+}
+
 static noinline int btrfs_ioctl_snap_destroy(struct file *file,
 					     void __user *arg)
 {
@@ -849,10 +1341,11 @@ out:
 	return err;
 }
 
-static int btrfs_ioctl_defrag(struct file *file)
+static int btrfs_ioctl_defrag(struct file *file, void __user *argp)
 {
 	struct inode *inode = fdentry(file)->d_inode;
 	struct btrfs_root *root = BTRFS_I(inode)->root;
+	struct btrfs_ioctl_defrag_range_args *range;
 	int ret;
 
 	ret = mnt_want_write(file->f_path.mnt);
@@ -873,7 +1366,31 @@ static int btrfs_ioctl_defrag(struct file *file)
 			ret = -EINVAL;
 			goto out;
 		}
-		btrfs_defrag_file(file);
+
+		range = kzalloc(sizeof(*range), GFP_KERNEL);
+		if (!range) {
+			ret = -ENOMEM;
+			goto out;
+		}
+
+		if (argp) {
+			if (copy_from_user(range, argp,
+					   sizeof(*range))) {
+				ret = -EFAULT;
+				kfree(range);
+				goto out;
+			}
+			/* compression requires us to start the IO */
+			if ((range->flags & BTRFS_DEFRAG_RANGE_COMPRESS)) {
+				range->flags |= BTRFS_DEFRAG_RANGE_START_IO;
+				range->extent_thresh = (u32)-1;
+			}
+		} else {
+			/* the rest are all set to zero by kzalloc */
+			range->len = (u64)-1;
+		}
+		btrfs_defrag_file(file, range);
+		kfree(range);
 		break;
 	}
 out:
@@ -1274,6 +1791,157 @@ out:
 	return ret;
 }
 
+static long btrfs_ioctl_default_subvol(struct file *file, void __user *argp)
+{
+	struct inode *inode = fdentry(file)->d_inode;
+	struct btrfs_root *root = BTRFS_I(inode)->root;
+	struct btrfs_root *new_root;
+	struct btrfs_dir_item *di;
+	struct btrfs_trans_handle *trans;
+	struct btrfs_path *path;
+	struct btrfs_key location;
+	struct btrfs_disk_key disk_key;
+	struct btrfs_super_block *disk_super;
+	u64 features;
+	u64 objectid = 0;
+	u64 dir_id;
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EPERM;
+
+	if (copy_from_user(&objectid, argp, sizeof(objectid)))
+		return -EFAULT;
+
+	if (!objectid)
+		objectid = root->root_key.objectid;
+
+	location.objectid = objectid;
+	location.type = BTRFS_ROOT_ITEM_KEY;
+	location.offset = (u64)-1;
+
+	new_root = btrfs_read_fs_root_no_name(root->fs_info, &location);
+	if (IS_ERR(new_root))
+		return PTR_ERR(new_root);
+
+	if (btrfs_root_refs(&new_root->root_item) == 0)
+		return -ENOENT;
+
+	path = btrfs_alloc_path();
+	if (!path)
+		return -ENOMEM;
+	path->leave_spinning = 1;
+
+	trans = btrfs_start_transaction(root, 1);
+	if (!trans) {
+		btrfs_free_path(path);
+		return -ENOMEM;
+	}
+
+	dir_id = btrfs_super_root_dir(&root->fs_info->super_copy);
+	di = btrfs_lookup_dir_item(trans, root->fs_info->tree_root, path,
+				   dir_id, "default", 7, 1);
+	if (!di) {
+		btrfs_free_path(path);
+		btrfs_end_transaction(trans, root);
+		printk(KERN_ERR "Umm, you don't have the default dir item, "
+		       "this isn't going to work\n");
+		return -ENOENT;
+	}
+
+	btrfs_cpu_key_to_disk(&disk_key, &new_root->root_key);
+	btrfs_set_dir_item_key(path->nodes[0], di, &disk_key);
+	btrfs_mark_buffer_dirty(path->nodes[0]);
+	btrfs_free_path(path);
+
+	disk_super = &root->fs_info->super_copy;
+	features = btrfs_super_incompat_flags(disk_super);
+	if (!(features & BTRFS_FEATURE_INCOMPAT_DEFAULT_SUBVOL)) {
+		features |= BTRFS_FEATURE_INCOMPAT_DEFAULT_SUBVOL;
+		btrfs_set_super_incompat_flags(disk_super, features);
+	}
+	btrfs_end_transaction(trans, root);
+
+	return 0;
+}
+
+long btrfs_ioctl_space_info(struct btrfs_root *root, void __user *arg)
+{
+	struct btrfs_ioctl_space_args space_args;
+	struct btrfs_ioctl_space_info space;
+	struct btrfs_ioctl_space_info *dest;
+	struct btrfs_ioctl_space_info *dest_orig;
+	struct btrfs_ioctl_space_info *user_dest;
+	struct btrfs_space_info *info;
+	int alloc_size;
+	int ret = 0;
+	int slot_count = 0;
+
+	if (copy_from_user(&space_args,
+			   (struct btrfs_ioctl_space_args __user *)arg,
+			   sizeof(space_args)))
+		return -EFAULT;
+
+	/* first we count slots */
+	rcu_read_lock();
+	list_for_each_entry_rcu(info, &root->fs_info->space_info, list)
+		slot_count++;
+	rcu_read_unlock();
+
+	/* space_slots == 0 means they are asking for a count */
+	if (space_args.space_slots == 0) {
+		space_args.total_spaces = slot_count;
+		goto out;
+	}
+	alloc_size = sizeof(*dest) * slot_count;
+	/* we generally have at most 6 or so space infos, one for each raid
+	 * level.  So, a whole page should be more than enough for everyone
+	 */
+	if (alloc_size > PAGE_CACHE_SIZE)
+		return -ENOMEM;
+
+	space_args.total_spaces = 0;
+	dest = kmalloc(alloc_size, GFP_NOFS);
+	if (!dest)
+		return -ENOMEM;
+	dest_orig = dest;
+
+	/* now we have a buffer to copy into */
+	rcu_read_lock();
+	list_for_each_entry_rcu(info, &root->fs_info->space_info, list) {
+		/* make sure we don't copy more than we allocated
+		 * in our buffer
+		 */
+		if (slot_count == 0)
+			break;
+		slot_count--;
+
+		/* make sure userland has enough room in their buffer */
+		if (space_args.total_spaces >= space_args.space_slots)
+			break;
+
+		space.flags = info->flags;
+		space.total_bytes = info->total_bytes;
+		space.used_bytes = info->bytes_used;
+		memcpy(dest, &space, sizeof(space));
+		dest++;
+		space_args.total_spaces++;
+	}
+	rcu_read_unlock();
+
+	user_dest = (struct btrfs_ioctl_space_info *)
+		(arg + sizeof(struct btrfs_ioctl_space_args));
+
+	if (copy_to_user(user_dest, dest_orig, alloc_size))
+		ret = -EFAULT;
+
+	kfree(dest_orig);
+out:
+	if (ret == 0 && copy_to_user(arg, &space_args, sizeof(space_args)))
+		ret = -EFAULT;
+
+	return ret;
+}
+
 /*
  * there are many ways the trans_start and trans_end ioctls can lead
  * to deadlocks.  They should only be used by applications that
@@ -1320,8 +1988,12 @@ long btrfs_ioctl(struct file *file, unsigned int
 		return btrfs_ioctl_snap_create(file, argp, 1);
 	case BTRFS_IOC_SNAP_DESTROY:
 		return btrfs_ioctl_snap_destroy(file, argp);
+	case BTRFS_IOC_DEFAULT_SUBVOL:
+		return btrfs_ioctl_default_subvol(file, argp);
 	case BTRFS_IOC_DEFRAG:
-		return btrfs_ioctl_defrag(file);
+		return btrfs_ioctl_defrag(file, NULL);
+	case BTRFS_IOC_DEFRAG_RANGE:
+		return btrfs_ioctl_defrag(file, argp);
 	case BTRFS_IOC_RESIZE:
 		return btrfs_ioctl_resize(root, argp);
 	case BTRFS_IOC_ADD_DEV:
@@ -1338,6 +2010,12 @@ long btrfs_ioctl(struct file *file, unsigned int
 		return btrfs_ioctl_trans_start(file);
 	case BTRFS_IOC_TRANS_END:
 		return btrfs_ioctl_trans_end(file);
+	case BTRFS_IOC_TREE_SEARCH:
+		return btrfs_ioctl_tree_search(file, argp);
+	case BTRFS_IOC_INO_LOOKUP:
+		return btrfs_ioctl_ino_lookup(file, argp);
+	case BTRFS_IOC_SPACE_INFO:
+		return btrfs_ioctl_space_info(root, argp);
 	case BTRFS_IOC_SYNC:
 		btrfs_sync_fs(file->f_dentry->d_sb, 1);
 		return 0;
diff --git a/fs/btrfs/ioctl.h b/fs/btrfs/ioctl.h
index bc49914..424694a 100644
--- a/fs/btrfs/ioctl.h
+++ b/fs/btrfs/ioctl.h
@@ -30,12 +30,114 @@ struct btrfs_ioctl_vol_args {
 	char name[BTRFS_PATH_NAME_MAX + 1];
 };
 
