Update Linux to v5.4.2

Change-Id: Idf6911045d9d382da2cfe01b1edff026404ac8fd
diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c
index 0526b6c..5cd42b6 100644
--- a/fs/btrfs/relocation.c
+++ b/fs/btrfs/relocation.c
@@ -20,6 +20,8 @@
 #include "inode-map.h"
 #include "qgroup.h"
 #include "print-tree.h"
+#include "delalloc-space.h"
+#include "block-group.h"
 
 /*
  * backref_node, mapping_node and tree_block start with this
@@ -162,6 +164,8 @@
 	struct mapping_tree reloc_root_tree;
 	/* list of reloc trees */
 	struct list_head reloc_roots;
+	/* list of subvolume trees that get relocated */
+	struct list_head dirty_subvol_roots;
 	/* size of metadata reservation for merging reloc trees */
 	u64 merging_rsv_size;
 	/* size of relocated tree nodes */
@@ -648,8 +652,8 @@
 					int level, u64 bytenr)
 {
 	struct backref_cache *cache = &rc->backref_cache;
-	struct btrfs_path *path1;
-	struct btrfs_path *path2;
+	struct btrfs_path *path1; /* For searching extent root */
+	struct btrfs_path *path2; /* For searching parent of TREE_BLOCK_REF */
 	struct extent_buffer *eb;
 	struct btrfs_root *root;
 	struct backref_node *cur;
@@ -662,7 +666,7 @@
 	struct btrfs_key key;
 	unsigned long end;
 	unsigned long ptr;
-	LIST_HEAD(list);
+	LIST_HEAD(list); /* Pending edge list, upper node needs to be checked */
 	LIST_HEAD(useless);
 	int cowonly;
 	int ret;
@@ -778,6 +782,10 @@
 				key.type != BTRFS_SHARED_BLOCK_REF_KEY);
 		}
 
+		/*
+		 * Parent node found and matches current inline ref, no need to
+		 * rebuild this node for this inline ref.
+		 */
 		if (exist &&
 		    ((key.type == BTRFS_TREE_BLOCK_REF_KEY &&
 		      exist->owner == key.offset) ||
@@ -787,11 +795,12 @@
 			goto next;
 		}
 
+		/* SHARED_BLOCK_REF means key.offset is the parent bytenr */
 		if (key.type == BTRFS_SHARED_BLOCK_REF_KEY) {
 			if (key.objectid == key.offset) {
 				/*
-				 * only root blocks of reloc trees use
-				 * backref of this type.
+				 * Only root blocks of reloc trees use backref
+				 * pointing to itself.
 				 */
 				root = find_reloc_root(rc, cur->bytenr);
 				ASSERT(root);
@@ -840,7 +849,11 @@
 			goto next;
 		}
 
-		/* key.type == BTRFS_TREE_BLOCK_REF_KEY */
+		/*
+		 * key.type == BTRFS_TREE_BLOCK_REF_KEY, inline ref offset
+		 * means the root objectid. We need to search the tree to get
+		 * its parent bytenr.
+		 */
 		root = read_fs_root(rc->extent_root->fs_info, key.offset);
 		if (IS_ERR(root)) {
 			err = PTR_ERR(root);
@@ -863,10 +876,7 @@
 
 		level = cur->level + 1;
 
-		/*
-		 * searching the tree to find upper level blocks
-		 * reference the block.
-		 */
+		/* Search the tree to find parent blocks referring the block. */
 		path2->search_commit_root = 1;
 		path2->skip_locking = 1;
 		path2->lowest_level = level;
@@ -884,7 +894,8 @@
 		    cur->bytenr) {
 			btrfs_err(root->fs_info,
 	"couldn't find block (%llu) (level %d) in tree (%llu) with key (%llu %u %llu)",
-				  cur->bytenr, level - 1, root->objectid,
+				  cur->bytenr, level - 1,
+				  root->root_key.objectid,
 				  node_key->objectid, node_key->type,
 				  node_key->offset);
 			err = -ENOENT;
@@ -892,6 +903,8 @@
 		}
 		lower = cur;
 		need_check = true;
+
+		/* Add all nodes and edges in the path */
 		for (; level < BTRFS_MAX_LEVEL; level++) {
 			if (!path2->nodes[level]) {
 				ASSERT(btrfs_root_bytenr(&root->root_item) ==
@@ -1422,6 +1435,13 @@
 	int clear_rsv = 0;
 	int ret;
 
