1 files changed, 296 insertions, 244 deletions

diff --git a/fs/btrfs/block-group.c b/fs/btrfs/block-group.c
index 404e050ce8ee..613920c17ac1 100644
--- a/fs/btrfs/block-group.c
+++ b/fs/btrfs/block-group.c
@@ -7,7 +7,6 @@
 #include "disk-io.h"
 #include "free-space-cache.h"
 #include "free-space-tree.h"
-#include "disk-io.h"
 #include "volumes.h"
 #include "transaction.h"
 #include "ref-verify.h"
@@ -66,11 +65,8 @@ static u64 btrfs_reduce_alloc_profile(struct btrfs_fs_info *fs_info, u64 flags)
 	spin_lock(&fs_info->balance_lock);
 	target = get_restripe_target(fs_info, flags);
 	if (target) {
-		/* Pick target profile only if it's already available */
-		if ((flags & target) & BTRFS_EXTENDED_PROFILE_MASK) {
-			spin_unlock(&fs_info->balance_lock);
-			return extended_to_chunk(target);
-		}
+		spin_unlock(&fs_info->balance_lock);
+		return extended_to_chunk(target);
 	}
 	spin_unlock(&fs_info->balance_lock);
 
@@ -119,12 +115,12 @@ u64 btrfs_get_alloc_profile(struct btrfs_fs_info *fs_info, u64 orig_flags)
 
 void btrfs_get_block_group(struct btrfs_block_group *cache)
 {
-	atomic_inc(&cache->count);
+	refcount_inc(&cache->refs);
 }
 
 void btrfs_put_block_group(struct btrfs_block_group *cache)
 {
-	if (atomic_dec_and_test(&cache->count)) {
+	if (refcount_dec_and_test(&cache->refs)) {
 		WARN_ON(cache->pinned > 0);
 		WARN_ON(cache->reserved > 0);
 
@@ -161,6 +157,8 @@ static int btrfs_add_block_group_cache(struct btrfs_fs_info *info,
 	struct rb_node *parent = NULL;
 	struct btrfs_block_group *cache;
 
+	ASSERT(block_group->length != 0);
+
 	spin_lock(&info->block_group_cache_lock);
 	p = &info->block_group_cache_tree.rb_node;
 
@@ -460,7 +458,7 @@ u64 add_new_free_space(struct btrfs_block_group *block_group, u64 start, u64 end
 	int ret;
 
 	while (start < end) {
-		ret = find_first_extent_bit(info->pinned_extents, start,
+		ret = find_first_extent_bit(&info->excluded_extents, start,
 					    &extent_start, &extent_end,
 					    EXTENT_DIRTY | EXTENT_UPTODATE,
 					    NULL);
@@ -856,18 +854,41 @@ static void clear_incompat_bg_bits(struct btrfs_fs_info *fs_info, u64 flags)
 				found_raid1c34 = true;
 			up_read(&sinfo->groups_sem);
 		}
-		if (found_raid56)
+		if (!found_raid56)
 			btrfs_clear_fs_incompat(fs_info, RAID56);
-		if (found_raid1c34)
+		if (!found_raid1c34)
 			btrfs_clear_fs_incompat(fs_info, RAID1C34);
 	}
 }
 
+static int remove_block_group_item(struct btrfs_trans_handle *trans,
+				   struct btrfs_path *path,
+				   struct btrfs_block_group *block_group)
+{
+	struct btrfs_fs_info *fs_info = trans->fs_info;
+	struct btrfs_root *root;
+	struct btrfs_key key;
+	int ret;
+
+	root = fs_info->extent_root;
+	key.objectid = block_group->start;
+	key.type = BTRFS_BLOCK_GROUP_ITEM_KEY;
+	key.offset = block_group->length;
+
+	ret = btrfs_search_slot(trans, root, &key, path, -1, 1);
+	if (ret > 0)
+		ret = -ENOENT;
+	if (ret < 0)
+		return ret;
+
+	ret = btrfs_del_item(trans, root, path);
+	return ret;
+}
+
 int btrfs_remove_block_group(struct btrfs_trans_handle *trans,
 			     u64 group_start, struct extent_map *em)
 {
 	struct btrfs_fs_info *fs_info = trans->fs_info;
-	struct btrfs_root *root = fs_info->extent_root;
 	struct btrfs_path *path;
 	struct btrfs_block_group *block_group;
 	struct btrfs_free_cluster *cluster;
@@ -992,6 +1013,9 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans,
 		 &fs_info->block_group_cache_tree);
 	RB_CLEAR_NODE(&block_group->cache_node);
 
+	/* Once for the block groups rbtree */
+	btrfs_put_block_group(block_group);
+
 	if (fs_info->first_logical_byte == block_group->start)
 		fs_info->first_logical_byte = (u64)-1;
 	spin_unlock(&fs_info->block_group_cache_lock);
@@ -1065,26 +1089,43 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans,
 
 	spin_unlock(&block_group->space_info->lock);
 