+#define BTRFS_INO_LOOKUP_PATH_MAX 4080
+struct btrfs_ioctl_ino_lookup_args {
+	__u64 treeid;
+	__u64 objectid;
+	char name[BTRFS_INO_LOOKUP_PATH_MAX];
+};
+
+struct btrfs_ioctl_search_key {
+	/* which root are we searching.  0 is the tree of tree roots */
+	__u64 tree_id;
+
+	/* keys returned will be >= min and <= max */
+	__u64 min_objectid;
+	__u64 max_objectid;
+
+	/* keys returned will be >= min and <= max */
+	__u64 min_offset;
+	__u64 max_offset;
+
+	/* max and min transids to search for */
+	__u64 min_transid;
+	__u64 max_transid;
+
+	/* keys returned will be >= min and <= max */
+	__u32 min_type;
+	__u32 max_type;
+
+	/*
+	 * how many items did userland ask for, and how many are we
+	 * returning
+	 */
+	__u32 nr_items;
+
+	/* align to 64 bits */
+	__u32 unused;
+
+	/* some extra for later */
+	__u64 unused1;
+	__u64 unused2;
+	__u64 unused3;
+	__u64 unused4;
+};
+
+struct btrfs_ioctl_search_header {
+	__u64 transid;
+	__u64 objectid;
+	__u64 offset;
+	__u32 type;
+	__u32 len;
+};
+
+#define BTRFS_SEARCH_ARGS_BUFSIZE (4096 - sizeof(struct btrfs_ioctl_search_key))
+/*
+ * the buf is an array of search headers where
+ * each header is followed by the actual item
+ * the type field is expanded to 32 bits for alignment
+ */
+struct btrfs_ioctl_search_args {
+	struct btrfs_ioctl_search_key key;
+	char buf[BTRFS_SEARCH_ARGS_BUFSIZE];
+};
+
 struct btrfs_ioctl_clone_range_args {
   __s64 src_fd;
   __u64 src_offset, src_length;
   __u64 dest_offset;
 };
 
+/* flags for the defrag range ioctl */
+#define BTRFS_DEFRAG_RANGE_COMPRESS 1
+#define BTRFS_DEFRAG_RANGE_START_IO 2
+
+struct btrfs_ioctl_defrag_range_args {
+	/* start of the defrag operation */
+	__u64 start;
+
+	/* number of bytes to defrag, use (u64)-1 to say all */
+	__u64 len;
+
+	/*
+	 * flags for the operation, which can include turning
+	 * on compression for this one defrag
+	 */
+	__u64 flags;
+
+	/*
+	 * any extent bigger than this will be considered
+	 * already defragged.  Use 0 to take the kernel default
+	 * Use 1 to say every single extent must be rewritten
+	 */
+	__u32 extent_thresh;
+
+	/* spare for later */
+	__u32 unused[5];
+};
+
+struct btrfs_ioctl_space_info {
+	__u64 flags;
+	__u64 total_bytes;
+	__u64 used_bytes;
+};
+
+struct btrfs_ioctl_space_args {
+	__u64 space_slots;
+	__u64 total_spaces;
+	struct btrfs_ioctl_space_info spaces[0];
+};
+
 #define BTRFS_IOC_SNAP_CREATE _IOW(BTRFS_IOCTL_MAGIC, 1, \
 				   struct btrfs_ioctl_vol_args)
 #define BTRFS_IOC_DEFRAG _IOW(BTRFS_IOCTL_MAGIC, 2, \
@@ -67,4 +169,13 @@ struct btrfs_ioctl_clone_range_args {
 				   struct btrfs_ioctl_vol_args)
 #define BTRFS_IOC_SNAP_DESTROY _IOW(BTRFS_IOCTL_MAGIC, 15, \
 				struct btrfs_ioctl_vol_args)
+#define BTRFS_IOC_DEFRAG_RANGE _IOW(BTRFS_IOCTL_MAGIC, 16, \
+				struct btrfs_ioctl_defrag_range_args)
+#define BTRFS_IOC_TREE_SEARCH _IOWR(BTRFS_IOCTL_MAGIC, 17, \
+				   struct btrfs_ioctl_search_args)
+#define BTRFS_IOC_INO_LOOKUP _IOWR(BTRFS_IOCTL_MAGIC, 18, \
+				   struct btrfs_ioctl_ino_lookup_args)
+#define BTRFS_IOC_DEFAULT_SUBVOL _IOW(BTRFS_IOCTL_MAGIC, 19, u64)
+#define BTRFS_IOC_SPACE_INFO _IOWR(BTRFS_IOCTL_MAGIC, 20, \
+				    struct btrfs_ioctl_space_args)
 #endif
diff --git a/fs/btrfs/locking.c b/fs/btrfs/locking.c
index 1c36e5c..6151f2e 100644
--- a/fs/btrfs/locking.c
+++ b/fs/btrfs/locking.c
@@ -16,7 +16,6 @@
  * Boston, MA 021110-1307, USA.
  */
 #include <linux/sched.h>
-#include <linux/gfp.h>
 #include <linux/pagemap.h>
 #include <linux/spinlock.h>
 #include <linux/page-flags.h>
diff --git a/fs/btrfs/ordered-data.c b/fs/btrfs/ordered-data.c
index 5c2a9e7..a127c0e 100644
--- a/fs/btrfs/ordered-data.c
+++ b/fs/btrfs/ordered-data.c
@@ -16,7 +16,6 @@
  * Boston, MA 021110-1307, USA.
  */
 
