Btrfs: don't return ino to ino cache if inode item removal fails

In btrfs_evict_inode(), if btrfs_truncate_inode_items() fails, the inode
item will still be in the tree but we still return the ino to the ino
cache. That will blow up later when someone tries to allocate that ino,
so don't return it to the cache.

Fixes: 581bb050 ("Btrfs: Cache free inode numbers in memory")
Reviewed-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Omar Sandoval <osandov@fb.com>
Signed-off-by: David Sterba <dsterba@suse.com>

fs/btrfs/inode.c

@@ -5593,13 +5593,18 @@ void btrfs_evict_inode(struct inode *inode)
 		trans->block_rsv = rsv;
 
 		ret = btrfs_truncate_inode_items(trans, root, inode, 0, 0);
-		if (ret != -ENOSPC && ret != -EAGAIN)
+		if (ret) {
+			trans->block_rsv = &fs_info->trans_block_rsv;
+			btrfs_end_transaction(trans);
+			btrfs_btree_balance_dirty(fs_info);
+			if (ret != -ENOSPC && ret != -EAGAIN) {
+				btrfs_orphan_del(NULL, BTRFS_I(inode));
+				btrfs_free_block_rsv(fs_info, rsv);
+				goto no_delete;
+			}
+		} else {
 			break;
-
-		trans->block_rsv = &fs_info->trans_block_rsv;
-		btrfs_end_transaction(trans);
-		trans = NULL;
-		btrfs_btree_balance_dirty(fs_info);
+		}
 	}
 
 	btrfs_free_block_rsv(fs_info, rsv);
@@ -5608,12 +5613,8 @@ void btrfs_evict_inode(struct inode *inode)
 	 * Errors here aren't a big deal, it just means we leave orphan items
 	 * in the tree.  They will be cleaned up on the next mount.
 	 */
-	if (ret == 0) {
-		trans->block_rsv = root->orphan_block_rsv;
-		btrfs_orphan_del(trans, BTRFS_I(inode));
-	} else {
-		btrfs_orphan_del(NULL, BTRFS_I(inode));
-	}
+	trans->block_rsv = root->orphan_block_rsv;
+	btrfs_orphan_del(trans, BTRFS_I(inode));
 
 	trans->block_rsv = &fs_info->trans_block_rsv;
 	if (!(root == fs_info->tree_root ||