1 files changed, 251 insertions, 187 deletions

diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
index fdce8799b98d..98b6a71decba 100644
--- a/fs/btrfs/disk-io.c
+++ b/fs/btrfs/disk-io.c
@@ -29,6 +29,8 @@
 #include <linux/crc32c.h>
 #include <linux/slab.h>
 #include <linux/migrate.h>
+#include <linux/ratelimit.h>
+#include <asm/unaligned.h>
 #include "compat.h"
 #include "ctree.h"
 #include "disk-io.h"
@@ -40,6 +42,7 @@
 #include "locking.h"
 #include "tree-log.h"
 #include "free-space-cache.h"
+#include "inode-map.h"
 
 static struct extent_io_ops btree_extent_io_ops;
 static void end_workqueue_fn(struct btrfs_work *work);
@@ -136,7 +139,7 @@ static const char *btrfs_eb_name[BTRFS_MAX_LEVEL + 1] = {
  * that covers the entire device
  */
 static struct extent_map *btree_get_extent(struct inode *inode,
-		struct page *page, size_t page_offset, u64 start, u64 len,
+		struct page *page, size_t pg_offset, u64 start, u64 len,
 		int create)
 {
 	struct extent_map_tree *em_tree = &BTRFS_I(inode)->extent_tree;
@@ -153,7 +156,7 @@ static struct extent_map *btree_get_extent(struct inode *inode,
 	}
 	read_unlock(&em_tree->lock);
 
-	em = alloc_extent_map(GFP_NOFS);
+	em = alloc_extent_map();
 	if (!em) {
 		em = ERR_PTR(-ENOMEM);
 		goto out;
@@ -198,7 +201,7 @@ u32 btrfs_csum_data(struct btrfs_root *root, char *data, u32 seed, size_t len)
 
 void btrfs_csum_final(u32 crc, char *result)
 {
-	*(__le32 *)result = ~cpu_to_le32(crc);
+	put_unaligned_le32(~crc, result);
 }
 
 /*
@@ -253,14 +256,12 @@ static int csum_tree_block(struct btrfs_root *root, struct extent_buffer *buf,
 			memcpy(&found, result, csum_size);
 
 			read_extent_buffer(buf, &val, 0, csum_size);
-			if (printk_ratelimit()) {
-				printk(KERN_INFO "btrfs: %s checksum verify "
+			printk_ratelimited(KERN_INFO "btrfs: %s checksum verify "
 				       "failed on %llu wanted %X found %X "
 				       "level %d\n",
 				       root->fs_info->sb->s_id,
 				       (unsigned long long)buf->start, val, found,
 				       btrfs_header_level(buf));
-			}
 			if (result != (char *)&inline_result)
 				kfree(result);
 			return 1;
@@ -295,13 +296,11 @@ static int verify_parent_transid(struct extent_io_tree *io_tree,
 		ret = 0;
 		goto out;
 	}
-	if (printk_ratelimit()) {
-		printk("parent transid verify failed on %llu wanted %llu "
+	printk_ratelimited("parent transid verify failed on %llu wanted %llu "
 		       "found %llu\n",
 		       (unsigned long long)eb->start,
 		       (unsigned long long)parent_transid,
 		       (unsigned long long)btrfs_header_generation(eb));
-	}
 	ret = 1;
 	clear_extent_buffer_uptodate(io_tree, eb, &cached_state);
 out:
@@ -323,6 +322,7 @@ static int btree_read_extent_buffer_pages(struct btrfs_root *root,
 	int num_copies = 0;
 	int mirror_num = 0;
 
+	clear_bit(EXTENT_BUFFER_CORRUPT, &eb->bflags);
 	io_tree = &BTRFS_I(root->fs_info->btree_inode)->io_tree;
 	while (1) {
 		ret = read_extent_buffer_pages(io_tree, eb, start, 1,
@@ -331,6 +331,14 @@ static int btree_read_extent_buffer_pages(struct btrfs_root *root,
 		    !verify_parent_transid(io_tree, eb, parent_transid))
 			return ret;
 
+		/*
+		 * This buffer's crc is fine, but its contents are corrupted, so
+		 * there is no reason to read the other copies, they won't be
+		 * any less wrong.
+		 */
+		if (test_bit(EXTENT_BUFFER_CORRUPT, &eb->bflags))
+			return ret;
+
 		num_copies = btrfs_num_copies(&root->fs_info->mapping_tree,
 					      eb->start, eb->len);
 		if (num_copies == 1)
@@ -359,14 +367,18 @@ static int csum_dirty_buffer(struct btrfs_root *root, struct page *page)
 
 	tree = &BTRFS_I(page->mapping->host)->io_tree;
 
-	if (page->private == EXTENT_PAGE_PRIVATE)
+	if (page->private == EXTENT_PAGE_PRIVATE) {
+		WARN_ON(1);
 		goto out;
-	if (!page->private)
+	}
+	if (!page->private) {
+		WARN_ON(1);
 		goto out;
+	}
 	len = page->private >> 2;
 	WARN_ON(len == 0);
 