-#include <linux/gfp.h>
 #include <linux/slab.h>
 #include <linux/blkdev.h>
 #include <linux/writeback.h>
@@ -174,7 +173,6 @@ int btrfs_add_ordered_extent(struct inode *inode, u64 file_offset,
 	if (!entry)
 		return -ENOMEM;
 
-	mutex_lock(&tree->mutex);
 	entry->file_offset = file_offset;
 	entry->start = start;
 	entry->len = len;
@@ -190,16 +188,17 @@ int btrfs_add_ordered_extent(struct inode *inode, u64 file_offset,
 	INIT_LIST_HEAD(&entry->list);
 	INIT_LIST_HEAD(&entry->root_extent_list);
 
+	spin_lock(&tree->lock);
 	node = tree_insert(&tree->tree, file_offset,
 			   &entry->rb_node);
 	BUG_ON(node);
+	spin_unlock(&tree->lock);
 
 	spin_lock(&BTRFS_I(inode)->root->fs_info->ordered_extent_lock);
 	list_add_tail(&entry->root_extent_list,
 		      &BTRFS_I(inode)->root->fs_info->ordered_extents);
 	spin_unlock(&BTRFS_I(inode)->root->fs_info->ordered_extent_lock);
 
-	mutex_unlock(&tree->mutex);
 	BUG_ON(node);
 	return 0;
 }
@@ -216,9 +215,9 @@ int btrfs_add_ordered_sum(struct inode *inode,
 	struct btrfs_ordered_inode_tree *tree;
 
 	tree = &BTRFS_I(inode)->ordered_tree;
-	mutex_lock(&tree->mutex);
+	spin_lock(&tree->lock);
 	list_add_tail(&sum->list, &entry->list);
-	mutex_unlock(&tree->mutex);
+	spin_unlock(&tree->lock);
 	return 0;
 }
 
@@ -232,15 +231,16 @@ int btrfs_add_ordered_sum(struct inode *inode,
  * to make sure this function only returns 1 once for a given ordered extent.
  */
 int btrfs_dec_test_ordered_pending(struct inode *inode,
+				   struct btrfs_ordered_extent **cached,
 				   u64 file_offset, u64 io_size)
 {
 	struct btrfs_ordered_inode_tree *tree;
 	struct rb_node *node;
-	struct btrfs_ordered_extent *entry;
+	struct btrfs_ordered_extent *entry = NULL;
 	int ret;
 
 	tree = &BTRFS_I(inode)->ordered_tree;
-	mutex_lock(&tree->mutex);
+	spin_lock(&tree->lock);
 	node = tree_search(tree, file_offset);
 	if (!node) {
 		ret = 1;
@@ -264,7 +264,11 @@ int btrfs_dec_test_ordered_pending(struct inode *inode,
 	else
 		ret = 1;
 out:
-	mutex_unlock(&tree->mutex);
+	if (!ret && cached && entry) {
+		*cached = entry;
+		atomic_inc(&entry->refs);
+	}
+	spin_unlock(&tree->lock);
 	return ret == 0;
 }
 
@@ -291,13 +295,14 @@ int btrfs_put_ordered_extent(struct btrfs_ordered_extent *entry)
 
 /*
  * remove an ordered extent from the tree.  No references are dropped
- * and you must wake_up entry->wait.  You must hold the tree mutex
+ * and you must wake_up entry->wait.  You must hold the tree lock
  * while you call this function.
  */
 static int __btrfs_remove_ordered_extent(struct inode *inode,
 				struct btrfs_ordered_extent *entry)
 {
 	struct btrfs_ordered_inode_tree *tree;
+	struct btrfs_root *root = BTRFS_I(inode)->root;
 	struct rb_node *node;
 
 	tree = &BTRFS_I(inode)->ordered_tree;
@@ -307,12 +312,13 @@ static int __btrfs_remove_ordered_extent(struct inode *inode,
 	set_bit(BTRFS_ORDERED_COMPLETE, &entry->flags);
 
 	spin_lock(&BTRFS_I(inode)->accounting_lock);
+	WARN_ON(!BTRFS_I(inode)->outstanding_extents);
 	BTRFS_I(inode)->outstanding_extents--;
 	spin_unlock(&BTRFS_I(inode)->accounting_lock);
 	btrfs_unreserve_metadata_for_delalloc(BTRFS_I(inode)->root,
 					      inode, 1);
 
-	spin_lock(&BTRFS_I(inode)->root->fs_info->ordered_extent_lock);
+	spin_lock(&root->fs_info->ordered_extent_lock);
 	list_del_init(&entry->root_extent_list);
 
 	/*
@@ -324,7 +330,7 @@ static int __btrfs_remove_ordered_extent(struct inode *inode,
 	    !mapping_tagged(inode->i_mapping, PAGECACHE_TAG_DIRTY)) {
 		list_del_init(&BTRFS_I(inode)->ordered_operations);
 	}
-	spin_unlock(&BTRFS_I(inode)->root->fs_info->ordered_extent_lock);
+	spin_unlock(&root->fs_info->ordered_extent_lock);
 
 	return 0;
 }
@@ -340,9 +346,9 @@ int btrfs_remove_ordered_extent(struct inode *inode,
 	int ret;
 
 	tree = &BTRFS_I(inode)->ordered_tree;
-	mutex_lock(&tree->mutex);
+	spin_lock(&tree->lock);
 	ret = __btrfs_remove_ordered_extent(inode, entry);
-	mutex_unlock(&tree->mutex);
+	spin_unlock(&tree->lock);
 	wake_up(&entry->wait);
 
 	return ret;
@@ -567,7 +573,7 @@ struct btrfs_ordered_extent *btrfs_lookup_ordered_extent(struct inode *inode,
 	struct btrfs_ordered_extent *entry = NULL;
 
 	tree = &BTRFS_I(inode)->ordered_tree;
-	mutex_lock(&tree->mutex);
+	spin_lock(&tree->lock);
 	node = tree_search(tree, file_offset);
 	if (!node)
 		goto out;
@@ -578,7 +584,7 @@ struct btrfs_ordered_extent *btrfs_lookup_ordered_extent(struct inode *inode,
 	if (entry)
 		atomic_inc(&entry->refs);
 out:
-	mutex_unlock(&tree->mutex);
+	spin_unlock(&tree->lock);
 	return entry;
 }
 