-	key.objectid = block_group->start;
-	key.type = BTRFS_BLOCK_GROUP_ITEM_KEY;
-	key.offset = block_group->length;
+	/*
+	 * Remove the free space for the block group from the free space tree
+	 * and the block group's item from the extent tree before marking the
+	 * block group as removed. This is to prevent races with tasks that
+	 * freeze and unfreeze a block group, this task and another task
+	 * allocating a new block group - the unfreeze task ends up removing
+	 * the block group's extent map before the task calling this function
+	 * deletes the block group item from the extent tree, allowing for
+	 * another task to attempt to create another block group with the same
+	 * item key (and failing with -EEXIST and a transaction abort).
+	 */
+	ret = remove_block_group_free_space(trans, block_group);
+	if (ret)
+		goto out;
+
+	ret = remove_block_group_item(trans, path, block_group);
+	if (ret < 0)
+		goto out;
 
-	mutex_lock(&fs_info->chunk_mutex);
 	spin_lock(&block_group->lock);
 	block_group->removed = 1;
 	/*
-	 * At this point trimming can't start on this block group, because we
-	 * removed the block group from the tree fs_info->block_group_cache_tree
-	 * so no one can't find it anymore and even if someone already got this
-	 * block group before we removed it from the rbtree, they have already
-	 * incremented block_group->trimming - if they didn't, they won't find
-	 * any free space entries because we already removed them all when we
-	 * called btrfs_remove_free_space_cache().
+	 * At this point trimming or scrub can't start on this block group,
+	 * because we removed the block group from the rbtree
+	 * fs_info->block_group_cache_tree so no one can't find it anymore and
+	 * even if someone already got this block group before we removed it
+	 * from the rbtree, they have already incremented block_group->frozen -
+	 * if they didn't, for the trimming case they won't find any free space
+	 * entries because we already removed them all when we called
+	 * btrfs_remove_free_space_cache().
 	 *
 	 * And we must not remove the extent map from the fs_info->mapping_tree
 	 * to prevent the same logical address range and physical device space
-	 * ranges from being reused for a new block group. This is because our
-	 * fs trim operation (btrfs_trim_fs() / btrfs_ioctl_fitrim()) is
+	 * ranges from being reused for a new block group. This is needed to
+	 * avoid races with trimming and scrub.
+	 *
+	 * An fs trim operation (btrfs_trim_fs() / btrfs_ioctl_fitrim()) is
 	 * completely transactionless, so while it is trimming a range the
 	 * currently running transaction might finish and a new one start,
 	 * allowing for new block groups to be created that can reuse the same
@@ -1095,28 +1136,9 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans,
 	 * in place until the extents have been discarded completely when
 	 * the transaction commit has completed.
 	 */
-	remove_em = (atomic_read(&block_group->trimming) == 0);
+	remove_em = (atomic_read(&block_group->frozen) == 0);
 	spin_unlock(&block_group->lock);
 
-	mutex_unlock(&fs_info->chunk_mutex);
-
-	ret = remove_block_group_free_space(trans, block_group);
-	if (ret)
-		goto out;
-
-	btrfs_put_block_group(block_group);
-	btrfs_put_block_group(block_group);
-
-	ret = btrfs_search_slot(trans, root, &key, path, -1, 1);
-	if (ret > 0)
-		ret = -EIO;
-	if (ret < 0)
-		goto out;
-
-	ret = btrfs_del_item(trans, root, path);
-	if (ret)
-		goto out;
-
 	if (remove_em) {
 		struct extent_map_tree *em_tree;
 
@@ -1127,7 +1149,10 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans,
 		/* once for the tree */
 		free_extent_map(em);
 	}
+
 out:
+	/* Once for the lookup reference */
+	btrfs_put_block_group(block_group);
 	if (remove_rsv)
 		btrfs_delayed_refs_rsv_release(fs_info, 1);
 	btrfs_free_path(path);
@@ -1171,7 +1196,7 @@ struct btrfs_trans_handle *btrfs_start_trans_remove_block_group(
 	free_extent_map(em);
 
 	return btrfs_start_transaction_fallback_global_rsv(fs_info->extent_root,
-							   num_items, 1);
+							   num_items);
 }
 