-	eb = alloc_extent_buffer(tree, start, len, page, GFP_NOFS);
+	eb = alloc_extent_buffer(tree, start, len, page);
 	if (eb == NULL) {
 		WARN_ON(1);
 		goto out;
@@ -415,6 +427,73 @@ static int check_tree_block_fsid(struct btrfs_root *root,
 	return ret;
 }
 
+#define CORRUPT(reason, eb, root, slot)				\
+	printk(KERN_CRIT "btrfs: corrupt leaf, %s: block=%llu,"	\
+	       "root=%llu, slot=%d\n", reason,			\
+	       (unsigned long long)btrfs_header_bytenr(eb),	\
+	       (unsigned long long)root->objectid, slot)
+
+static noinline int check_leaf(struct btrfs_root *root,
+			       struct extent_buffer *leaf)
+{
+	struct btrfs_key key;
+	struct btrfs_key leaf_key;
+	u32 nritems = btrfs_header_nritems(leaf);
+	int slot;
+
+	if (nritems == 0)
+		return 0;
+
+	/* Check the 0 item */
+	if (btrfs_item_offset_nr(leaf, 0) + btrfs_item_size_nr(leaf, 0) !=
+	    BTRFS_LEAF_DATA_SIZE(root)) {
+		CORRUPT("invalid item offset size pair", leaf, root, 0);
+		return -EIO;
+	}
+
+	/*
+	 * Check to make sure each items keys are in the correct order and their
+	 * offsets make sense.  We only have to loop through nritems-1 because
+	 * we check the current slot against the next slot, which verifies the
+	 * next slot's offset+size makes sense and that the current's slot
+	 * offset is correct.
+	 */
+	for (slot = 0; slot < nritems - 1; slot++) {
+		btrfs_item_key_to_cpu(leaf, &leaf_key, slot);
+		btrfs_item_key_to_cpu(leaf, &key, slot + 1);
+
+		/* Make sure the keys are in the right order */
+		if (btrfs_comp_cpu_keys(&leaf_key, &key) >= 0) {
+			CORRUPT("bad key order", leaf, root, slot);
+			return -EIO;
+		}
+
+		/*
+		 * Make sure the offset and ends are right, remember that the
+		 * item data starts at the end of the leaf and grows towards the
+		 * front.
+		 */
+		if (btrfs_item_offset_nr(leaf, slot) !=
+			btrfs_item_end_nr(leaf, slot + 1)) {
+			CORRUPT("slot offset bad", leaf, root, slot);
+			return -EIO;
+		}
+
+		/*
+		 * Check to make sure that we don't point outside of the leaf,
+		 * just incase all the items are consistent to eachother, but
+		 * all point outside of the leaf.
+		 */
+		if (btrfs_item_end_nr(leaf, slot) >
+		    BTRFS_LEAF_DATA_SIZE(root)) {
+			CORRUPT("slot end outside of leaf", leaf, root, slot);
+			return -EIO;
+		}
+	}
+
+	return 0;
+}
+
 #ifdef CONFIG_DEBUG_LOCK_ALLOC
 void btrfs_set_buffer_lockdep_class(struct extent_buffer *eb, int level)
 {
@@ -444,7 +523,7 @@ static int btree_readpage_end_io_hook(struct page *page, u64 start, u64 end,
 	len = page->private >> 2;
 	WARN_ON(len == 0);
 
-	eb = alloc_extent_buffer(tree, start, len, page, GFP_NOFS);
+	eb = alloc_extent_buffer(tree, start, len, page);
 	if (eb == NULL) {
 		ret = -EIO;
 		goto out;
@@ -452,12 +531,10 @@ static int btree_readpage_end_io_hook(struct page *page, u64 start, u64 end,
 
 	found_start = btrfs_header_bytenr(eb);
 	if (found_start != start) {
-		if (printk_ratelimit()) {
-			printk(KERN_INFO "btrfs bad tree block start "
+		printk_ratelimited(KERN_INFO "btrfs bad tree block start "
 			       "%llu %llu\n",
 			       (unsigned long long)found_start,
 			       (unsigned long long)eb->start);
-		}
 		ret = -EIO;
 		goto err;
 	}
@@ -469,10 +546,8 @@ static int btree_readpage_end_io_hook(struct page *page, u64 start, u64 end,
 		goto err;
 	}
 	if (check_tree_block_fsid(root, eb)) {
-		if (printk_ratelimit()) {
-			printk(KERN_INFO "btrfs bad fsid on block %llu\n",
+		printk_ratelimited(KERN_INFO "btrfs bad fsid on block %llu\n",
 			       (unsigned long long)eb->start);
-		}
 		ret = -EIO;
 		goto err;
 	}
@@ -481,8 +556,20 @@ static int btree_readpage_end_io_hook(struct page *page, u64 start, u64 end,
 	btrfs_set_buffer_lockdep_class(eb, found_level);
 
 	ret = csum_tree_block(root, eb, 1);
-	if (ret)
+	if (ret) {
 		ret = -EIO;
+		goto err;
+	}
+
+	/*
+	 * If this is a leaf block and it is corrupt, set the corrupt bit so
+	 * that we don't try and read the other copies of this block, just
+	 * return -EIO.
+	 */
+	if (found_level == 0 && check_leaf(root, eb)) {
+		set_bit(EXTENT_BUFFER_CORRUPT, &eb->bflags);
+		ret = -EIO;
+	}
 
 	end = min_t(u64, eb->len, PAGE_CACHE_SIZE);
 	end = eb->start + end - 1;
@@ -557,12 +644,6 @@ unsigned long btrfs_async_submit_limit(struct btrfs_fs_info *info)
 	return 256 * limit;
 }
 