@@ -594,7 +600,7 @@ btrfs_lookup_first_ordered_extent(struct inode *inode, u64 file_offset)
 	struct btrfs_ordered_extent *entry = NULL;
 
 	tree = &BTRFS_I(inode)->ordered_tree;
-	mutex_lock(&tree->mutex);
+	spin_lock(&tree->lock);
 	node = tree_search(tree, file_offset);
 	if (!node)
 		goto out;
@@ -602,7 +608,7 @@ btrfs_lookup_first_ordered_extent(struct inode *inode, u64 file_offset)
 	entry = rb_entry(node, struct btrfs_ordered_extent, rb_node);
 	atomic_inc(&entry->refs);
 out:
-	mutex_unlock(&tree->mutex);
+	spin_unlock(&tree->lock);
 	return entry;
 }
 
@@ -629,7 +635,7 @@ int btrfs_ordered_update_i_size(struct inode *inode, u64 offset,
 	else
 		offset = ALIGN(offset, BTRFS_I(inode)->root->sectorsize);
 
-	mutex_lock(&tree->mutex);
+	spin_lock(&tree->lock);
 	disk_i_size = BTRFS_I(inode)->disk_i_size;
 
 	/* truncate file */
@@ -735,7 +741,7 @@ out:
 	 */
 	if (ordered)
 		__btrfs_remove_ordered_extent(inode, ordered);
-	mutex_unlock(&tree->mutex);
+	spin_unlock(&tree->lock);
 	if (ordered)
 		wake_up(&ordered->wait);
 	return ret;
@@ -762,7 +768,7 @@ int btrfs_find_ordered_sum(struct inode *inode, u64 offset, u64 disk_bytenr,
 	if (!ordered)
 		return 1;
 
-	mutex_lock(&tree->mutex);
+	spin_lock(&tree->lock);
 	list_for_each_entry_reverse(ordered_sum, &ordered->list, list) {
 		if (disk_bytenr >= ordered_sum->bytenr) {
 			num_sectors = ordered_sum->len / sectorsize;
@@ -777,7 +783,7 @@ int btrfs_find_ordered_sum(struct inode *inode, u64 offset, u64 disk_bytenr,
 		}
 	}
 out:
-	mutex_unlock(&tree->mutex);
+	spin_unlock(&tree->lock);
 	btrfs_put_ordered_extent(ordered);
 	return ret;
 }
diff --git a/fs/btrfs/ordered-data.h b/fs/btrfs/ordered-data.h
index 9116c6d..c82f76a 100644
--- a/fs/btrfs/ordered-data.h
+++ b/fs/btrfs/ordered-data.h
@@ -21,7 +21,7 @@
 
 /* one of these per inode */
 struct btrfs_ordered_inode_tree {
-	struct mutex mutex;
+	spinlock_t lock;
 	struct rb_root tree;
 	struct rb_node *last;
 };
@@ -128,7 +128,7 @@ static inline int btrfs_ordered_sum_size(struct btrfs_root *root,
 static inline void
 btrfs_ordered_inode_tree_init(struct btrfs_ordered_inode_tree *t)
 {
-	mutex_init(&t->mutex);
+	spin_lock_init(&t->lock);
 	t->tree = RB_ROOT;
 	t->last = NULL;
 }
@@ -137,7 +137,8 @@ int btrfs_put_ordered_extent(struct btrfs_ordered_extent *entry);
 int btrfs_remove_ordered_extent(struct inode *inode,
 				struct btrfs_ordered_extent *entry);
 int btrfs_dec_test_ordered_pending(struct inode *inode,
-				       u64 file_offset, u64 io_size);
+				   struct btrfs_ordered_extent **cached,
+				   u64 file_offset, u64 io_size);
 int btrfs_add_ordered_extent(struct inode *inode, u64 file_offset,
 			     u64 start, u64 len, u64 disk_len, int tyep);
 int btrfs_add_ordered_sum(struct inode *inode,
diff --git a/fs/btrfs/ref-cache.c b/fs/btrfs/ref-cache.c
index d0cc62b..a97314c 100644
--- a/fs/btrfs/ref-cache.c
+++ b/fs/btrfs/ref-cache.c
@@ -17,6 +17,7 @@
  */
 
 #include <linux/sched.h>
+#include <linux/slab.h>
 #include <linux/sort.h>
 #include "ctree.h"
 #include "ref-cache.h"
diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c
index 0109e56..e558dd9 100644
--- a/fs/btrfs/relocation.c
+++ b/fs/btrfs/relocation.c
@@ -21,6 +21,7 @@
 #include <linux/writeback.h>
 #include <linux/blkdev.h>
 #include <linux/rbtree.h>
+#include <linux/slab.h>
 #include "ctree.h"
 #include "disk-io.h"
 #include "transaction.h"
@@ -2659,7 +2660,7 @@ static int relocate_file_extent_cluster(struct inode *inode,
 					EXTENT_BOUNDARY, GFP_NOFS);
 			nr++;
 		}
-		btrfs_set_extent_delalloc(inode, page_start, page_end);
+		btrfs_set_extent_delalloc(inode, page_start, page_end, NULL);
 
 		set_page_dirty(page);
 		dirty_page++;
@@ -3487,7 +3488,7 @@ static struct inode *create_reloc_inode(struct btrfs_fs_info *fs_info,
 	key.objectid = objectid;
 	key.type = BTRFS_INODE_ITEM_KEY;
 	key.offset = 0;
-	inode = btrfs_iget(root->fs_info->sb, &key, root);
+	inode = btrfs_iget(root->fs_info->sb, &key, root, NULL);
 	BUG_ON(IS_ERR(inode) || is_bad_inode(inode));
 	BTRFS_I(inode)->index_cnt = group->key.objectid;
 
diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c
index f8b4521..1866dff 100644
--- a/fs/btrfs/super.c
+++ b/fs/btrfs/super.c
@@ -38,6 +38,7 @@
 #include <linux/namei.h>
 #include <linux/miscdevice.h>
 #include <linux/magic.h>
+#include <linux/slab.h>
 #include "compat.h"
 #include "ctree.h"
 #include "disk-io.h"
@@ -63,22 +64,21 @@ static void btrfs_put_super(struct super_block *sb)
 }
 
 enum {
-	Opt_degraded, Opt_subvol, Opt_device, Opt_nodatasum, Opt_nodatacow,
-	Opt_max_extent, Opt_max_inline, Opt_alloc_start, Opt_nobarrier,
-	Opt_ssd, Opt_nossd, Opt_ssd_spread, Opt_thread_pool, Opt_noacl,
-	Opt_compress, Opt_compress_force, Opt_notreelog, Opt_ratio,
-	Opt_flushoncommit,
+	Opt_degraded, Opt_subvol, Opt_subvolid, Opt_device, Opt_nodatasum,
+	Opt_nodatacow, Opt_max_inline, Opt_alloc_start, Opt_nobarrier, Opt_ssd,
+	Opt_nossd, Opt_ssd_spread, Opt_thread_pool, Opt_noacl, Opt_compress,
+	Opt_compress_force, Opt_notreelog, Opt_ratio, Opt_flushoncommit,
 	Opt_discard, Opt_err,
 };
 
 static match_table_t tokens = {
 	{Opt_degraded, "degraded"},
 	{Opt_subvol, "subvol=%s"},
+	{Opt_subvolid, "subvolid=%d"},
 	{Opt_device, "device=%s"},
 	{Opt_nodatasum, "nodatasum"},
 	{Opt_nodatacow, "nodatacow"},
 	{Opt_nobarrier, "nobarrier"},
-	{Opt_max_extent, "max_extent=%s"},
 	{Opt_max_inline, "max_inline=%s"},
 	{Opt_alloc_start, "alloc_start=%s"},
 	{Opt_thread_pool, "thread_pool=%d"},
@@ -95,31 +95,6 @@ static match_table_t tokens = {
 	{Opt_err, NULL},
 };
 
