Hello Georg

On 03-Feb-99, Georg Gollmann wrote:
>  At 18:38 Uhr +0000 02.02.1999, Douglas Brebner wrote:
> > On 02-Feb-99, Markus Kohler wrote:
> >>  Ian Piumarta wrote:
> >>> 
> >>> ftp://alix.inria.fr/pub/squeak/unix/2.3 contains the usual stuff for:
> >>> 
> >>>         alpha-osf1-v4.0
> >>>         i386-linux-2.0.36       (linked against glibc-2.0.7)
> >>>         i386-linux-2.1.24       (linked against libc-5.4.38)
> >>>         sparc-solaris-2.5.1
> >>>         sparc-sunos-4.1.4
> >>> 
> >> 
> >>  Can anyone confirm that Ballon works on these platforms ?
> >>  I got 2.3 running on HP-UX now, but Ballon doesn't work.
> >>  In fact the primitives for Ballon are failing.
> >>  I really don't understand why because the failing primitive is
> >>  defined in my interpreter C-file. It is the primitive
> >>  primInitializeBuffer.
> > 
> > I had the same problem. My platform (NetBSD/i386 1.3.3) stored the VM
> > symbols with an underscore e.g.  _primInitializeBuffer
> > 
> > You need to change ioLoadExternalFunctionOfLengthFromModuleOfLength()
> > to add the underscore to the symbol name before dysym() if this is the
> > problem.
>  
>  In my case (HP/UX) the symbol not found is 'gePrimitiveInitializeBuffer'
>  although 'nm' lists it as external...

Does nm show 'gePrimitiveInitializeBuffer' or
            '_gePrimitiveInitializeBuffer' ?

The leading underscore *is* significant. Squeak looks for the former.
If your VM contains the latter then dlsym() will fail. *All* of my
external primitives failed because of this. :(

They work fine now though.

-- 
Regards
Douglas

E-Mail:  douglas+mail at fang.demon.co.uk