+	/*
+	 * The subvolume has reloc tree but the swap is finished, no need to
+	 * create/update the dead reloc tree
+	 */
+	if (test_bit(BTRFS_ROOT_DEAD_RELOC_TREE, &root->state))
+		return 0;
+
 	if (root->reloc_root) {
 		reloc_root = root->reloc_root;
 		reloc_root->last_trans = trans->transid;
@@ -1458,15 +1478,17 @@
 	struct btrfs_root_item *root_item;
 	int ret;
 
-	if (!root->reloc_root)
+	if (test_bit(BTRFS_ROOT_DEAD_RELOC_TREE, &root->state) ||
+	    !root->reloc_root)
 		goto out;
 
 	reloc_root = root->reloc_root;
 	root_item = &reloc_root->root_item;
 
+	/* root->reloc_root will stay until current relocation finished */
 	if (fs_info->reloc_ctl->merge_reloc_tree &&
 	    btrfs_root_refs(root_item) == 0) {
-		root->reloc_root = NULL;
+		set_bit(BTRFS_ROOT_DEAD_RELOC_TREE, &root->state);
 		__del_reloc_root(reloc_root);
 	}
 
@@ -1630,6 +1652,8 @@
 
 	nritems = btrfs_header_nritems(leaf);
 	for (i = 0; i < nritems; i++) {
+		struct btrfs_ref ref = { 0 };
+
 		cond_resched();
 		btrfs_item_key_to_cpu(leaf, &key, i);
 		if (key.type != BTRFS_EXTENT_DATA_KEY)
@@ -1690,18 +1714,23 @@
 		dirty = 1;
 
 		key.offset -= btrfs_file_extent_offset(leaf, fi);
-		ret = btrfs_inc_extent_ref(trans, root, new_bytenr,
-					   num_bytes, parent,
-					   btrfs_header_owner(leaf),
-					   key.objectid, key.offset);
+		btrfs_init_generic_ref(&ref, BTRFS_ADD_DELAYED_REF, new_bytenr,
+				       num_bytes, parent);
+		ref.real_root = root->root_key.objectid;
+		btrfs_init_data_ref(&ref, btrfs_header_owner(leaf),
+				    key.objectid, key.offset);
+		ret = btrfs_inc_extent_ref(trans, &ref);
 		if (ret) {
 			btrfs_abort_transaction(trans, ret);
 			break;
 		}
 
-		ret = btrfs_free_extent(trans, root, bytenr, num_bytes,
-					parent, btrfs_header_owner(leaf),
-					key.objectid, key.offset);
+		btrfs_init_generic_ref(&ref, BTRFS_DROP_DELAYED_REF, bytenr,
+				       num_bytes, parent);
+		ref.real_root = root->root_key.objectid;
+		btrfs_init_data_ref(&ref, btrfs_header_owner(leaf),
+				    key.objectid, key.offset);
+		ret = btrfs_free_extent(trans, &ref);
 		if (ret) {
 			btrfs_abort_transaction(trans, ret);
 			break;
@@ -1735,7 +1764,7 @@
  * errors, a negative error number is returned.
  */
 static noinline_for_stack
-int replace_path(struct btrfs_trans_handle *trans,
+int replace_path(struct btrfs_trans_handle *trans, struct reloc_control *rc,
 		 struct btrfs_root *dest, struct btrfs_root *src,
 		 struct btrfs_path *path, struct btrfs_key *next_key,
 		 int lowest_level, int max_level)
@@ -1743,6 +1772,7 @@
 	struct btrfs_fs_info *fs_info = dest->fs_info;
 	struct extent_buffer *eb;
 	struct extent_buffer *parent;
+	struct btrfs_ref ref = { 0 };
 	struct btrfs_key key;
 	u64 old_bytenr;
 	u64 new_bytenr;
@@ -1764,7 +1794,7 @@
 	btrfs_node_key_to_cpu(path->nodes[lowest_level], &key, slot);
 
