>>>>> "Will" == Will  <[EMAIL PROTECTED]> writes:

    Will> This is consistent.  The reported list of traps is always
    Will> nil.  I tried setting the traps in a .cmucl-init file, but
    Will> that didn't work either.  The code works as expected in the
    Will> Linux version, so I'm starting to think that floating point
    Will> traps don't work on the G5 architecture (or powerpc in
    Will> general).  Is this the case?

Yes.  It's a known problem.  I don't know what the solution is.

Ray


Reply via email to