 /*
@@ -1248,6 +1273,51 @@ out:
 	return ret;
 }
 
+static bool clean_pinned_extents(struct btrfs_trans_handle *trans,
+				 struct btrfs_block_group *bg)
+{
+	struct btrfs_fs_info *fs_info = bg->fs_info;
+	struct btrfs_transaction *prev_trans = NULL;
+	const u64 start = bg->start;
+	const u64 end = start + bg->length - 1;
+	int ret;
+
+	spin_lock(&fs_info->trans_lock);
+	if (trans->transaction->list.prev != &fs_info->trans_list) {
+		prev_trans = list_last_entry(&trans->transaction->list,
+					     struct btrfs_transaction, list);
+		refcount_inc(&prev_trans->use_count);
+	}
+	spin_unlock(&fs_info->trans_lock);
+
+	/*
+	 * Hold the unused_bg_unpin_mutex lock to avoid racing with
+	 * btrfs_finish_extent_commit(). If we are at transaction N, another
+	 * task might be running finish_extent_commit() for the previous
+	 * transaction N - 1, and have seen a range belonging to the block
+	 * group in pinned_extents before we were able to clear the whole block
+	 * group range from pinned_extents. This means that task can lookup for
+	 * the block group after we unpinned it from pinned_extents and removed
+	 * it, leading to a BUG_ON() at unpin_extent_range().
+	 */
+	mutex_lock(&fs_info->unused_bg_unpin_mutex);
+	if (prev_trans) {
+		ret = clear_extent_bits(&prev_trans->pinned_extents, start, end,
+					EXTENT_DIRTY);
+		if (ret)
+			goto out;
+	}
+
+	ret = clear_extent_bits(&trans->transaction->pinned_extents, start, end,
+				EXTENT_DIRTY);
+out:
+	mutex_unlock(&fs_info->unused_bg_unpin_mutex);
+	if (prev_trans)
+		btrfs_put_transaction(prev_trans);
+
+	return ret == 0;
+}
+
 /*
  * Process the unused_bgs list and remove any that don't have any allocated
  * space inside of them.
@@ -1265,7 +1335,6 @@ void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info)
 
 	spin_lock(&fs_info->unused_bgs_lock);
 	while (!list_empty(&fs_info->unused_bgs)) {
-		u64 start, end;
 		int trimming;
 
 		block_group = list_first_entry(&fs_info->unused_bgs,
@@ -1344,35 +1413,10 @@ void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info)
 		 * We could have pending pinned extents for this block group,
 		 * just delete them, we don't care about them anymore.
 		 */
-		start = block_group->start;
-		end = start + block_group->length - 1;
-		/*
-		 * Hold the unused_bg_unpin_mutex lock to avoid racing with
-		 * btrfs_finish_extent_commit(). If we are at transaction N,
-		 * another task might be running finish_extent_commit() for the
-		 * previous transaction N - 1, and have seen a range belonging
-		 * to the block group in freed_extents[] before we were able to
-		 * clear the whole block group range from freed_extents[]. This
-		 * means that task can lookup for the block group after we
-		 * unpinned it from freed_extents[] and removed it, leading to
-		 * a BUG_ON() at btrfs_unpin_extent_range().
-		 */
-		mutex_lock(&fs_info->unused_bg_unpin_mutex);
-		ret = clear_extent_bits(&fs_info->freed_extents[0], start, end,
-				  EXTENT_DIRTY);
-		if (ret) {
-			mutex_unlock(&fs_info->unused_bg_unpin_mutex);
+		if (!clean_pinned_extents(trans, block_group)) {
 			btrfs_dec_block_group_ro(block_group);
 			goto end_trans;
 		}
-		ret = clear_extent_bits(&fs_info->freed_extents[1], start, end,
-				  EXTENT_DIRTY);
-		if (ret) {
-			mutex_unlock(&fs_info->unused_bg_unpin_mutex);
-			btrfs_dec_block_group_ro(block_group);
-			goto end_trans;
-		}
-		mutex_unlock(&fs_info->unused_bg_unpin_mutex);
 
 		/*
 		 * At this point, the block_group is read only and should fail
@@ -1421,7 +1465,7 @@ void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info)
 
 		/* Implicit trim during transaction commit. */
 		if (trimming)
-			btrfs_get_block_group_trimming(block_group);
+			btrfs_freeze_block_group(block_group);
 
 		/*
 		 * Btrfs_remove_chunk will abort the transaction if things go
@@ -1431,7 +1475,7 @@ void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info)
 
 		if (ret) {
 			if (trimming)
-				btrfs_put_block_group_trimming(block_group);
+				btrfs_unfreeze_block_group(block_group);
 			goto end_trans;
 		}
 
@@ -1482,21 +1526,70 @@ void btrfs_mark_bg_unused(struct btrfs_block_group *bg)
 	spin_unlock(&fs_info->unused_bgs_lock);
 }
 