 	eb = btrfs_lock_root_node(dest);
-	btrfs_set_lock_blocking(eb);
+	btrfs_set_lock_blocking_write(eb);
 	level = btrfs_header_level(eb);
 
 	if (level < lowest_level) {
@@ -1777,7 +1807,7 @@
 		ret = btrfs_cow_block(trans, dest, eb, NULL, 0, &eb);
 		BUG_ON(ret);
 	}
-	btrfs_set_lock_blocking(eb);
+	btrfs_set_lock_blocking_write(eb);
 
 	if (next_key) {
 		next_key->objectid = (u64)-1;
@@ -1793,6 +1823,8 @@
 		BUG_ON(level < lowest_level);
 
 		ret = btrfs_bin_search(parent, &key, level, &slot);
+		if (ret < 0)
+			break;
 		if (ret && slot > 0)
 			slot--;
 
@@ -1843,7 +1875,7 @@
 						      slot, &eb);
 				BUG_ON(ret);
 			}
-			btrfs_set_lock_blocking(eb);
+			btrfs_set_lock_blocking_write(eb);
 
 			btrfs_tree_unlock(parent);
 			free_extent_buffer(parent);
@@ -1876,20 +1908,18 @@
 		 *    If not traced, we will leak data numbers
 		 * 2) Fs subtree
 		 *    If not traced, we will double count old data
-		 *    and tree block numbers, if current trans doesn't free
-		 *    data reloc tree inode.
+		 *
+		 * We don't scan the subtree right now, but only record
+		 * the swapped tree blocks.
+		 * The real subtree rescan is delayed until we have new
+		 * CoW on the subtree root node before transaction commit.
 		 */
-		ret = btrfs_qgroup_trace_subtree(trans, parent,
-				btrfs_header_generation(parent),
-				btrfs_header_level(parent));
+		ret = btrfs_qgroup_add_swapped_blocks(trans, dest,
+				rc->block_group, parent, slot,
+				path->nodes[level], path->slots[level],
+				last_snapshot);
 		if (ret < 0)
 			break;
-		ret = btrfs_qgroup_trace_subtree(trans, path->nodes[level],
-				btrfs_header_generation(path->nodes[level]),
-				btrfs_header_level(path->nodes[level]));
-		if (ret < 0)
-			break;
-
 		/*
 		 * swap blocks in fs tree and reloc tree.
 		 */
@@ -1903,23 +1933,31 @@
 					      path->slots[level], old_ptr_gen);
 		btrfs_mark_buffer_dirty(path->nodes[level]);
 
-		ret = btrfs_inc_extent_ref(trans, src, old_bytenr,
-					blocksize, path->nodes[level]->start,
-					src->root_key.objectid, level - 1, 0);
+		btrfs_init_generic_ref(&ref, BTRFS_ADD_DELAYED_REF, old_bytenr,
+				       blocksize, path->nodes[level]->start);
+		ref.skip_qgroup = true;
+		btrfs_init_tree_ref(&ref, level - 1, src->root_key.objectid);
+		ret = btrfs_inc_extent_ref(trans, &ref);
 		BUG_ON(ret);
-		ret = btrfs_inc_extent_ref(trans, dest, new_bytenr,
-					blocksize, 0, dest->root_key.objectid,
-					level - 1, 0);
+		btrfs_init_generic_ref(&ref, BTRFS_ADD_DELAYED_REF, new_bytenr,
+				       blocksize, 0);
+		ref.skip_qgroup = true;
+		btrfs_init_tree_ref(&ref, level - 1, dest->root_key.objectid);
+		ret = btrfs_inc_extent_ref(trans, &ref);
 		BUG_ON(ret);
 
-		ret = btrfs_free_extent(trans, src, new_bytenr, blocksize,
-					path->nodes[level]->start,
-					src->root_key.objectid, level - 1, 0);
+		btrfs_init_generic_ref(&ref, BTRFS_DROP_DELAYED_REF, new_bytenr,
+				       blocksize, path->nodes[level]->start);
+		btrfs_init_tree_ref(&ref, level - 1, src->root_key.objectid);
+		ref.skip_qgroup = true;
+		ret = btrfs_free_extent(trans, &ref);
 		BUG_ON(ret);
 
