GNU Clisp problem

I have been using this lisp implementation for a while, when I got this out of the blue:

*** - invalid byte sequence #xD7 #xED in CHARSET:UTF-8 conversion
Break 1 [6]> 

I am using clisp version 2.44.1 (64 bit) on karmic Ubuntu.
Any help please? Is it really a problem? Is there any other information I should post here?

Not all byte sequences are valid UTF-8 character codes. For example, some source files use a different encoding (maybe latin-1?) that triggers this error when clisp reads the file assuming utf-8. In many cases, it is sufficient to open the source file in a text editor and change the offending characters (often vowels with accents), or you can use a proper conversion tool.

Here's a couple useful links
http://techessence.info/node/60
http://www.unicodetools.com/

This happens before I load source files (manually). It happens after I start it, as well as when I try to load files. It is always the same bad bytes, whatever I load. I tried uninstalling and reinstalling clisp. I really do not know what the problem is, and clisp never complained before. What on earth is 0xD7 0xED? Latin-1 control characters? it is certainly not ASCII. Unless some files have been corrupted, where would there be strange foreign symbols on my system?

Does it happen if you start `clisp -norc`? Do you have a file named .clisprc.lisp or .clisprc.fas in your home directory?