+static int read_bg_from_eb(struct btrfs_fs_info *fs_info, struct btrfs_key *key,
+			   struct btrfs_path *path)
+{
+	struct extent_map_tree *em_tree;
+	struct extent_map *em;
+	struct btrfs_block_group_item bg;
+	struct extent_buffer *leaf;
+	int slot;
+	u64 flags;
+	int ret = 0;
+
+	slot = path->slots[0];
+	leaf = path->nodes[0];
+
+	em_tree = &fs_info->mapping_tree;
+	read_lock(&em_tree->lock);
+	em = lookup_extent_mapping(em_tree, key->objectid, key->offset);
+	read_unlock(&em_tree->lock);
+	if (!em) {
+		btrfs_err(fs_info,
+			  "logical %llu len %llu found bg but no related chunk",
+			  key->objectid, key->offset);
+		return -ENOENT;
+	}
+
+	if (em->start != key->objectid || em->len != key->offset) {
+		btrfs_err(fs_info,
+			"block group %llu len %llu mismatch with chunk %llu len %llu",
+			key->objectid, key->offset, em->start, em->len);
+		ret = -EUCLEAN;
+		goto out_free_em;
+	}
+
+	read_extent_buffer(leaf, &bg, btrfs_item_ptr_offset(leaf, slot),
+			   sizeof(bg));
+	flags = btrfs_stack_block_group_flags(&bg) &
+		BTRFS_BLOCK_GROUP_TYPE_MASK;
+
+	if (flags != (em->map_lookup->type & BTRFS_BLOCK_GROUP_TYPE_MASK)) {
+		btrfs_err(fs_info,
+"block group %llu len %llu type flags 0x%llx mismatch with chunk type flags 0x%llx",
+			  key->objectid, key->offset, flags,
+			  (BTRFS_BLOCK_GROUP_TYPE_MASK & em->map_lookup->type));
+		ret = -EUCLEAN;
+	}
+
+out_free_em:
+	free_extent_map(em);
+	return ret;
+}
+
 static int find_first_block_group(struct btrfs_fs_info *fs_info,
 				  struct btrfs_path *path,
 				  struct btrfs_key *key)
 {
 	struct btrfs_root *root = fs_info->extent_root;
-	int ret = 0;
+	int ret;
 	struct btrfs_key found_key;
 	struct extent_buffer *leaf;
-	struct btrfs_block_group_item bg;
-	u64 flags;
 	int slot;
 
 	ret = btrfs_search_slot(NULL, root, key, path, 0, 0);
 	if (ret < 0)
-		goto out;
+		return ret;
 
 	while (1) {
 		slot = path->slots[0];
@@ -1513,49 +1606,10 @@ static int find_first_block_group(struct btrfs_fs_info *fs_info,
 
 		if (found_key.objectid >= key->objectid &&
 		    found_key.type == BTRFS_BLOCK_GROUP_ITEM_KEY) {
-			struct extent_map_tree *em_tree;
-			struct extent_map *em;
-
-			em_tree = &root->fs_info->mapping_tree;
-			read_lock(&em_tree->lock);
-			em = lookup_extent_mapping(em_tree, found_key.objectid,
-						   found_key.offset);
-			read_unlock(&em_tree->lock);
-			if (!em) {
-				btrfs_err(fs_info,
-			"logical %llu len %llu found bg but no related chunk",
-					  found_key.objectid, found_key.offset);
-				ret = -ENOENT;
-			} else if (em->start != found_key.objectid ||
-				   em->len != found_key.offset) {
-				btrfs_err(fs_info,
-		"block group %llu len %llu mismatch with chunk %llu len %llu",
-					  found_key.objectid, found_key.offset,
-					  em->start, em->len);
-				ret = -EUCLEAN;
-			} else {
-				read_extent_buffer(leaf, &bg,
-					btrfs_item_ptr_offset(leaf, slot),
-					sizeof(bg));
-				flags = btrfs_stack_block_group_flags(&bg) &
-					BTRFS_BLOCK_GROUP_TYPE_MASK;
-
-				if (flags != (em->map_lookup->type &
-					      BTRFS_BLOCK_GROUP_TYPE_MASK)) {
-					btrfs_err(fs_info,
-"block group %llu len %llu type flags 0x%llx mismatch with chunk type flags 0x%llx",
-						found_key.objectid,
-						found_key.offset, flags,
-						(BTRFS_BLOCK_GROUP_TYPE_MASK &
-						 em->map_lookup->type));
-					ret = -EUCLEAN;
-				} else {
-					ret = 0;
-				}
-			}
-			free_extent_map(em);
-			goto out;
+			ret = read_bg_from_eb(fs_info, &found_key, path);
+			break;
 		}
+
 		path->slots[0]++;
 	}
 out:
@@ -1607,19 +1661,12 @@ int btrfs_rmap_block(struct btrfs_fs_info *fs_info, u64 chunk_start,
 		return -EIO;
 
 	map = em->map_lookup;
-	data_stripe_length = em->len;
+	data_stripe_length = em->orig_block_len;
 	io_stripe_size = map->stripe_len;
 
-	if (map->type & BTRFS_BLOCK_GROUP_RAID10)
-		data_stripe_length = div_u64(data_stripe_length,
-					     map->num_stripes / map->sub_stripes);
-	else if (map->type & BTRFS_BLOCK_GROUP_RAID0)
-		data_stripe_length = div_u64(data_stripe_length, map->num_stripes);
-	else if (map->type & BTRFS_BLOCK_GROUP_RAID56_MASK) {
-		data_stripe_length = div_u64(data_stripe_length,
-					     nr_data_stripes(map));
+	/* For RAID5/6 adjust to a full IO stripe length */
+	if (map->type & BTRFS_BLOCK_GROUP_RAID56_MASK)
 		io_stripe_size = map->stripe_len * nr_data_stripes(map);
-	}
 
 	buf = kcalloc(map->num_stripes, sizeof(u64), GFP_NOFS);
 	if (!buf) {
@@ -1698,25 +1745,12 @@ static int exclude_super_stripes(struct btrfs_block_group *cache)
 			return ret;
 
