New C++ ABI: patches.

[Alexandre Petit-Bianco]

Jeff Sturm writes:
> Is the first vtable assignment bogus?

Yes. Thanks.

> Why not inline?  (You didn't mention isAssignableFrom in the
> ChangeLog.)

Because I don't really know what's happening -- I'm assuming that
turning inlining off in order not to have a undefined symbol is
wrong. But this patch is necessary for those who want to try their
luck at building and testing with the new ABI (I wanted to investigate
but I was running out of time.)

> Looks like this #ifndef block should just go away?

Yes, you're right.

I'm working on a new set of patches I'll post today. I'll probably
commit the C++ piece today too.

./A