AltiVec [RFH] Overload resolution of vec_subs intrinsic fails

[Paolo Bonzini]

> > What I do not understand is if the treatment of opaque types in my patch
goes
> > against comptypes' definition of compatibility as in "same type-ness".
>
> Because opaque types have absolutely nothing to do with altivec.
> Altivec doesn't have opaque types.

That's also mixing issues.  As part of sorting out Altivec, is it ok to sort
out (the way I suggested it) the handling of *SPE* opaque types in comptypes?
It does not change behavior; the SPE tests pass.

> I also don't see how you can implement this overloading business with
> just opaque types and back-end magic.  But I am willing to review such
> patch.

I could try if somebody volunteers to help with testing.

Paolo