 		while (nr--) {
-			u64 start, len;
-
-			if (logical[nr] > cache->start + cache->length)
-				continue;
-
-			if (logical[nr] + stripe_len <= cache->start)
-				continue;
-
-			start = logical[nr];
-			if (start < cache->start) {
-				start = cache->start;
-				len = (logical[nr] + stripe_len) - start;
-			} else {
-				len = min_t(u64, stripe_len,
-					    cache->start + cache->length - start);
-			}
+			u64 len = min_t(u64, stripe_len,
+				cache->start + cache->length - logical[nr]);
 
 			cache->bytes_super += len;
-			ret = btrfs_add_excluded_extent(fs_info, start, len);
+			ret = btrfs_add_excluded_extent(fs_info, logical[nr],
+							len);
 			if (ret) {
 				kfree(logical);
 				return ret;
@@ -1745,7 +1779,7 @@ static void link_block_group(struct btrfs_block_group *cache)
 }
 
 static struct btrfs_block_group *btrfs_create_block_group_cache(
-		struct btrfs_fs_info *fs_info, u64 start, u64 size)
+		struct btrfs_fs_info *fs_info, u64 start)
 {
 	struct btrfs_block_group *cache;
 
@@ -1761,7 +1795,6 @@ static struct btrfs_block_group *btrfs_create_block_group_cache(
 	}
 
 	cache->start = start;
-	cache->length = size;
 
 	cache->fs_info = fs_info;
 	cache->full_stripe_len = btrfs_full_stripe_len(fs_info, start);
@@ -1769,7 +1802,7 @@ static struct btrfs_block_group *btrfs_create_block_group_cache(
 
 	cache->discard_index = BTRFS_DISCARD_INDEX_UNUSED;
 
-	atomic_set(&cache->count, 1);
+	refcount_set(&cache->refs, 1);
 	spin_lock_init(&cache->lock);
 	init_rwsem(&cache->data_rwsem);
 	INIT_LIST_HEAD(&cache->list);
@@ -1780,7 +1813,7 @@ static struct btrfs_block_group *btrfs_create_block_group_cache(
 	INIT_LIST_HEAD(&cache->dirty_list);
 	INIT_LIST_HEAD(&cache->io_list);
 	btrfs_init_free_space_ctl(cache);
-	atomic_set(&cache->trimming, 0);
+	atomic_set(&cache->frozen, 0);
 	mutex_init(&cache->free_space_lock);
 	btrfs_init_full_stripe_locks_tree(&cache->full_stripe_locks_root);
 
@@ -1841,25 +1874,44 @@ static int check_chunk_block_group_mappings(struct btrfs_fs_info *fs_info)
 	return ret;
 }
 
+static int read_block_group_item(struct btrfs_block_group *cache,
+				 struct btrfs_path *path,
+				 const struct btrfs_key *key)
+{
+	struct extent_buffer *leaf = path->nodes[0];
+	struct btrfs_block_group_item bgi;
+	int slot = path->slots[0];
+
+	cache->length = key->offset;
+
+	read_extent_buffer(leaf, &bgi, btrfs_item_ptr_offset(leaf, slot),
+			   sizeof(bgi));
+	cache->used = btrfs_stack_block_group_used(&bgi);
+	cache->flags = btrfs_stack_block_group_flags(&bgi);
+
+	return 0;
+}
+
 static int read_one_block_group(struct btrfs_fs_info *info,
 				struct btrfs_path *path,
 				const struct btrfs_key *key,
 				int need_clear)
 {
-	struct extent_buffer *leaf = path->nodes[0];
 	struct btrfs_block_group *cache;
 	struct btrfs_space_info *space_info;
-	struct btrfs_block_group_item bgi;
 	const bool mixed = btrfs_fs_incompat(info, MIXED_GROUPS);
-	int slot = path->slots[0];
 	int ret;
 
 	ASSERT(key->type == BTRFS_BLOCK_GROUP_ITEM_KEY);
 
-	cache = btrfs_create_block_group_cache(info, key->objectid, key->offset);
+	cache = btrfs_create_block_group_cache(info, key->objectid);
 	if (!cache)
 		return -ENOMEM;
 
