Btrfs: stop refusing the relocation of chunk 0

AFAICT chunk 0 is no longer special, and so it should be restriped just
like every other chunk.  One reason for this change is us refusing the
relocation can lead to filesystems that can only be mounted ro, and
never rw -- see the bugzilla [1] for details.  The other reason is that
device removal code is already doing this: it will happily relocate
chunk 0 is part of shrinking the device.

[1] https://bugzilla.kernel.org/show_bug.cgi?id=60594

Reported-by: Xavier Bassery <xavier@bartica.org>
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Signed-off-by: Chris Mason <chris.mason@fusionio.com>

fs/btrfs/volumes.c

@@ -2997,10 +2997,6 @@ static int __btrfs_balance(struct btrfs_fs_info *fs_info)
 		if (found_key.objectid != key.objectid)
 			break;
 
-		/* chunk zero is special */
-		if (found_key.offset == 0)
-			break;
-
 		chunk = btrfs_item_ptr(leaf, slot, struct btrfs_chunk);
 
 		if (!counting) {
@@ -3036,6 +3032,8 @@ static int __btrfs_balance(struct btrfs_fs_info *fs_info)
 			spin_unlock(&fs_info->balance_lock);
 		}
 loop:
+		if (found_key.offset == 0)
+			break;
 		key.offset = found_key.offset - 1;
 	}