-int btrfs_congested_async(struct btrfs_fs_info *info, int iodone)
-{
-	return atomic_read(&info->nr_async_bios) >
-		btrfs_async_submit_limit(info);
-}
-
 static void run_one_async_start(struct btrfs_work *work)
 {
 	struct async_submit_bio *async;
@@ -843,7 +924,6 @@ static const struct address_space_operations btree_aops = {
 	.writepages	= btree_writepages,
 	.releasepage	= btree_releasepage,
 	.invalidatepage = btree_invalidatepage,
-	.sync_page	= block_sync_page,
 #ifdef CONFIG_MIGRATION
 	.migratepage	= btree_migratepage,
 #endif
@@ -871,7 +951,7 @@ struct extent_buffer *btrfs_find_tree_block(struct btrfs_root *root,
 	struct inode *btree_inode = root->fs_info->btree_inode;
 	struct extent_buffer *eb;
 	eb = find_extent_buffer(&BTRFS_I(btree_inode)->io_tree,
-				bytenr, blocksize, GFP_NOFS);
+				bytenr, blocksize);
 	return eb;
 }
 
@@ -882,7 +962,7 @@ struct extent_buffer *btrfs_find_create_tree_block(struct btrfs_root *root,
 	struct extent_buffer *eb;
 
 	eb = alloc_extent_buffer(&BTRFS_I(btree_inode)->io_tree,
-				 bytenr, blocksize, NULL, GFP_NOFS);
+				 bytenr, blocksize, NULL);
 	return eb;
 }
 
@@ -966,13 +1046,13 @@ static int __setup_root(u32 nodesize, u32 leafsize, u32 sectorsize,
 	root->name = NULL;
 	root->in_sysfs = 0;
 	root->inode_tree = RB_ROOT;
+	INIT_RADIX_TREE(&root->delayed_nodes_tree, GFP_ATOMIC);
 	root->block_rsv = NULL;
 	root->orphan_block_rsv = NULL;
 
 	INIT_LIST_HEAD(&root->dirty_list);
 	INIT_LIST_HEAD(&root->orphan_list);
 	INIT_LIST_HEAD(&root->root_list);
-	spin_lock_init(&root->node_lock);
 	spin_lock_init(&root->orphan_lock);
 	spin_lock_init(&root->inode_lock);
 	spin_lock_init(&root->accounting_lock);
@@ -988,7 +1068,7 @@ static int __setup_root(u32 nodesize, u32 leafsize, u32 sectorsize,
 	root->log_transid = 0;
 	root->last_log_commit = 0;
 	extent_io_tree_init(&root->dirty_log_pages,
-			     fs_info->btree_inode->i_mapping, GFP_NOFS);
+			     fs_info->btree_inode->i_mapping);
 
 	memset(&root->root_key, 0, sizeof(root->root_key));
 	memset(&root->root_item, 0, sizeof(root->root_item));
@@ -1156,7 +1236,10 @@ struct btrfs_root *btrfs_read_fs_root_no_radix(struct btrfs_root *tree_root,
 		     root, fs_info, location->objectid);
 
 	path = btrfs_alloc_path();
-	BUG_ON(!path);
+	if (!path) {
+		kfree(root);
+		return ERR_PTR(-ENOMEM);
+	}
 	ret = btrfs_search_slot(NULL, tree_root, location, path, 0, 0);
 	if (ret == 0) {
 		l = path->nodes[0];
@@ -1180,27 +1263,14 @@ struct btrfs_root *btrfs_read_fs_root_no_radix(struct btrfs_root *tree_root,
 	root->commit_root = btrfs_root_node(root);
 	BUG_ON(!root->node);
 out:
-	if (location->objectid != BTRFS_TREE_LOG_OBJECTID)
+	if (location->objectid != BTRFS_TREE_LOG_OBJECTID) {
 		root->ref_cows = 1;
+		btrfs_check_and_init_root_item(&root->root_item);
+	}
 
 	return root;
 }
 
-struct btrfs_root *btrfs_lookup_fs_root(struct btrfs_fs_info *fs_info,
-					u64 root_objectid)
-{
-	struct btrfs_root *root;
-
-	if (root_objectid == BTRFS_ROOT_TREE_OBJECTID)
-		return fs_info->tree_root;
-	if (root_objectid == BTRFS_EXTENT_TREE_OBJECTID)
-		return fs_info->extent_root;
-
-	root = radix_tree_lookup(&fs_info->fs_roots_radix,
-				 (unsigned long)root_objectid);
-	return root;
-}
-
 struct btrfs_root *btrfs_read_fs_root_no_name(struct btrfs_fs_info *fs_info,
 					      struct btrfs_key *location)
 {
@@ -1229,6 +1299,19 @@ again:
 	if (IS_ERR(root))
 		return root;
 