-u64 btrfs_parse_size(char *str)
-{
-	u64 res;
-	int mult = 1;
-	char *end;
-	char last;
-
-	res = simple_strtoul(str, &end, 10);
-
-	last = end[0];
-	if (isalpha(last)) {
-		last = tolower(last);
-		switch (last) {
-		case 'g':
-			mult *= 1024;
-		case 'm':
-			mult *= 1024;
-		case 'k':
-			mult *= 1024;
-		}
-		res = res * mult;
-	}
-	return res;
-}
-
 /*
  * Regular mount options parser.  Everything that is needed only when
  * reading in a new superblock is parsed here.
@@ -157,6 +132,7 @@ int btrfs_parse_options(struct btrfs_root *root, char *options)
 			btrfs_set_opt(info->mount_opt, DEGRADED);
 			break;
 		case Opt_subvol:
+		case Opt_subvolid:
 		case Opt_device:
 			/*
 			 * These are parsed by btrfs_parse_early_options
@@ -211,22 +187,10 @@ int btrfs_parse_options(struct btrfs_root *root, char *options)
 				       info->thread_pool_size);
 			}
 			break;
-		case Opt_max_extent:
-			num = match_strdup(&args[0]);
-			if (num) {
-				info->max_extent = btrfs_parse_size(num);
-				kfree(num);
-
-				info->max_extent = max_t(u64,
-					info->max_extent, root->sectorsize);
-				printk(KERN_INFO "btrfs: max_extent at %llu\n",
-				       (unsigned long long)info->max_extent);
-			}
-			break;
 		case Opt_max_inline:
 			num = match_strdup(&args[0]);
 			if (num) {
-				info->max_inline = btrfs_parse_size(num);
+				info->max_inline = memparse(num, NULL);
 				kfree(num);
 
 				if (info->max_inline) {
@@ -241,7 +205,7 @@ int btrfs_parse_options(struct btrfs_root *root, char *options)
 		case Opt_alloc_start:
 			num = match_strdup(&args[0]);
 			if (num) {
-				info->alloc_start = btrfs_parse_size(num);
+				info->alloc_start = memparse(num, NULL);
 				kfree(num);
 				printk(KERN_INFO
 					"btrfs: allocations start at %llu\n",
@@ -292,12 +256,13 @@ out:
  * only when we need to allocate a new super block.
  */
 static int btrfs_parse_early_options(const char *options, fmode_t flags,
-		void *holder, char **subvol_name,
+		void *holder, char **subvol_name, u64 *subvol_objectid,
 		struct btrfs_fs_devices **fs_devices)
 {
 	substring_t args[MAX_OPT_ARGS];
 	char *opts, *p;
 	int error = 0;
+	int intarg;
 
 	if (!options)
 		goto out;
@@ -320,6 +285,18 @@ static int btrfs_parse_early_options(const char *options, fmode_t flags,
 		case Opt_subvol:
 			*subvol_name = match_strdup(&args[0]);
 			break;
+		case Opt_subvolid:
+			intarg = 0;
+			error = match_int(&args[0], &intarg);
+			if (!error) {
+				/* we want the original fs_tree */
+				if (!intarg)
+					*subvol_objectid =
+						BTRFS_FS_TREE_OBJECTID;
+				else
+					*subvol_objectid = intarg;
+			}
+			break;
 		case Opt_device:
 			error = btrfs_scan_one_device(match_strdup(&args[0]),
 					flags, holder, fs_devices);
@@ -347,6 +324,110 @@ static int btrfs_parse_early_options(const char *options, fmode_t flags,
 	return error;
 }
 
+static struct dentry *get_default_root(struct super_block *sb,
+				       u64 subvol_objectid)
+{
+	struct btrfs_root *root = sb->s_fs_info;
+	struct btrfs_root *new_root;
+	struct btrfs_dir_item *di;
+	struct btrfs_path *path;
+	struct btrfs_key location;
+	struct inode *inode;
+	struct dentry *dentry;
+	u64 dir_id;
+	int new = 0;
+
+	/*
+	 * We have a specific subvol we want to mount, just setup location and
+	 * go look up the root.
+	 */
+	if (subvol_objectid) {
+		location.objectid = subvol_objectid;
+		location.type = BTRFS_ROOT_ITEM_KEY;
+		location.offset = (u64)-1;
+		goto find_root;
+	}
+
+	path = btrfs_alloc_path();
+	if (!path)
+		return ERR_PTR(-ENOMEM);
+	path->leave_spinning = 1;
+
+	/*
+	 * Find the "default" dir item which points to the root item that we
+	 * will mount by default if we haven't been given a specific subvolume
+	 * to mount.
+	 */
+	dir_id = btrfs_super_root_dir(&root->fs_info->super_copy);
+	di = btrfs_lookup_dir_item(NULL, root, path, dir_id, "default", 7, 0);
+	if (!di) {
+		/*
+		 * Ok the default dir item isn't there.  This is weird since
+		 * it's always been there, but don't freak out, just try and
+		 * mount to root most subvolume.
+		 */
+		btrfs_free_path(path);
+		dir_id = BTRFS_FIRST_FREE_OBJECTID;
+		new_root = root->fs_info->fs_root;
+		goto setup_root;
+	}
+
+	btrfs_dir_item_key_to_cpu(path->nodes[0], di, &location);
+	btrfs_free_path(path);
+
+find_root:
+	new_root = btrfs_read_fs_root_no_name(root->fs_info, &location);
+	if (IS_ERR(new_root))
+		return ERR_PTR(PTR_ERR(new_root));
+
+	if (btrfs_root_refs(&new_root->root_item) == 0)
+		return ERR_PTR(-ENOENT);
+
+	dir_id = btrfs_root_dirid(&new_root->root_item);
+setup_root:
+	location.objectid = dir_id;
+	location.type = BTRFS_INODE_ITEM_KEY;
+	location.offset = 0;
+
+	inode = btrfs_iget(sb, &location, new_root, &new);
+	if (!inode)
+		return ERR_PTR(-ENOMEM);
+
+	/*
+	 * If we're just mounting the root most subvol put the inode and return
+	 * a reference to the dentry.  We will have already gotten a reference
+	 * to the inode in btrfs_fill_super so we're good to go.
+	 */
+	if (!new && sb->s_root->d_inode == inode) {
+		iput(inode);
+		return dget(sb->s_root);
+	}
+
+	if (new) {
+		const struct qstr name = { .name = "/", .len = 1 };
+
+		/*
+		 * New inode, we need to make the dentry a sibling of s_root so
+		 * everything gets cleaned up properly on unmount.
+		 */
+		dentry = d_alloc(sb->s_root, &name);
+		if (!dentry) {
+			iput(inode);
+			return ERR_PTR(-ENOMEM);
+		}
+		d_splice_alias(inode, dentry);
+	} else {
+		/*
+		 * We found the inode in cache, just find a dentry for it and
+		 * put the reference to the inode we just got.
+		 */
+		dentry = d_find_alias(inode);
+		iput(inode);
+	}
+
+	return dentry;
+}
+
 static int btrfs_fill_super(struct super_block *sb,
 			    struct btrfs_fs_devices *fs_devices,
 			    void *data, int silent)
@@ -380,7 +461,7 @@ static int btrfs_fill_super(struct super_block *sb,
 	key.objectid = BTRFS_FIRST_FREE_OBJECTID;
 	key.type = BTRFS_INODE_ITEM_KEY;
 	key.offset = 0;
-	inode = btrfs_iget(sb, &key, tree_root->fs_info->fs_root);
+	inode = btrfs_iget(sb, &key, tree_root->fs_info->fs_root, NULL);
 	if (IS_ERR(inode)) {
 		err = PTR_ERR(inode);
 		goto fail_close;
@@ -392,12 +473,6 @@ static int btrfs_fill_super(struct super_block *sb,
 		err = -ENOMEM;
 		goto fail_close;
 	}
-#if 0
-	/* this does the super kobj at the same time */
-	err = btrfs_sysfs_add_super(tree_root->fs_info);
-	if (err)
-		goto fail_close;
-#endif
 
