[RFA:] PR37170, extern weak broken, like for gcc.dg/weak/weak-1.c

[Hans-Peter Nilsson]

> Date: Mon, 25 Aug 2008 08:23:27 +0100
> From: "Rafael Espindola" <espindola@google.com>

> > I have to retract this patch: it's not altogether wrong, but a
> > bit incomplet.
> 
> If it is just incomplete, shouldn't it be better to commit it and then
> fix the remaining parts?

Because I wasn't really sure.  Now that I have more confidence
in the current patch, I'm just waiting for test-results.  See
the PR.

> Is there something I can do to help?

Yes, there's a related change in revision 138310.  From the
mailing list conversation it looks like it's just papering over
the output_operand machinery not being able to "catch" all
symbol_refs.  This should be fixed by my "take 4" patch.  I'd
just like to verify that with the "original" pair of eyes.

But, all of this seems just a hack anyway, and the
assemble_external calls should be only in the "expand" pass; not
in front-ends nor in the back-end (varasm.c, final.c).  If you
could fix *that*, it'd be nice.

brgds, H-P