+	root->free_ino_ctl = kzalloc(sizeof(*root->free_ino_ctl), GFP_NOFS);
+	if (!root->free_ino_ctl)
+		goto fail;
+	root->free_ino_pinned = kzalloc(sizeof(*root->free_ino_pinned),
+					GFP_NOFS);
+	if (!root->free_ino_pinned)
+		goto fail;
+
+	btrfs_init_free_ino_ctl(root);
+	mutex_init(&root->fs_commit_mutex);
+	spin_lock_init(&root->cache_lock);
+	init_waitqueue_head(&root->cache_wait);
+
 	set_anon_super(&root->anon_super, NULL);
 
 	if (btrfs_root_refs(&root->root_item) == 0) {
@@ -1272,41 +1355,6 @@ fail:
 	return ERR_PTR(ret);
 }
 
-struct btrfs_root *btrfs_read_fs_root(struct btrfs_fs_info *fs_info,
-				      struct btrfs_key *location,
-				      const char *name, int namelen)
-{
-	return btrfs_read_fs_root_no_name(fs_info, location);
-#if 0
-	struct btrfs_root *root;
-	int ret;
-
-	root = btrfs_read_fs_root_no_name(fs_info, location);
-	if (!root)
-		return NULL;
-
-	if (root->in_sysfs)
-		return root;
-
-	ret = btrfs_set_root_name(root, name, namelen);
-	if (ret) {
-		free_extent_buffer(root->node);
-		kfree(root);
-		return ERR_PTR(ret);
-	}
-
-	ret = btrfs_sysfs_add_root(root);
-	if (ret) {
-		free_extent_buffer(root->node);
-		kfree(root->name);
-		kfree(root);
-		return ERR_PTR(ret);
-	}
-	root->in_sysfs = 1;
-	return root;
-#endif
-}
-
 static int btrfs_congested_fn(void *congested_data, int bdi_bits)
 {
 	struct btrfs_fs_info *info = (struct btrfs_fs_info *)congested_data;
@@ -1314,7 +1362,8 @@ static int btrfs_congested_fn(void *congested_data, int bdi_bits)
 	struct btrfs_device *device;
 	struct backing_dev_info *bdi;
 
-	list_for_each_entry(device, &info->fs_devices->devices, dev_list) {
+	rcu_read_lock();
+	list_for_each_entry_rcu(device, &info->fs_devices->devices, dev_list) {
 		if (!device->bdev)
 			continue;
 		bdi = blk_get_backing_dev_info(device->bdev);
@@ -1323,86 +1372,11 @@ static int btrfs_congested_fn(void *congested_data, int bdi_bits)
 			break;
 		}
 	}
+	rcu_read_unlock();
 	return ret;
 }
 
 /*
- * this unplugs every device on the box, and it is only used when page
- * is null
- */
-static void __unplug_io_fn(struct backing_dev_info *bdi, struct page *page)
-{
-	struct btrfs_device *device;
-	struct btrfs_fs_info *info;
-
-	info = (struct btrfs_fs_info *)bdi->unplug_io_data;
-	list_for_each_entry(device, &info->fs_devices->devices, dev_list) {
-		if (!device->bdev)
-			continue;
-
-		bdi = blk_get_backing_dev_info(device->bdev);
-		if (bdi->unplug_io_fn)
-			bdi->unplug_io_fn(bdi, page);
-	}
-}
-
-static void btrfs_unplug_io_fn(struct backing_dev_info *bdi, struct page *page)
-{
-	struct inode *inode;
-	struct extent_map_tree *em_tree;
-	struct extent_map *em;
-	struct address_space *mapping;
-	u64 offset;
-
-	/* the generic O_DIRECT read code does this */
-	if (1 || !page) {
-		__unplug_io_fn(bdi, page);
-		return;
-	}
-
-	/*
-	 * page->mapping may change at any time.  Get a consistent copy
-	 * and use that for everything below
-	 */
-	smp_mb();
-	mapping = page->mapping;
-	if (!mapping)
-		return;
-
-	inode = mapping->host;
-
-	/*
-	 * don't do the expensive searching for a small number of
-	 * devices
-	 */
-	if (BTRFS_I(inode)->root->fs_info->fs_devices->open_devices <= 2) {
-		__unplug_io_fn(bdi, page);
-		return;
-	}
-
-	offset = page_offset(page);
-
-	em_tree = &BTRFS_I(inode)->extent_tree;
-	read_lock(&em_tree->lock);
-	em = lookup_extent_mapping(em_tree, offset, PAGE_CACHE_SIZE);
-	read_unlock(&em_tree->lock);
-	if (!em) {
-		__unplug_io_fn(bdi, page);
-		return;
-	}
-
-	if (em->block_start >= EXTENT_MAP_LAST_BYTE) {
-		free_extent_map(em);
-		__unplug_io_fn(bdi, page);
-		return;
-	}
-	offset = offset - em->start;
-	btrfs_unplug_page(&BTRFS_I(inode)->root->fs_info->mapping_tree,
-			  em->block_start + offset, page);
-	free_extent_map(em);
-}
-
-/*
  * If this fails, caller must call bdi_destroy() to get rid of the
  * bdi again.
  */
@@ -1416,8 +1390,6 @@ static int setup_bdi(struct btrfs_fs_info *info, struct backing_dev_info *bdi)
 		return err;
 