-		ret = btrfs_free_extent(trans, dest, old_bytenr, blocksize,
-					0, dest->root_key.objectid, level - 1,
-					0);
+		btrfs_init_generic_ref(&ref, BTRFS_DROP_DELAYED_REF, old_bytenr,
+				       blocksize, 0);
+		btrfs_init_tree_ref(&ref, level - 1, dest->root_key.objectid);
+		ref.skip_qgroup = true;
+		ret = btrfs_free_extent(trans, &ref);
 		BUG_ON(ret);
 
 		btrfs_unlock_up_safe(path, 0);
@@ -2117,6 +2155,66 @@
 }
 
 /*
+ * Insert current subvolume into reloc_control::dirty_subvol_roots
+ */
+static void insert_dirty_subvol(struct btrfs_trans_handle *trans,
+				struct reloc_control *rc,
+				struct btrfs_root *root)
+{
+	struct btrfs_root *reloc_root = root->reloc_root;
+	struct btrfs_root_item *reloc_root_item;
+
+	/* @root must be a subvolume tree root with a valid reloc tree */
+	ASSERT(root->root_key.objectid != BTRFS_TREE_RELOC_OBJECTID);
+	ASSERT(reloc_root);
+
+	reloc_root_item = &reloc_root->root_item;
+	memset(&reloc_root_item->drop_progress, 0,
+		sizeof(reloc_root_item->drop_progress));
+	reloc_root_item->drop_level = 0;
+	btrfs_set_root_refs(reloc_root_item, 0);
+	btrfs_update_reloc_root(trans, root);
+
+	if (list_empty(&root->reloc_dirty_list)) {
+		btrfs_grab_fs_root(root);
+		list_add_tail(&root->reloc_dirty_list, &rc->dirty_subvol_roots);
+	}
+}
+
+static int clean_dirty_subvols(struct reloc_control *rc)
+{
+	struct btrfs_root *root;
+	struct btrfs_root *next;
+	int ret = 0;
+	int ret2;
+
+	list_for_each_entry_safe(root, next, &rc->dirty_subvol_roots,
+				 reloc_dirty_list) {
+		if (root->root_key.objectid != BTRFS_TREE_RELOC_OBJECTID) {
+			/* Merged subvolume, cleanup its reloc root */
+			struct btrfs_root *reloc_root = root->reloc_root;
+
+			list_del_init(&root->reloc_dirty_list);
+			root->reloc_root = NULL;
+			if (reloc_root) {
+
+				ret2 = btrfs_drop_snapshot(reloc_root, NULL, 0, 1);
+				if (ret2 < 0 && !ret)
+					ret = ret2;
+			}
+			clear_bit(BTRFS_ROOT_DEAD_RELOC_TREE, &root->state);
+			btrfs_put_fs_root(root);
+		} else {
+			/* Orphan reloc tree, just clean it up */
+			ret2 = btrfs_drop_snapshot(root, NULL, 0, 1);
+			if (ret2 < 0 && !ret)
+				ret = ret2;
+		}
+	}
+	return ret;
+}
+
+/*
  * merge the relocated tree blocks in reloc tree with corresponding
  * fs tree.
  */
@@ -2124,7 +2222,6 @@
 					       struct btrfs_root *root)
 {
 	struct btrfs_fs_info *fs_info = rc->extent_root->fs_info;
-	LIST_HEAD(inode_list);
 	struct btrfs_key key;
 	struct btrfs_key next_key;
 	struct btrfs_trans_handle *trans = NULL;
@@ -2205,7 +2302,7 @@
 		    btrfs_comp_cpu_keys(&next_key, &key) >= 0) {
 			ret = 0;
 		} else {
-			ret = replace_path(trans, root, reloc_root, path,
+			ret = replace_path(trans, rc, root, reloc_root, path,
 					   &next_key, level, max_level);
 		}
 		if (ret < 0) {
@@ -2255,13 +2352,8 @@
 out:
 	btrfs_free_path(path);
 
