staging headers fix (was Re: libstdc++ problem)

[Gabriel Dos Reis]

Phil Edwards <pedwards@disaster.jaj.com> writes:

| Short term... sigh.  That patch fixes so much stuff I'd hate to revert
| the whole thing.

I'm not questioning the fact that it was intended to fix many things.
But reality is that, we've got three bootstrap failure reports (I'm
not counting mine on solaris).  Java folks rely on functional C++, so
failure isn't stopping just we, V3 hackers, but also other guys
relying on a functional C++.  It is not question not to have the patch
in, but we have to preserve the "bootstrapability".

|  Why don't we just force the headers to be copied rather
| than symlinked?  It's only a single line (patch below), it preserves all
| the other semantics we've been trying to achieve, it lets the autobuilders
| succeed, and it lets other people depending on this patch go forward.
| Everybody wins.  (At least until Monday when we try to find a real solution.)

As a short term solution, yes.  If it bootstraps then please check it in.

-- Gaby