 	bdi->ra_pages	= default_backing_dev_info.ra_pages;
-	bdi->unplug_io_fn	= btrfs_unplug_io_fn;
-	bdi->unplug_io_data	= info;
 	bdi->congested_fn	= btrfs_congested_fn;
 	bdi->congested_data	= info;
 	return 0;
@@ -1503,6 +1475,7 @@ static int cleaner_kthread(void *arg)
 			btrfs_run_delayed_iputs(root);
 			btrfs_clean_old_snapshots(root);
 			mutex_unlock(&root->fs_info->cleaner_mutex);
+			btrfs_run_defrag_inodes(root->fs_info);
 		}
 
 		if (freezing(current)) {
@@ -1592,7 +1565,7 @@ struct btrfs_root *open_ctree(struct super_block *sb,
 	struct btrfs_root *csum_root = kzalloc(sizeof(struct btrfs_root),
 						 GFP_NOFS);
 	struct btrfs_root *tree_root = btrfs_sb(sb);
-	struct btrfs_fs_info *fs_info = tree_root->fs_info;
+	struct btrfs_fs_info *fs_info = NULL;
 	struct btrfs_root *chunk_root = kzalloc(sizeof(struct btrfs_root),
 						GFP_NOFS);
 	struct btrfs_root *dev_root = kzalloc(sizeof(struct btrfs_root),
@@ -1604,11 +1577,12 @@ struct btrfs_root *open_ctree(struct super_block *sb,
 
 	struct btrfs_super_block *disk_super;
 
-	if (!extent_root || !tree_root || !fs_info ||
+	if (!extent_root || !tree_root || !tree_root->fs_info ||
 	    !chunk_root || !dev_root || !csum_root) {
 		err = -ENOMEM;
 		goto fail;
 	}
+	fs_info = tree_root->fs_info;
 
 	ret = init_srcu_struct(&fs_info->subvol_srcu);
 	if (ret) {
@@ -1628,6 +1602,8 @@ struct btrfs_root *open_ctree(struct super_block *sb,
 		goto fail_bdi;
 	}
 
+	fs_info->btree_inode->i_mapping->flags &= ~__GFP_FS;
+
 	INIT_RADIX_TREE(&fs_info->fs_roots_radix, GFP_ATOMIC);
 	INIT_LIST_HEAD(&fs_info->trans_list);
 	INIT_LIST_HEAD(&fs_info->dead_roots);
@@ -1641,6 +1617,7 @@ struct btrfs_root *open_ctree(struct super_block *sb,
 	spin_lock_init(&fs_info->ref_cache_lock);
 	spin_lock_init(&fs_info->fs_roots_radix_lock);
 	spin_lock_init(&fs_info->delayed_iput_lock);
+	spin_lock_init(&fs_info->defrag_inodes_lock);
 
 	init_completion(&fs_info->kobj_unregister);
 	fs_info->tree_root = tree_root;
@@ -1663,15 +1640,35 @@ struct btrfs_root *open_ctree(struct super_block *sb,
 	atomic_set(&fs_info->async_delalloc_pages, 0);
 	atomic_set(&fs_info->async_submit_draining, 0);
 	atomic_set(&fs_info->nr_async_bios, 0);
+	atomic_set(&fs_info->defrag_running, 0);
 	fs_info->sb = sb;
 	fs_info->max_inline = 8192 * 1024;
 	fs_info->metadata_ratio = 0;
+	fs_info->defrag_inodes = RB_ROOT;
 
 	fs_info->thread_pool_size = min_t(unsigned long,
 					  num_online_cpus() + 2, 8);
 