-	if (err == 0) {
-		memset(&root_item->drop_progress, 0,
-		       sizeof(root_item->drop_progress));
-		root_item->drop_level = 0;
-		btrfs_set_root_refs(root_item, 0);
-		btrfs_update_reloc_root(trans, root);
-	}
+	if (err == 0)
+		insert_dirty_subvol(trans, rc, root);
 
 	if (trans)
 		btrfs_end_transaction_throttle(trans);
@@ -2405,14 +2497,9 @@
 			}
 		} else {
 			list_del_init(&reloc_root->root_list);
-		}
-
-		ret = btrfs_drop_snapshot(reloc_root, rc->block_rsv, 0, 1);
-		if (ret < 0) {
-			if (list_empty(&reloc_root->root_list))
-				list_add_tail(&reloc_root->root_list,
-					      &reloc_roots);
-			goto out;
+			/* Don't forget to queue this reloc root for cleanup */
+			list_add_tail(&reloc_root->reloc_dirty_list,
+				      &rc->dirty_subvol_roots);
 		}
 	}
 
@@ -2627,7 +2714,7 @@
 		 * only one thread can access block_rsv at this point,
 		 * so we don't need hold lock to protect block_rsv.
 		 * we expand more reservation size here to allow enough
-		 * space for relocation and we will return eailer in
+		 * space for relocation and we will return earlier in
 		 * enospc case.
 		 */
 		rc->block_rsv->size = tmp + fs_info->nodesize *
@@ -2670,6 +2757,7 @@
 	rc->backref_cache.path[node->level] = node;
 	list_for_each_entry(edge, &node->upper, list[LOWER]) {
 		struct btrfs_key first_key;
+		struct btrfs_ref ref = { 0 };
 
 		cond_resched();
 
@@ -2681,6 +2769,10 @@
 			if (!lowest) {
 				ret = btrfs_bin_search(upper->eb, key,
 						       upper->level, &slot);
+				if (ret < 0) {
+					err = ret;
+					goto next;
+				}
 				BUG_ON(ret);
 				bytenr = btrfs_node_blockptr(upper->eb, slot);
 				if (node->eb->start == bytenr)
@@ -2716,6 +2808,10 @@
 		} else {
 			ret = btrfs_bin_search(upper->eb, key, upper->level,
 					       &slot);
+			if (ret < 0) {
+				err = ret;
+				goto next;
+			}
 			BUG_ON(ret);
 		}
 
@@ -2748,7 +2844,7 @@
 			goto next;
 		}
 		btrfs_tree_lock(eb);
-		btrfs_set_lock_blocking(eb);
+		btrfs_set_lock_blocking_write(eb);
 
 		if (!node->eb) {
 			ret = btrfs_cow_block(trans, root, eb, upper->eb,
@@ -2767,11 +2863,13 @@
 						      trans->transid);
 			btrfs_mark_buffer_dirty(upper->eb);
 
-			ret = btrfs_inc_extent_ref(trans, root,
-						node->eb->start, blocksize,
-						upper->eb->start,
-						btrfs_header_owner(upper->eb),
-						node->level, 0);
+			btrfs_init_generic_ref(&ref, BTRFS_ADD_DELAYED_REF,
+					       node->eb->start, blocksize,
+					       upper->eb->start);
+			ref.real_root = root->root_key.objectid;
+			btrfs_init_tree_ref(&ref, node->level,
+					    btrfs_header_owner(upper->eb));
+			ret = btrfs_inc_extent_ref(trans, &ref);
 			BUG_ON(ret);
 
 			ret = btrfs_drop_subtree(trans, root, eb, upper->eb);
@@ -2911,7 +3009,6 @@
 		free_extent_buffer(eb);
 		return -EIO;
 	}
-	WARN_ON(btrfs_header_level(eb) != block->level);
 	if (block->level == 0)
 		btrfs_item_key_to_cpu(eb, &block->key, 0);
 	else
@@ -2987,7 +3084,7 @@
 	struct backref_node *node;
 	struct btrfs_path *path;
 	struct tree_block *block;
-	struct rb_node *rb_node;
+	struct tree_block *next;
 	int ret;
 	int err = 0;
 