 	sb->s_root = root_dentry;
 
@@ -441,9 +516,6 @@ static int btrfs_show_options(struct seq_file *seq, struct vfsmount *vfs)
 		seq_puts(seq, ",nodatacow");
 	if (btrfs_test_opt(root, NOBARRIER))
 		seq_puts(seq, ",nobarrier");
-	if (info->max_extent != (u64)-1)
-		seq_printf(seq, ",max_extent=%llu",
-			   (unsigned long long)info->max_extent);
 	if (info->max_inline != 8192 * 1024)
 		seq_printf(seq, ",max_inline=%llu",
 			   (unsigned long long)info->max_inline);
@@ -489,19 +561,22 @@ static int btrfs_test_super(struct super_block *s, void *data)
 static int btrfs_get_sb(struct file_system_type *fs_type, int flags,
 		const char *dev_name, void *data, struct vfsmount *mnt)
 {
-	char *subvol_name = NULL;
 	struct block_device *bdev = NULL;
 	struct super_block *s;
 	struct dentry *root;
 	struct btrfs_fs_devices *fs_devices = NULL;
 	fmode_t mode = FMODE_READ;
+	char *subvol_name = NULL;
+	u64 subvol_objectid = 0;
 	int error = 0;
+	int found = 0;
 
 	if (!(flags & MS_RDONLY))
 		mode |= FMODE_WRITE;
 
 	error = btrfs_parse_early_options(data, mode, fs_type,
-					  &subvol_name, &fs_devices);
+					  &subvol_name, &subvol_objectid,
+					  &fs_devices);
 	if (error)
 		return error;
 
@@ -530,6 +605,7 @@ static int btrfs_get_sb(struct file_system_type *fs_type, int flags,
 			goto error_close_devices;
 		}
 
+		found = 1;
 		btrfs_close_devices(fs_devices);
 	} else {
 		char b[BDEVNAME_SIZE];
@@ -547,25 +623,35 @@ static int btrfs_get_sb(struct file_system_type *fs_type, int flags,
 		s->s_flags |= MS_ACTIVE;
 	}
 
-	if (!strcmp(subvol_name, "."))
-		root = dget(s->s_root);
-	else {
-		mutex_lock(&s->s_root->d_inode->i_mutex);
-		root = lookup_one_len(subvol_name, s->s_root,
+	root = get_default_root(s, subvol_objectid);
+	if (IS_ERR(root)) {
+		error = PTR_ERR(root);
+		deactivate_locked_super(s);
+		goto error;
+	}
+	/* if they gave us a subvolume name bind mount into that */
+	if (strcmp(subvol_name, ".")) {
+		struct dentry *new_root;
+		mutex_lock(&root->d_inode->i_mutex);
+		new_root = lookup_one_len(subvol_name, root,
 				      strlen(subvol_name));
-		mutex_unlock(&s->s_root->d_inode->i_mutex);
+		mutex_unlock(&root->d_inode->i_mutex);
 
-		if (IS_ERR(root)) {
+		if (IS_ERR(new_root)) {
 			deactivate_locked_super(s);
-			error = PTR_ERR(root);
-			goto error_free_subvol_name;
+			error = PTR_ERR(new_root);
+			dput(root);
+			goto error_close_devices;
 		}
-		if (!root->d_inode) {
+		if (!new_root->d_inode) {
 			dput(root);
+			dput(new_root);
 			deactivate_locked_super(s);
 			error = -ENXIO;
-			goto error_free_subvol_name;
+			goto error_close_devices;
 		}
+		dput(root);
+		root = new_root;
 	}
 
 	mnt->mnt_sb = s;
@@ -580,6 +666,7 @@ error_close_devices:
 	btrfs_close_devices(fs_devices);
 error_free_subvol_name:
 	kfree(subvol_name);
+error:
 	return error;
 }
 
@@ -624,14 +711,37 @@ static int btrfs_statfs(struct dentry *dentry, struct kstatfs *buf)
 {
 	struct btrfs_root *root = btrfs_sb(dentry->d_sb);
 	struct btrfs_super_block *disk_super = &root->fs_info->super_copy;
+	struct list_head *head = &root->fs_info->space_info;
+	struct btrfs_space_info *found;
+	u64 total_used = 0;
+	u64 data_used = 0;
 	int bits = dentry->d_sb->s_blocksize_bits;
 	__be32 *fsid = (__be32 *)root->fs_info->fsid;
 
+	rcu_read_lock();
+	list_for_each_entry_rcu(found, head, list) {
+		if (found->flags & (BTRFS_BLOCK_GROUP_DUP|
+				    BTRFS_BLOCK_GROUP_RAID10|
+				    BTRFS_BLOCK_GROUP_RAID1)) {
+			total_used += found->bytes_used;
+			if (found->flags & BTRFS_BLOCK_GROUP_DATA)
+				data_used += found->bytes_used;
+			else
+				data_used += found->total_bytes;
+		}
+
+		total_used += found->bytes_used;
+		if (found->flags & BTRFS_BLOCK_GROUP_DATA)
+			data_used += found->bytes_used;
+		else
+			data_used += found->total_bytes;
+	}
+	rcu_read_unlock();
+
 	buf->f_namelen = BTRFS_NAME_LEN;
 	buf->f_blocks = btrfs_super_total_bytes(disk_super) >> bits;
-	buf->f_bfree = buf->f_blocks -
-		(btrfs_super_bytes_used(disk_super) >> bits);
-	buf->f_bavail = buf->f_bfree;
+	buf->f_bfree = buf->f_blocks - (total_used >> bits);
+	buf->f_bavail = buf->f_blocks - (data_used >> bits);
 	buf->f_bsize = dentry->d_sb->s_blocksize;
 	buf->f_type = BTRFS_SUPER_MAGIC;
 
diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c
index 2a36e23..2cb1160 100644
--- a/fs/btrfs/transaction.c
+++ b/fs/btrfs/transaction.c
@@ -17,6 +17,7 @@
  */
 