 	INIT_LIST_HEAD(&fs_info->ordered_extents);
 	spin_lock_init(&fs_info->ordered_extent_lock);
+	fs_info->delayed_root = kmalloc(sizeof(struct btrfs_delayed_root),
+					GFP_NOFS);
+	if (!fs_info->delayed_root) {
+		err = -ENOMEM;
+		goto fail_iput;
+	}
+	btrfs_init_delayed_root(fs_info->delayed_root);
+
+	mutex_init(&fs_info->scrub_lock);
+	atomic_set(&fs_info->scrubs_running, 0);
+	atomic_set(&fs_info->scrub_pause_req, 0);
+	atomic_set(&fs_info->scrubs_paused, 0);
+	atomic_set(&fs_info->scrub_cancel_req, 0);
+	init_waitqueue_head(&fs_info->scrub_pause_wait);
+	init_rwsem(&fs_info->scrub_super_lock);
+	fs_info->scrub_workers_refcnt = 0;
+	btrfs_init_workers(&fs_info->scrub_workers, "scrub",
+			   fs_info->thread_pool_size, &fs_info->generic_worker);
 
 	sb->s_blocksize = 4096;
 	sb->s_blocksize_bits = blksize_bits(4096);
@@ -1690,10 +1687,8 @@ struct btrfs_root *open_ctree(struct super_block *sb,
 
 	RB_CLEAR_NODE(&BTRFS_I(fs_info->btree_inode)->rb_node);
 	extent_io_tree_init(&BTRFS_I(fs_info->btree_inode)->io_tree,
-			     fs_info->btree_inode->i_mapping,
-			     GFP_NOFS);
-	extent_map_tree_init(&BTRFS_I(fs_info->btree_inode)->extent_tree,
-			     GFP_NOFS);
+			     fs_info->btree_inode->i_mapping);
+	extent_map_tree_init(&BTRFS_I(fs_info->btree_inode)->extent_tree);
 
 	BTRFS_I(fs_info->btree_inode)->io_tree.ops = &btree_extent_io_ops;
 
@@ -1707,9 +1702,9 @@ struct btrfs_root *open_ctree(struct super_block *sb,
 	fs_info->block_group_cache_tree = RB_ROOT;
 
 	extent_io_tree_init(&fs_info->freed_extents[0],
-			     fs_info->btree_inode->i_mapping, GFP_NOFS);
+			     fs_info->btree_inode->i_mapping);
 	extent_io_tree_init(&fs_info->freed_extents[1],
-			     fs_info->btree_inode->i_mapping, GFP_NOFS);
+			     fs_info->btree_inode->i_mapping);
 	fs_info->pinned_extents = &fs_info->freed_extents[0];
 	fs_info->do_barriers = 1;
 
@@ -1739,7 +1734,7 @@ struct btrfs_root *open_ctree(struct super_block *sb,
 	bh = btrfs_read_dev_super(fs_devices->latest_bdev);
 	if (!bh) {
 		err = -EINVAL;
-		goto fail_iput;
+		goto fail_alloc;
 	}
 
 	memcpy(&fs_info->super_copy, bh->b_data, sizeof(fs_info->super_copy));
@@ -1751,17 +1746,23 @@ struct btrfs_root *open_ctree(struct super_block *sb,
 
 	disk_super = &fs_info->super_copy;
 	if (!btrfs_super_root(disk_super))
-		goto fail_iput;
+		goto fail_alloc;
 
 	/* check FS state, whether FS is broken. */
 	fs_info->fs_state |= btrfs_super_flags(disk_super);
 
 	btrfs_check_super_valid(fs_info, sb->s_flags & MS_RDONLY);
 
+	/*
+	 * In the long term, we'll store the compression type in the super
+	 * block, and it'll be used for per file compression control.
+	 */
+	fs_info->compress_type = BTRFS_COMPRESS_ZLIB;
+
 	ret = btrfs_parse_options(tree_root, options);
 	if (ret) {
 		err = ret;
-		goto fail_iput;
+		goto fail_alloc;
 	}
 
 	features = btrfs_super_incompat_flags(disk_super) &
@@ -1771,7 +1772,7 @@ struct btrfs_root *open_ctree(struct super_block *sb,
 		       "unsupported optional features (%Lx).\n",
 		       (unsigned long long)features);
 		err = -EINVAL;
-		goto fail_iput;
+		goto fail_alloc;
 	}
 
 	features = btrfs_super_incompat_flags(disk_super);
@@ -1787,7 +1788,7 @@ struct btrfs_root *open_ctree(struct super_block *sb,
 		       "unsupported option features (%Lx).\n",
 		       (unsigned long long)features);
 		err = -EINVAL;
-		goto fail_iput;
+		goto fail_alloc;
 	}
 
 	btrfs_init_workers(&fs_info->generic_worker,
@@ -1834,6 +1835,9 @@ struct btrfs_root *open_ctree(struct super_block *sb,
 			   &fs_info->generic_worker);
 	btrfs_init_workers(&fs_info->endio_freespace_worker, "freespace-write",
 			   1, &fs_info->generic_worker);
+	btrfs_init_workers(&fs_info->delayed_workers, "delayed-meta",
+			   fs_info->thread_pool_size,
+			   &fs_info->generic_worker);
 