@@ -2997,29 +3094,23 @@
 		goto out_free_blocks;
 	}
 
-	rb_node = rb_first(blocks);
-	while (rb_node) {
-		block = rb_entry(rb_node, struct tree_block, rb_node);
+	/* Kick in readahead for tree blocks with missing keys */
+	rbtree_postorder_for_each_entry_safe(block, next, blocks, rb_node) {
 		if (!block->key_ready)
 			readahead_tree_block(fs_info, block->bytenr);
-		rb_node = rb_next(rb_node);
 	}
 
-	rb_node = rb_first(blocks);
-	while (rb_node) {
-		block = rb_entry(rb_node, struct tree_block, rb_node);
+	/* Get first keys */
+	rbtree_postorder_for_each_entry_safe(block, next, blocks, rb_node) {
 		if (!block->key_ready) {
 			err = get_tree_block_key(fs_info, block);
 			if (err)
 				goto out_free_path;
 		}
-		rb_node = rb_next(rb_node);
 	}
 
-	rb_node = rb_first(blocks);
-	while (rb_node) {
-		block = rb_entry(rb_node, struct tree_block, rb_node);
-
+	/* Do tree relocation */
+	rbtree_postorder_for_each_entry_safe(block, next, blocks, rb_node) {
 		node = build_backref_tree(rc, &block->key,
 					  block->level, block->bytenr);
 		if (IS_ERR(node)) {
@@ -3030,11 +3121,10 @@
 		ret = relocate_tree_block(trans, rc, node, &block->key,
 					  path);
 		if (ret < 0) {
-			if (ret != -EAGAIN || rb_node == rb_first(blocks))
+			if (ret != -EAGAIN || &block->rb_node == rb_first(blocks))
 				err = ret;
 			goto out;
 		}
-		rb_node = rb_next(rb_node);
 	}
 out:
 	err = finish_pending_nodes(trans, rc, path, err);
@@ -3187,6 +3277,8 @@
 			if (!page) {
 				btrfs_delalloc_release_metadata(BTRFS_I(inode),
 							PAGE_SIZE, true);
+				btrfs_delalloc_release_extents(BTRFS_I(inode),
+							PAGE_SIZE);
 				ret = -ENOMEM;
 				goto out;
 			}
@@ -3207,7 +3299,7 @@
 				btrfs_delalloc_release_metadata(BTRFS_I(inode),
 							PAGE_SIZE, true);
 				btrfs_delalloc_release_extents(BTRFS_I(inode),
-							       PAGE_SIZE, true);
+							       PAGE_SIZE);
 				ret = -EIO;
 				goto out;
 			}
@@ -3229,14 +3321,14 @@
 		}
 
 		ret = btrfs_set_extent_delalloc(inode, page_start, page_end, 0,
-						NULL, 0);
+						NULL);
 		if (ret) {
 			unlock_page(page);
 			put_page(page);
 			btrfs_delalloc_release_metadata(BTRFS_I(inode),
 							 PAGE_SIZE, true);
 			btrfs_delalloc_release_extents(BTRFS_I(inode),
-			                               PAGE_SIZE, true);
+			                               PAGE_SIZE);
 
 			clear_extent_bits(&BTRFS_I(inode)->io_tree,
 					  page_start, page_end,
@@ -3252,8 +3344,7 @@
 		put_page(page);
 
 		index++;
-		btrfs_delalloc_release_extents(BTRFS_I(inode), PAGE_SIZE,
-					       false);
+		btrfs_delalloc_release_extents(BTRFS_I(inode), PAGE_SIZE);
 		balance_dirty_pages_ratelimited(inode->i_mapping);
 		btrfs_throttle(fs_info);
 	}
@@ -4083,6 +4174,9 @@
 		goto out_free;
 	}
 	btrfs_commit_transaction(trans);
+	ret = clean_dirty_subvols(rc);
+	if (ret < 0 && !err)
+		err = ret;
 out_free:
 	btrfs_free_block_rsv(fs_info, rc->block_rsv);
 	btrfs_free_path(path);
@@ -4168,7 +4262,7 @@
 	return inode;
 }
 
-static struct reloc_control *alloc_reloc_control(void)
+static struct reloc_control *alloc_reloc_control(struct btrfs_fs_info *fs_info)
 {
 	struct reloc_control *rc;
 
