[Ecls-list] Adding long long support to FFI.

[Julian Stecklina]

"Goffioul Michael" <goffioul at imec.be> writes:

> Looking at the configuration script (and following the ECL internal
> representation of objects), I think that cl_fixnum is typedef'ed to
> something that can fit into an address (void*). The question is then
> "does a LONG always fit into an address?". I'm not a platform expert,
> so I can't tell if such cases exist.

C implementors are free to make long a 64-bit integer even on 32-bit
platforms. I do not know an example of this, though. ;) 

What about using ptrdiff_t instead of long: "The type ptrdiff_t is the
signed integral type for the result of subtracting two pointers."

Regards,
-- 
Julian Stecklina

"I object to doing things that computers can do." - Olin Shivers