 	/*
 	 * endios are largely parallel and should have a very
@@ -1855,6 +1859,7 @@ struct btrfs_root *open_ctree(struct super_block *sb,
 	btrfs_start_workers(&fs_info->endio_meta_write_workers, 1);
 	btrfs_start_workers(&fs_info->endio_write_workers, 1);
 	btrfs_start_workers(&fs_info->endio_freespace_worker, 1);
+	btrfs_start_workers(&fs_info->delayed_workers, 1);
 
 	fs_info->bdi.ra_pages *= btrfs_super_num_devices(disk_super);
 	fs_info->bdi.ra_pages = max(fs_info->bdi.ra_pages,
@@ -1963,6 +1968,12 @@ struct btrfs_root *open_ctree(struct super_block *sb,
 	fs_info->metadata_alloc_profile = (u64)-1;
 	fs_info->system_alloc_profile = fs_info->metadata_alloc_profile;
 
+	ret = btrfs_init_space_info(fs_info);
+	if (ret) {
+		printk(KERN_ERR "Failed to initial space info: %d\n", ret);
+		goto fail_block_groups;
+	}
+
 	ret = btrfs_read_block_groups(extent_root);
 	if (ret) {
 		printk(KERN_ERR "Failed to read block groups: %d\n", ret);
@@ -2054,9 +2065,14 @@ struct btrfs_root *open_ctree(struct super_block *sb,
 
 	if (!(sb->s_flags & MS_RDONLY)) {
 		down_read(&fs_info->cleanup_work_sem);
-		btrfs_orphan_cleanup(fs_info->fs_root);
-		btrfs_orphan_cleanup(fs_info->tree_root);
+		err = btrfs_orphan_cleanup(fs_info->fs_root);
+		if (!err)
+			err = btrfs_orphan_cleanup(fs_info->tree_root);
 		up_read(&fs_info->cleanup_work_sem);
+		if (err) {
+			close_ctree(tree_root);
+			return ERR_PTR(err);
+		}
 	}
 
 	return tree_root;
@@ -2100,6 +2116,9 @@ fail_sb_buffer:
 	btrfs_stop_workers(&fs_info->endio_write_workers);
 	btrfs_stop_workers(&fs_info->endio_freespace_worker);
 	btrfs_stop_workers(&fs_info->submit_workers);
+	btrfs_stop_workers(&fs_info->delayed_workers);
+fail_alloc:
+	kfree(fs_info->delayed_root);
 fail_iput:
 	invalidate_inode_pages2(fs_info->btree_inode->i_mapping);
 	iput(fs_info->btree_inode);
@@ -2127,11 +2146,9 @@ static void btrfs_end_buffer_write_sync(struct buffer_head *bh, int uptodate)
 	if (uptodate) {
 		set_buffer_uptodate(bh);
 	} else {
-		if (printk_ratelimit()) {
-			printk(KERN_WARNING "lost page write due to "
+		printk_ratelimited(KERN_WARNING "lost page write due to "
 					"I/O error on %s\n",
 				       bdevname(bh->b_bdev, b));
-		}
 		/* note, we dont' set_buffer_write_io_error because we have
 		 * our own ways of dealing with the IO errors
 		 */
@@ -2295,7 +2312,7 @@ int write_all_supers(struct btrfs_root *root, int max_mirrors)
 
 	mutex_lock(&root->fs_info->fs_devices->device_list_mutex);
 	head = &root->fs_info->fs_devices->devices;
-	list_for_each_entry(dev, head, dev_list) {
+	list_for_each_entry_rcu(dev, head, dev_list) {
 		if (!dev->bdev) {
 			total_errors++;
 			continue;
@@ -2328,7 +2345,7 @@ int write_all_supers(struct btrfs_root *root, int max_mirrors)
 	}
 
 	total_errors = 0;
-	list_for_each_entry(dev, head, dev_list) {
+	list_for_each_entry_rcu(dev, head, dev_list) {
 		if (!dev->bdev)
 			continue;
 		if (!dev->in_fs_metadata || !dev->writeable)
@@ -2366,12 +2383,15 @@ int btrfs_free_fs_root(struct btrfs_fs_info *fs_info, struct btrfs_root *root)
 	if (btrfs_root_refs(&root->root_item) == 0)
 		synchronize_srcu(&fs_info->subvol_srcu);
 
+	__btrfs_remove_free_space_cache(root->free_ino_pinned);
+	__btrfs_remove_free_space_cache(root->free_ino_ctl);
 	free_fs_root(root);
 	return 0;
 }
 
 static void free_fs_root(struct btrfs_root *root)
 {
+	iput(root->cache_inode);
 	WARN_ON(!RB_EMPTY_ROOT(&root->inode_tree));
 	if (root->anon_super.s_dev) {
 		down_write(&root->anon_super.s_umount);
@@ -2379,6 +2399,8 @@ static void free_fs_root(struct btrfs_root *root)
 	}
 	free_extent_buffer(root->node);
 	free_extent_buffer(root->commit_root);
+	kfree(root->free_ino_ctl);
+	kfree(root->free_ino_pinned);
 	kfree(root->name);
 	kfree(root);
 }
@@ -2431,8 +2453,12 @@ int btrfs_cleanup_fs_roots(struct btrfs_fs_info *fs_info)
 
 		root_objectid = gang[ret - 1]->root_key.objectid + 1;
 		for (i = 0; i < ret; i++) {
+			int err;
+
 			root_objectid = gang[i]->root_key.objectid;
-			btrfs_orphan_cleanup(gang[i]);
+			err = btrfs_orphan_cleanup(gang[i]);
+			if (err)
+				return err;
 		}
 		root_objectid++;
 	}
@@ -2478,6 +2504,15 @@ int close_ctree(struct btrfs_root *root)
 	fs_info->closing = 1;
 	smp_mb();
 