+	ret = read_block_group_item(cache, path, key);
+	if (ret < 0)
+		goto error;
+
 	if (need_clear) {
 		/*
 		 * When we mount with old space cache, we need to
@@ -1874,10 +1926,6 @@ static int read_one_block_group(struct btrfs_fs_info *info,
 		if (btrfs_test_opt(info, SPACE_CACHE))
 			cache->disk_cache_state = BTRFS_DC_CLEAR;
 	}
-	read_extent_buffer(leaf, &bgi, btrfs_item_ptr_offset(leaf, slot),
-			   sizeof(bgi));
-	cache->used = btrfs_stack_block_group_used(&bgi);
-	cache->flags = btrfs_stack_block_group_flags(&bgi);
 	if (!mixed && ((cache->flags & BTRFS_BLOCK_GROUP_METADATA) &&
 	    (cache->flags & BTRFS_BLOCK_GROUP_DATA))) {
 			btrfs_err(info,
@@ -1905,15 +1953,15 @@ static int read_one_block_group(struct btrfs_fs_info *info,
 	 * are empty, and we can just add all the space in and be done with it.
 	 * This saves us _a_lot_ of time, particularly in the full case.
 	 */
-	if (key->offset == cache->used) {
+	if (cache->length == cache->used) {
 		cache->last_byte_to_unpin = (u64)-1;
 		cache->cached = BTRFS_CACHE_FINISHED;
 		btrfs_free_excluded_extents(cache);
 	} else if (cache->used == 0) {
 		cache->last_byte_to_unpin = (u64)-1;
 		cache->cached = BTRFS_CACHE_FINISHED;
-		add_new_free_space(cache, key->objectid,
-				   key->objectid + key->offset);
+		add_new_free_space(cache, cache->start,
+				   cache->start + cache->length);
 		btrfs_free_excluded_extents(cache);
 	}
 
@@ -1923,7 +1971,7 @@ static int read_one_block_group(struct btrfs_fs_info *info,
 		goto error;
 	}
 	trace_btrfs_add_block_group(info, cache, 0);
-	btrfs_update_space_info(info, cache->flags, key->offset,
+	btrfs_update_space_info(info, cache->flags, cache->length,
 				cache->used, cache->bytes_super, &space_info);
 
 	cache->space_info = space_info;
@@ -1962,7 +2010,6 @@ int btrfs_read_block_groups(struct btrfs_fs_info *info)
 	path = btrfs_alloc_path();
 	if (!path)
 		return -ENOMEM;
-	path->reada = READA_FORWARD;
 
 	cache_gen = btrfs_super_cache_generation(info->super_copy);
 	if (btrfs_test_opt(info, SPACE_CACHE) &&
@@ -1987,6 +2034,7 @@ int btrfs_read_block_groups(struct btrfs_fs_info *info)
 		btrfs_release_path(path);
 	}
 
+	rcu_read_lock();
 	list_for_each_entry_rcu(space_info, &info->space_info, list) {
 		if (!(btrfs_get_alloc_profile(info, space_info->flags) &
 		      (BTRFS_BLOCK_GROUP_RAID10 |
@@ -2007,6 +2055,7 @@ int btrfs_read_block_groups(struct btrfs_fs_info *info)
 				list)
 			inc_block_group_ro(cache, 1);
 	}
+	rcu_read_unlock();
 
 	btrfs_init_global_block_rsv(info);
 	ret = check_chunk_block_group_mappings(info);
@@ -2015,13 +2064,32 @@ error:
 	return ret;
 }
 
+static int insert_block_group_item(struct btrfs_trans_handle *trans,
+				   struct btrfs_block_group *block_group)
+{
+	struct btrfs_fs_info *fs_info = trans->fs_info;
+	struct btrfs_block_group_item bgi;
+	struct btrfs_root *root;
+	struct btrfs_key key;
+
+	spin_lock(&block_group->lock);
+	btrfs_set_stack_block_group_used(&bgi, block_group->used);
+	btrfs_set_stack_block_group_chunk_objectid(&bgi,
+				BTRFS_FIRST_CHUNK_TREE_OBJECTID);
+	btrfs_set_stack_block_group_flags(&bgi, block_group->flags);
+	key.objectid = block_group->start;
+	key.type = BTRFS_BLOCK_GROUP_ITEM_KEY;
+	key.offset = block_group->length;
+	spin_unlock(&block_group->lock);
+
+	root = fs_info->extent_root;
+	return btrfs_insert_item(trans, root, &key, &bgi, sizeof(bgi));
+}
+
 void btrfs_create_pending_block_groups(struct btrfs_trans_handle *trans)
 {
 	struct btrfs_fs_info *fs_info = trans->fs_info;
 	struct btrfs_block_group *block_group;
-	struct btrfs_root *extent_root = fs_info->extent_root;
-	struct btrfs_block_group_item item;
-	struct btrfs_key key;
 	int ret = 0;
 
 	if (!trans->can_flush_pending_bgs)
@@ -2034,21 +2102,11 @@ void btrfs_create_pending_block_groups(struct btrfs_trans_handle *trans)
 		if (ret)
 			goto next;
 
-		spin_lock(&block_group->lock);
-		btrfs_set_stack_block_group_used(&item, block_group->used);
-		btrfs_set_stack_block_group_chunk_objectid(&item,
-				BTRFS_FIRST_CHUNK_TREE_OBJECTID);
-		btrfs_set_stack_block_group_flags(&item, block_group->flags);
-		key.objectid = block_group->start;
-		key.type = BTRFS_BLOCK_GROUP_ITEM_KEY;
-		key.offset = block_group->length;
-		spin_unlock(&block_group->lock);
-
-		ret = btrfs_insert_item(trans, extent_root, &key, &item,
-					sizeof(item));
+		ret = insert_block_group_item(trans, block_group);
 		if (ret)
 			btrfs_abort_transaction(trans, ret);
-		ret = btrfs_finish_chunk_alloc(trans, key.objectid, key.offset);
+		ret = btrfs_finish_chunk_alloc(trans, block_group->start,
+					block_group->length);
 		if (ret)
 			btrfs_abort_transaction(trans, ret);
 		add_block_group_free_space(trans, block_group);
@@ -2069,10 +2127,11 @@ int btrfs_make_block_group(struct btrfs_trans_handle *trans, u64 bytes_used,
 
 	btrfs_set_log_full_commit(trans);
 
