[Openmcl-devel] dppccl64 %SET-SINGLE-FLOAT

[Gary Byers]

Nope; that's just bogosity.

There's a compiler primitive - %SETF-SHORT-FLOAT - that tries to
destructively modify a heap-allocated short/single float; I'm 
not sure that that's even used, but since SINGLE-FLOATs are
immediate objects on PPC64, it doesn't make sense to talk
about destructivelby modifying one.  -That- should signal an
error (or something) if it's invoked on PPC64; %SET-SINGLE-FLOAT
is just a (foreign) memory operation, and should work more-or-less
the same on both platforms.

I'll try to check this in a little while.  After the compiler's
de-bogosified, the function %SET-SINGLE-FLOAT will need to be
recompiled.

On Tue, 12 Jul 2005, James Bielman wrote:

> (Found running the CFFI test suite):
>
> %SET-SINGLE-FLOAT appears to tail-call itself forever.  I noticed that
> there's a special case to treat this as a function call in nx1.lisp on
> :PPC64, is there a missing LAP function?
>
> [jamesjb at g5 193:~] openmcl64 -n                                | 19:00 05-07-12
> Welcome to OpenMCL Version (Alpha: DarwinPPC64) 0.14.3-050711!
> ? (ccl:%stack-block ((x 4))
>    (setf (ccl:%get-single-float x) 1.0))
> [...nada...]
>
> James
> _______________________________________________
> Openmcl-devel mailing list
> Openmcl-devel at clozure.com
> http://clozure.com/mailman/listinfo/openmcl-devel
>
>