+	btrfs_scrub_cancel(root);
+
+	/* wait for any defraggers to finish */
+	wait_event(fs_info->transaction_wait,
+		   (atomic_read(&fs_info->defrag_running) == 0));
+
+	/* clear out the rbtree of defraggable inodes */
+	btrfs_run_defrag_inodes(root->fs_info);
+
 	btrfs_put_block_group_cache(fs_info);
 
 	/*
@@ -2489,7 +2524,7 @@ int close_ctree(struct btrfs_root *root)
 	 * ERROR state on disk.
 	 *
 	 * 2. when btrfs flips readonly just in btrfs_commit_super,
-	 * and in such case, btrfs cannnot write sb via btrfs_commit_super,
+	 * and in such case, btrfs cannot write sb via btrfs_commit_super,
 	 * and since fs_state has been set BTRFS_SUPER_FLAG_ERROR flag,
 	 * btrfs will cleanup all FS resources first and write sb then.
 	 */
@@ -2536,6 +2571,7 @@ int close_ctree(struct btrfs_root *root)
 	del_fs_roots(fs_info);
 
 	iput(fs_info->btree_inode);
+	kfree(fs_info->delayed_root);
 
 	btrfs_stop_workers(&fs_info->generic_worker);
 	btrfs_stop_workers(&fs_info->fixup_workers);
@@ -2547,6 +2583,7 @@ int close_ctree(struct btrfs_root *root)
 	btrfs_stop_workers(&fs_info->endio_write_workers);
 	btrfs_stop_workers(&fs_info->endio_freespace_worker);
 	btrfs_stop_workers(&fs_info->submit_workers);
+	btrfs_stop_workers(&fs_info->delayed_workers);
 
 	btrfs_close_devices(fs_info->fs_devices);
 	btrfs_mapping_tree_free(&fs_info->mapping_tree);
@@ -2623,6 +2660,29 @@ void btrfs_btree_balance_dirty(struct btrfs_root *root, unsigned long nr)
 	if (current->flags & PF_MEMALLOC)
 		return;
 
+	btrfs_balance_delayed_items(root);
+
+	num_dirty = root->fs_info->dirty_metadata_bytes;
+
+	if (num_dirty > thresh) {
+		balance_dirty_pages_ratelimited_nr(
+				   root->fs_info->btree_inode->i_mapping, 1);
+	}
+	return;
+}
+
+void __btrfs_btree_balance_dirty(struct btrfs_root *root, unsigned long nr)
+{
+	/*
+	 * looks as though older kernels can get into trouble with
+	 * this code, they end up stuck in balance_dirty_pages forever
+	 */
+	u64 num_dirty;
+	unsigned long thresh = 32 * 1024 * 1024;
+
+	if (current->flags & PF_MEMALLOC)
+		return;
+
 	num_dirty = root->fs_info->dirty_metadata_bytes;
 
 	if (num_dirty > thresh) {
@@ -2655,7 +2715,7 @@ int btree_lock_page_hook(struct page *page)
 		goto out;
 
 	len = page->private >> 2;
-	eb = find_extent_buffer(io_tree, bytenr, len, GFP_NOFS);
+	eb = find_extent_buffer(io_tree, bytenr, len);
 	if (!eb)
 		goto out;
 
@@ -2782,6 +2842,7 @@ static int btrfs_destroy_delayed_refs(struct btrfs_transaction *trans,
 
 	spin_lock(&delayed_refs->lock);
 	if (delayed_refs->num_entries == 0) {
+		spin_unlock(&delayed_refs->lock);
 		printk(KERN_INFO "delayed_refs has NO entry\n");
 		return ret;
 	}
@@ -2943,7 +3004,10 @@ static int btrfs_destroy_pinned_extent(struct btrfs_root *root,
 			break;
 
 		/* opt_discard */
-		ret = btrfs_error_discard_extent(root, start, end + 1 - start);
+		if (btrfs_test_opt(root, DISCARD))
+			ret = btrfs_error_discard_extent(root, start,
+							 end + 1 - start,
+							 NULL);
 
 		clear_extent_dirty(unpin, start, end, GFP_NOFS);
 		btrfs_error_unpin_extent_range(root, start, end);
@@ -3012,7 +3076,7 @@ static int btrfs_cleanup_transaction(struct btrfs_root *root)
 		btrfs_destroy_pinned_extent(root,
 					    root->fs_info->pinned_extents);
 
-		t->use_count = 0;
+		atomic_set(&t->use_count, 0);
 		list_del_init(&t->list);
 		memset(t, 0, sizeof(*t));
 		kmem_cache_free(btrfs_transaction_cachep, t);