 #include <linux/fs.h>
+#include <linux/slab.h>
 #include <linux/sched.h>
 #include <linux/writeback.h>
 #include <linux/pagemap.h>
@@ -147,18 +148,13 @@ static void wait_current_trans(struct btrfs_root *root)
 		while (1) {
 			prepare_to_wait(&root->fs_info->transaction_wait, &wait,
 					TASK_UNINTERRUPTIBLE);
-			if (cur_trans->blocked) {
-				mutex_unlock(&root->fs_info->trans_mutex);
-				schedule();
-				mutex_lock(&root->fs_info->trans_mutex);
-				finish_wait(&root->fs_info->transaction_wait,
-					    &wait);
-			} else {
-				finish_wait(&root->fs_info->transaction_wait,
-					    &wait);
+			if (!cur_trans->blocked)
 				break;
-			}
+			mutex_unlock(&root->fs_info->trans_mutex);
+			schedule();
+			mutex_lock(&root->fs_info->trans_mutex);
 		}
+		finish_wait(&root->fs_info->transaction_wait, &wait);
 		put_transaction(cur_trans);
 	}
 }
@@ -760,10 +756,17 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
 	struct btrfs_root_item *new_root_item;
 	struct btrfs_root *tree_root = fs_info->tree_root;
 	struct btrfs_root *root = pending->root;
+	struct btrfs_root *parent_root;
+	struct inode *parent_inode;
 	struct extent_buffer *tmp;
 	struct extent_buffer *old;
 	int ret;
 	u64 objectid;
+	int namelen;
+	u64 index = 0;
+
+	parent_inode = pending->dentry->d_parent->d_inode;
+	parent_root = BTRFS_I(parent_inode)->root;
 
 	new_root_item = kmalloc(sizeof(*new_root_item), GFP_NOFS);
 	if (!new_root_item) {
@@ -774,79 +777,59 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
 	if (ret)
 		goto fail;
 
-	record_root_in_trans(trans, root);
-	btrfs_set_root_last_snapshot(&root->root_item, trans->transid);
-	memcpy(new_root_item, &root->root_item, sizeof(*new_root_item));
-
 	key.objectid = objectid;
 	/* record when the snapshot was created in key.offset */
 	key.offset = trans->transid;
 	btrfs_set_key_type(&key, BTRFS_ROOT_ITEM_KEY);
 
-	old = btrfs_lock_root_node(root);
-	btrfs_cow_block(trans, root, old, NULL, 0, &old);
-	btrfs_set_lock_blocking(old);
-
-	btrfs_copy_root(trans, root, old, &tmp, objectid);
-	btrfs_tree_unlock(old);
-	free_extent_buffer(old);
-
-	btrfs_set_root_node(new_root_item, tmp);
-	ret = btrfs_insert_root(trans, root->fs_info->tree_root, &key,
-				new_root_item);
-	btrfs_tree_unlock(tmp);
-	free_extent_buffer(tmp);
-	if (ret)
-		goto fail;
-
-	key.offset = (u64)-1;
 	memcpy(&pending->root_key, &key, sizeof(key));
-fail:
-	kfree(new_root_item);
-	return ret;
-}
-
-static noinline int finish_pending_snapshot(struct btrfs_fs_info *fs_info,
-				   struct btrfs_pending_snapshot *pending)
-{
-	int ret;
-	int namelen;
-	u64 index = 0;
-	struct btrfs_trans_handle *trans;
-	struct inode *parent_inode;
-	struct btrfs_root *parent_root;
-
-	parent_inode = pending->dentry->d_parent->d_inode;
-	parent_root = BTRFS_I(parent_inode)->root;
-	trans = btrfs_join_transaction(parent_root, 1);
+	pending->root_key.offset = (u64)-1;
 
+	record_root_in_trans(trans, parent_root);
 	/*
 	 * insert the directory item
 	 */
 	namelen = strlen(pending->name);
 	ret = btrfs_set_inode_index(parent_inode, &index);
+	BUG_ON(ret);
 	ret = btrfs_insert_dir_item(trans, parent_root,
 			    pending->name, namelen,
 			    parent_inode->i_ino,
 			    &pending->root_key, BTRFS_FT_DIR, index);
-
-	if (ret)
-		goto fail;
+	BUG_ON(ret);
 
 	btrfs_i_size_write(parent_inode, parent_inode->i_size + namelen * 2);
 	ret = btrfs_update_inode(trans, parent_root, parent_inode);
 	BUG_ON(ret);
 
+	record_root_in_trans(trans, root);
+	btrfs_set_root_last_snapshot(&root->root_item, trans->transid);
+	memcpy(new_root_item, &root->root_item, sizeof(*new_root_item));
+
+	old = btrfs_lock_root_node(root);
+	btrfs_cow_block(trans, root, old, NULL, 0, &old);
+	btrfs_set_lock_blocking(old);
+
+	btrfs_copy_root(trans, root, old, &tmp, objectid);
+	btrfs_tree_unlock(old);
+	free_extent_buffer(old);
+
+	btrfs_set_root_node(new_root_item, tmp);
+	ret = btrfs_insert_root(trans, root->fs_info->tree_root, &key,
+				new_root_item);
+	BUG_ON(ret);
+	btrfs_tree_unlock(tmp);
+	free_extent_buffer(tmp);
+
 	ret = btrfs_add_root_ref(trans, parent_root->fs_info->tree_root,
 				 pending->root_key.objectid,
 				 parent_root->root_key.objectid,
 				 parent_inode->i_ino, index, pending->name,
 				 namelen);
-
 	BUG_ON(ret);
 
 fail:
-	btrfs_end_transaction(trans, fs_info->fs_root);
+	kfree(new_root_item);
 	return ret;
 }
 
@@ -867,25 +850,6 @@ static noinline int create_pending_snapshots(struct btrfs_trans_handle *trans,
 	return 0;
 }
 
-static noinline int finish_pending_snapshots(struct btrfs_trans_handle *trans,
-					     struct btrfs_fs_info *fs_info)
-{
-	struct btrfs_pending_snapshot *pending;
-	struct list_head *head = &trans->transaction->pending_snapshots;
-	int ret;
-
-	while (!list_empty(head)) {
-		pending = list_entry(head->next,
-				     struct btrfs_pending_snapshot, list);
-		ret = finish_pending_snapshot(fs_info, pending);
-		BUG_ON(ret);
-		list_del(&pending->list);
-		kfree(pending->name);
-		kfree(pending);
-	}
-	return 0;
-}
-
 static void update_super_roots(struct btrfs_root *root)
 {
 	struct btrfs_root_item *root_item;
@@ -997,13 +961,10 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans,
 
 		mutex_unlock(&root->fs_info->trans_mutex);
 
-		if (flush_on_commit) {
+		if (flush_on_commit || snap_pending) {
 			btrfs_start_delalloc_inodes(root, 1);
 			ret = btrfs_wait_ordered_extents(root, 0, 1);
 			BUG_ON(ret);
-		} else if (snap_pending) {
-			ret = btrfs_wait_ordered_extents(root, 0, 1);
-			BUG_ON(ret);
 		}
 