-	cache = btrfs_create_block_group_cache(fs_info, chunk_offset, size);
+	cache = btrfs_create_block_group_cache(fs_info, chunk_offset);
 	if (!cache)
 		return -ENOMEM;
 
+	cache->length = size;
 	cache->used = bytes_used;
 	cache->flags = type;
 	cache->last_byte_to_unpin = (u64)-1;
@@ -2132,54 +2191,6 @@ int btrfs_make_block_group(struct btrfs_trans_handle *trans, u64 bytes_used,
 	return 0;
 }
 
-static u64 update_block_group_flags(struct btrfs_fs_info *fs_info, u64 flags)
-{
-	u64 num_devices;
-	u64 stripped;
-
-	/*
-	 * if restripe for this chunk_type is on pick target profile and
-	 * return, otherwise do the usual balance
-	 */
-	stripped = get_restripe_target(fs_info, flags);
-	if (stripped)
-		return extended_to_chunk(stripped);
-
-	num_devices = fs_info->fs_devices->rw_devices;
-
-	stripped = BTRFS_BLOCK_GROUP_RAID0 | BTRFS_BLOCK_GROUP_RAID56_MASK |
-		BTRFS_BLOCK_GROUP_RAID1_MASK | BTRFS_BLOCK_GROUP_RAID10;
-
-	if (num_devices == 1) {
-		stripped |= BTRFS_BLOCK_GROUP_DUP;
-		stripped = flags & ~stripped;
-
-		/* turn raid0 into single device chunks */
-		if (flags & BTRFS_BLOCK_GROUP_RAID0)
-			return stripped;
-
-		/* turn mirroring into duplication */
-		if (flags & (BTRFS_BLOCK_GROUP_RAID1_MASK |
-			     BTRFS_BLOCK_GROUP_RAID10))
-			return stripped | BTRFS_BLOCK_GROUP_DUP;
-	} else {
-		/* they already had raid on here, just return */
-		if (flags & stripped)
-			return flags;
-
-		stripped |= BTRFS_BLOCK_GROUP_DUP;
-		stripped = flags & ~stripped;
-
-		/* switch duplicated blocks with raid1 */
-		if (flags & BTRFS_BLOCK_GROUP_DUP)
-			return stripped | BTRFS_BLOCK_GROUP_RAID1;
-
-		/* this is drive concat, leave it alone */
-	}
-
-	return flags;
-}
-
 /*
  * Mark one block group RO, can be called several times for the same block
  * group.
@@ -2225,7 +2236,7 @@ again:
 		 * If we are changing raid levels, try to allocate a
 		 * corresponding block group with the new raid level.
 		 */
-		alloc_flags = update_block_group_flags(fs_info, cache->flags);
+		alloc_flags = btrfs_get_alloc_profile(fs_info, cache->flags);
 		if (alloc_flags != cache->flags) {
 			ret = btrfs_chunk_alloc(trans, alloc_flags,
 						CHUNK_ALLOC_FORCE);
@@ -2252,7 +2263,7 @@ again:
 	ret = inc_block_group_ro(cache, 0);
 out:
 	if (cache->flags & BTRFS_BLOCK_GROUP_SYSTEM) {
-		alloc_flags = update_block_group_flags(fs_info, cache->flags);
+		alloc_flags = btrfs_get_alloc_profile(fs_info, cache->flags);
 		mutex_lock(&fs_info->chunk_mutex);
 		check_system_chunk(trans, alloc_flags);
 		mutex_unlock(&fs_info->chunk_mutex);
@@ -2283,13 +2294,13 @@ void btrfs_dec_block_group_ro(struct btrfs_block_group *cache)
 	spin_unlock(&sinfo->lock);
 }
 
-static int write_one_cache_group(struct btrfs_trans_handle *trans,
-				 struct btrfs_path *path,
-				 struct btrfs_block_group *cache)
+static int update_block_group_item(struct btrfs_trans_handle *trans,
+				   struct btrfs_path *path,
+				   struct btrfs_block_group *cache)
 {
 	struct btrfs_fs_info *fs_info = trans->fs_info;
 	int ret;
-	struct btrfs_root *extent_root = fs_info->extent_root;
+	struct btrfs_root *root = fs_info->extent_root;
 	unsigned long bi;
 	struct extent_buffer *leaf;
 	struct btrfs_block_group_item bgi;
@@ -2299,7 +2310,7 @@ static int write_one_cache_group(struct btrfs_trans_handle *trans,
 	key.type = BTRFS_BLOCK_GROUP_ITEM_KEY;
 	key.offset = cache->length;
 