@@ -4177,9 +4271,11 @@
 		return NULL;
 
 	INIT_LIST_HEAD(&rc->reloc_roots);
+	INIT_LIST_HEAD(&rc->dirty_subvol_roots);
 	backref_cache_init(&rc->backref_cache);
 	mapping_tree_init(&rc->reloc_root_tree);
-	extent_io_tree_init(&rc->processed_blocks, NULL);
+	extent_io_tree_init(fs_info, &rc->processed_blocks,
+			    IO_TREE_RELOC_BLOCKS, NULL);
 	return rc;
 }
 
@@ -4189,37 +4285,13 @@
 static void describe_relocation(struct btrfs_fs_info *fs_info,
 				struct btrfs_block_group_cache *block_group)
 {
-	char buf[128];		/* prefixed by a '|' that'll be dropped */
-	u64 flags = block_group->flags;
+	char buf[128] = {'\0'};
 
-	/* Shouldn't happen */
-	if (!flags) {
-		strcpy(buf, "|NONE");
-	} else {
-		char *bp = buf;
-
-#define DESCRIBE_FLAG(f, d) \
-		if (flags & BTRFS_BLOCK_GROUP_##f) { \
-			bp += snprintf(bp, buf - bp + sizeof(buf), "|%s", d); \
-			flags &= ~BTRFS_BLOCK_GROUP_##f; \
-		}
-		DESCRIBE_FLAG(DATA,     "data");
-		DESCRIBE_FLAG(SYSTEM,   "system");
-		DESCRIBE_FLAG(METADATA, "metadata");
-		DESCRIBE_FLAG(RAID0,    "raid0");
-		DESCRIBE_FLAG(RAID1,    "raid1");
-		DESCRIBE_FLAG(DUP,      "dup");
-		DESCRIBE_FLAG(RAID10,   "raid10");
-		DESCRIBE_FLAG(RAID5,    "raid5");
-		DESCRIBE_FLAG(RAID6,    "raid6");
-		if (flags)
-			snprintf(bp, buf - bp + sizeof(buf), "|0x%llx", flags);
-#undef DESCRIBE_FLAG
-	}
+	btrfs_describe_block_groups(block_group->flags, buf, sizeof(buf));
 
 	btrfs_info(fs_info,
 		   "relocating block group %llu flags %s",
-		   block_group->key.objectid, buf + 1);
+		   block_group->key.objectid, buf);
 }
 
 /*
@@ -4227,6 +4299,7 @@
  */
 int btrfs_relocate_block_group(struct btrfs_fs_info *fs_info, u64 group_start)
 {
+	struct btrfs_block_group_cache *bg;
 	struct btrfs_root *extent_root = fs_info->extent_root;
 	struct reloc_control *rc;
 	struct inode *inode;
@@ -4235,14 +4308,23 @@
 	int rw = 0;
 	int err = 0;
 
-	rc = alloc_reloc_control();
-	if (!rc)
+	bg = btrfs_lookup_block_group(fs_info, group_start);
+	if (!bg)
+		return -ENOENT;
+
+	if (btrfs_pinned_by_swapfile(fs_info, bg)) {
+		btrfs_put_block_group(bg);
+		return -ETXTBSY;
+	}
+
+	rc = alloc_reloc_control(fs_info);
+	if (!rc) {
+		btrfs_put_block_group(bg);
 		return -ENOMEM;
+	}
 
 	rc->extent_root = extent_root;
-
-	rc->block_group = btrfs_lookup_block_group(fs_info, group_start);
-	BUG_ON(!rc->block_group);
+	rc->block_group = bg;
 
 	ret = btrfs_inc_block_group_ro(rc->block_group);
 	if (ret) {
@@ -4257,7 +4339,7 @@
 		goto out;
 	}
 