 		/*
@@ -1100,9 +1061,6 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans,
 
 	btrfs_finish_extent_commit(trans, root);
 
-	/* do the directory inserts of any pending snapshot creations */
-	finish_pending_snapshots(trans, root->fs_info);
-
 	mutex_lock(&root->fs_info->trans_mutex);
 
 	cur_trans->commit_done = 1;
diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c
index 4a9434b..af57dd2 100644
--- a/fs/btrfs/tree-log.c
+++ b/fs/btrfs/tree-log.c
@@ -17,6 +17,7 @@
  */
 
 #include <linux/sched.h>
+#include <linux/slab.h>
 #include "ctree.h"
 #include "transaction.h"
 #include "disk-io.h"
@@ -445,7 +446,7 @@ static noinline struct inode *read_one_inode(struct btrfs_root *root,
 	key.objectid = objectid;
 	key.type = BTRFS_INODE_ITEM_KEY;
 	key.offset = 0;
-	inode = btrfs_iget(root->fs_info->sb, &key, root);
+	inode = btrfs_iget(root->fs_info->sb, &key, root, NULL);
 	if (IS_ERR(inode)) {
 		inode = NULL;
 	} else if (is_bad_inode(inode)) {
diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
index 41ecbb2..8db7b14 100644
--- a/fs/btrfs/volumes.c
+++ b/fs/btrfs/volumes.c
@@ -17,6 +17,7 @@
  */
 #include <linux/sched.h>
 #include <linux/bio.h>
+#include <linux/slab.h>
 #include <linux/buffer_head.h>
 #include <linux/blkdev.h>
 #include <linux/random.h>
@@ -256,13 +257,13 @@ loop_lock:
 			wake_up(&fs_info->async_submit_wait);
 
 		BUG_ON(atomic_read(&cur->bi_cnt) == 0);
-		submit_bio(cur->bi_rw, cur);
-		num_run++;
-		batch_run++;
 
 		if (bio_rw_flagged(cur, BIO_RW_SYNCIO))
 			num_sync_run++;
 
+		submit_bio(cur->bi_rw, cur);
+		num_run++;
+		batch_run++;
 		if (need_resched()) {
 			if (num_sync_run) {
 				blk_run_backing_dev(bdi, NULL);
@@ -325,16 +326,6 @@ loop_lock:
 		num_sync_run = 0;
 		blk_run_backing_dev(bdi, NULL);
 	}
-
-	cond_resched();
-	if (again)
-		goto loop;
-
-	spin_lock(&device->io_lock);
-	if (device->pending_bios.head || device->pending_sync_bios.head)
-		goto loop_lock;
-	spin_unlock(&device->io_lock);
-
 	/*
 	 * IO has already been through a long path to get here.  Checksumming,
 	 * async helper threads, perhaps compression.  We've done a pretty
@@ -346,6 +337,16 @@ loop_lock:
 	 * cared about found its way down here.
 	 */
 	blk_run_backing_dev(bdi, NULL);
+
+	cond_resched();
+	if (again)
+		goto loop;
+
+	spin_lock(&device->io_lock);
+	if (device->pending_bios.head || device->pending_sync_bios.head)
+		goto loop_lock;
+	spin_unlock(&device->io_lock);
+
 done:
 	return 0;
 }
@@ -365,6 +366,7 @@ static noinline int device_list_add(const char *path,
 	struct btrfs_device *device;
 	struct btrfs_fs_devices *fs_devices;
 	u64 found_transid = btrfs_super_generation(disk_super);
+	char *name;
 
 	fs_devices = find_fsid(disk_super->fsid);
 	if (!fs_devices) {
@@ -411,6 +413,12 @@ static noinline int device_list_add(const char *path,
 
 		device->fs_devices = fs_devices;
 		fs_devices->num_devices++;
+	} else if (strcmp(device->name, path)) {
+		name = kstrdup(path, GFP_NOFS);
+		if (!name)
+			return -ENOMEM;
+		kfree(device->name);
+		device->name = name;
 	}
 
 	if (found_transid > fs_devices->latest_trans) {
@@ -592,7 +600,7 @@ static int __btrfs_open_devices(struct btrfs_fs_devices *fs_devices,
 			goto error_close;
 
 		disk_super = (struct btrfs_super_block *)bh->b_data;
-		devid = le64_to_cpu(disk_super->dev_item.devid);
+		devid = btrfs_stack_device_id(&disk_super->dev_item);
 		if (devid != device->devid)
 			goto error_brelse;
 
@@ -694,7 +702,7 @@ int btrfs_scan_one_device(const char *path, fmode_t flags, void *holder,
 		goto error_close;
 	}
 	disk_super = (struct btrfs_super_block *)bh->b_data;
-	devid = le64_to_cpu(disk_super->dev_item.devid);
+	devid = btrfs_stack_device_id(&disk_super->dev_item);
 	transid = btrfs_super_generation(disk_super);
 	if (disk_super->label[0])
 		printk(KERN_INFO "device label %s ", disk_super->label);
@@ -1187,7 +1195,7 @@ int btrfs_rm_device(struct btrfs_root *root, char *device_path)
 			goto error_close;
 		}
 		disk_super = (struct btrfs_super_block *)bh->b_data;
-		devid = le64_to_cpu(disk_super->dev_item.devid);
+		devid = btrfs_stack_device_id(&disk_super->dev_item);
 		dev_uuid = disk_super->dev_item.uuid;
 		device = btrfs_find_device(root, devid, dev_uuid,
 					   disk_super->fsid);
@@ -2191,9 +2199,9 @@ static int __btrfs_alloc_chunk(struct btrfs_trans_handle *trans,
 		min_stripes = 2;
 	}
 	if (type & (BTRFS_BLOCK_GROUP_RAID1)) {
-		num_stripes = min_t(u64, 2, fs_devices->rw_devices);
-		if (num_stripes < 2)
+		if (fs_devices->rw_devices < 2)
 			return -ENOSPC;
+		num_stripes = 2;
 		min_stripes = 2;
 	}
 	if (type & (BTRFS_BLOCK_GROUP_RAID10)) {
@@ -2237,8 +2245,16 @@ again:
 		do_div(calc_size, stripe_len);
 		calc_size *= stripe_len;
 	}
+
 	/* we don't want tiny stripes */
-	calc_size = max_t(u64, min_stripe_size, calc_size);
+	if (!looped)
+		calc_size = max_t(u64, min_stripe_size, calc_size);
+
+	/*
+	 * we're about to do_div by the stripe_len so lets make sure
+	 * we end up with something bigger than a stripe
+	 */
+	calc_size = max_t(u64, calc_size, stripe_len * 4);
 
 	do_div(calc_size, stripe_len);
 	calc_size *= stripe_len;
@@ -3382,6 +3398,8 @@ int btrfs_read_chunk_tree(struct btrfs_root *root)
 	key.type = 0;
 again:
 	ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
+	if (ret < 0)
+		goto error;
 	while (1) {
 		leaf = path->nodes[0];
 		slot = path->slots[0];