-	ret = btrfs_search_slot(trans, extent_root, &key, path, 0, 1);
+	ret = btrfs_search_slot(trans, root, &key, path, 0, 1);
 	if (ret) {
 		if (ret > 0)
 			ret = -ENOENT;
@@ -2345,7 +2356,7 @@ static int cache_save_setup(struct btrfs_block_group *block_group,
 		return 0;
 	}
 
-	if (trans->aborted)
+	if (TRANS_ABORTED(trans))
 		return 0;
 again:
 	inode = lookup_free_space_inode(block_group, path);
@@ -2446,7 +2457,8 @@ again:
 	num_pages *= 16;
 	num_pages *= PAGE_SIZE;
 
-	ret = btrfs_check_data_free_space(inode, &data_reserved, 0, num_pages);
+	ret = btrfs_check_data_free_space(BTRFS_I(inode), &data_reserved, 0,
+					  num_pages);
 	if (ret)
 		goto out_put;
 
@@ -2611,7 +2623,7 @@ again:
 			}
 		}
 		if (!ret) {
-			ret = write_one_cache_group(trans, path, cache);
+			ret = update_block_group_item(trans, path, cache);
 			/*
 			 * Our block group might still be attached to the list
 			 * of new block groups in the transaction handle of some
@@ -2760,7 +2772,7 @@ int btrfs_write_dirty_block_groups(struct btrfs_trans_handle *trans)
 			}
 		}
 		if (!ret) {
-			ret = write_one_cache_group(trans, path, cache);
+			ret = update_block_group_item(trans, path, cache);
 			/*
 			 * One of the free space endio workers might have
 			 * created a new block group while updating a free space
@@ -2777,7 +2789,7 @@ int btrfs_write_dirty_block_groups(struct btrfs_trans_handle *trans)
 			if (ret == -ENOENT) {
 				wait_event(cur_trans->writer_wait,
 				   atomic_read(&cur_trans->num_writers) == 1);
-				ret = write_one_cache_group(trans, path, cache);
+				ret = update_block_group_item(trans, path, cache);
 			}
 			if (ret)
 				btrfs_abort_transaction(trans, ret);
@@ -2881,7 +2893,7 @@ int btrfs_update_block_group(struct btrfs_trans_handle *trans,
 					&cache->space_info->total_bytes_pinned,
 					num_bytes,
 					BTRFS_TOTAL_BYTES_PINNED_BATCH);
-			set_extent_dirty(info->pinned_extents,
+			set_extent_dirty(&trans->transaction->pinned_extents,
 					 bytenr, bytenr + num_bytes - 1,
 					 GFP_NOFS | __GFP_NOFAIL);
 		}
@@ -3317,7 +3329,7 @@ int btrfs_free_block_groups(struct btrfs_fs_info *info)
 		ASSERT(list_empty(&block_group->dirty_list));
 		ASSERT(list_empty(&block_group->io_list));
 		ASSERT(list_empty(&block_group->bg_list));
-		ASSERT(atomic_read(&block_group->count) == 1);
+		ASSERT(refcount_read(&block_group->refs) == 1);
 		btrfs_put_block_group(block_group);
 
 		spin_lock(&info->block_group_cache_lock);
@@ -3347,8 +3359,48 @@ int btrfs_free_block_groups(struct btrfs_fs_info *info)
 			    space_info->bytes_reserved > 0 ||
 			    space_info->bytes_may_use > 0))
 			btrfs_dump_space_info(info, space_info, 0, 0);
+		WARN_ON(space_info->reclaim_size > 0);
 		list_del(&space_info->list);
 		btrfs_sysfs_remove_space_info(space_info);
 	}
 	return 0;
 }
+
+void btrfs_freeze_block_group(struct btrfs_block_group *cache)
+{
+	atomic_inc(&cache->frozen);
+}
+
+void btrfs_unfreeze_block_group(struct btrfs_block_group *block_group)
+{
+	struct btrfs_fs_info *fs_info = block_group->fs_info;
+	struct extent_map_tree *em_tree;
+	struct extent_map *em;
+	bool cleanup;
+
+	spin_lock(&block_group->lock);
+	cleanup = (atomic_dec_and_test(&block_group->frozen) &&
+		   block_group->removed);
+	spin_unlock(&block_group->lock);
+
+	if (cleanup) {
+		em_tree = &fs_info->mapping_tree;
+		write_lock(&em_tree->lock);
+		em = lookup_extent_mapping(em_tree, block_group->start,
+					   1);
+		BUG_ON(!em); /* logic error, can't happen */
+		remove_extent_mapping(em_tree, em);
+		write_unlock(&em_tree->lock);
+
+		/* once for us and once for the tree */
+		free_extent_map(em);
+		free_extent_map(em);
+
+		/*
+		 * We may have left one free space entry and other possible
+		 * tasks trimming this block group have left 1 entry each one.
+		 * Free them if any.
+		 */
+		__btrfs_remove_free_space_cache(block_group->free_space_ctl);
+	}
+}