-	inode = lookup_free_space_inode(fs_info, rc->block_group, path);
+	inode = lookup_free_space_inode(rc->block_group, path);
 	btrfs_free_path(path);
 
 	if (!IS_ERR(inode))
@@ -4289,27 +4371,36 @@
 		mutex_lock(&fs_info->cleaner_mutex);
 		ret = relocate_block_group(rc);
 		mutex_unlock(&fs_info->cleaner_mutex);
-		if (ret < 0) {
+		if (ret < 0)
 			err = ret;
-			goto out;
+
+		/*
+		 * We may have gotten ENOSPC after we already dirtied some
+		 * extents.  If writeout happens while we're relocating a
+		 * different block group we could end up hitting the
+		 * BUG_ON(rc->stage == UPDATE_DATA_PTRS) in
+		 * btrfs_reloc_cow_block.  Make sure we write everything out
+		 * properly so we don't trip over this problem, and then break
+		 * out of the loop if we hit an error.
+		 */
+		if (rc->stage == MOVE_DATA_EXTENTS && rc->found_file_extent) {
+			ret = btrfs_wait_ordered_range(rc->data_inode, 0,
+						       (u64)-1);
+			if (ret)
+				err = ret;
+			invalidate_mapping_pages(rc->data_inode->i_mapping,
+						 0, -1);
+			rc->stage = UPDATE_DATA_PTRS;
 		}
 
+		if (err < 0)
+			goto out;
+
 		if (rc->extents_found == 0)
 			break;
 
 		btrfs_info(fs_info, "found %llu extents", rc->extents_found);
 
-		if (rc->stage == MOVE_DATA_EXTENTS && rc->found_file_extent) {
-			ret = btrfs_wait_ordered_range(rc->data_inode, 0,
-						       (u64)-1);
-			if (ret) {
-				err = ret;
-				goto out;
-			}
-			invalidate_mapping_pages(rc->data_inode->i_mapping,
-						 0, -1);
-			rc->stage = UPDATE_DATA_PTRS;
-		}
 	}
 
 	WARN_ON(rc->block_group->pinned > 0);
@@ -4431,7 +4522,7 @@
 	if (list_empty(&reloc_roots))
 		goto out;
 
-	rc = alloc_reloc_control();
+	rc = alloc_reloc_control(fs_info);
 	if (!rc) {
 		err = -ENOMEM;
 		goto out;
@@ -4486,6 +4577,10 @@
 		goto out_free;
 	}
 	err = btrfs_commit_transaction(trans);
+
+	ret = clean_dirty_subvols(rc);
+	if (ret < 0 && !err)
+		err = ret;
 out_free:
 	kfree(rc);
 out:
@@ -4549,7 +4644,7 @@
 		new_bytenr = ordered->start + (sums->bytenr - disk_bytenr);
 		sums->bytenr = new_bytenr;
 
-		btrfs_add_ordered_sum(inode, ordered, sums);
+		btrfs_add_ordered_sum(ordered, sums);
 	}
 out:
 	btrfs_put_ordered_extent(ordered);
@@ -4622,14 +4717,12 @@
 void btrfs_reloc_pre_snapshot(struct btrfs_pending_snapshot *pending,
 			      u64 *bytes_to_reserve)
 {
-	struct btrfs_root *root;
-	struct reloc_control *rc;
+	struct btrfs_root *root = pending->root;
+	struct reloc_control *rc = root->fs_info->reloc_ctl;
 
-	root = pending->root;
-	if (!root->reloc_root)
+	if (!root->reloc_root || !rc)
 		return;
 
-	rc = root->fs_info->reloc_ctl;
 	if (!rc->merge_reloc_tree)
 		return;
 
@@ -4658,10 +4751,10 @@
 	struct btrfs_root *root = pending->root;
 	struct btrfs_root *reloc_root;
 	struct btrfs_root *new_root;
-	struct reloc_control *rc;
+	struct reloc_control *rc = root->fs_info->reloc_ctl;
 	int ret;
 
-	if (!root->reloc_root)
+	if (!root->reloc_root || !rc)
 		return 0;
 
 	rc = root->fs_info->reloc_ctl;
@@ -4670,7 +4763,7 @@
 	if (rc->merge_reloc_tree) {
 		ret = btrfs_block_rsv_migrate(&pending->block_rsv,
 					      rc->block_rsv,
-					      rc->nodes_relocated, 1);
+					      rc->nodes_relocated, true);
 		if (ret)
 			return ret;
 	}