[Openmcl-devel] misc.527 (PFDietz ANSI Test)

[David Steuber]

On Apr 24, 2005, at 2:10 AM, Gary Byers wrote:

> Thanks; this is fixed in bleeding-edge CVS.  (The function that
> computes the dimension(s) of the result bit array - and examines the
> dimensions of the source bit-arrays - in BIT-BOOLE and related
> functions should ignore fill-pointers.)

I did a cvs up and also fetched the 050420 dppcl image and new 
interface databases file from /pub/testing on the clozure.com ftp 
server.  I was able to recompile OpenMCL and can confirm the fix.

http://paste.lisp.org/display/7678#3

Did you use ffigen based on gcc 3.4 for the new interface databases or 
just rerun parse-ffi?  I read this in ChangeLog:

2005-04-20 00:53  gb

         * lib/db-io.lisp, library/parse-ffi.lisp: Changes to how some
         things are encoded in .cdb files; field offsets and widths are 
now
         provided explicitly.  This will require new .cdb files and is a
         little hard to bootstrap ...

Since my build script went smoothly, I can't confirm the hardness of 
bootstrapping.  I guess you must suffer alone ;-)