state of libstdcxx_so_7-branch (was: rvalue references in libstdc++)

[Benjamin Kosnik <bkoz at redhat dot com>]

> I've been looking at updating the v7 branch to support "true" rvalue
> references. The code already there which implements it with a
> wrapper-hack will (purposefully) map very easily to the "true"
> semantics.

Hey Chris: delighted to see you tackling this.

Just a quick question for you, about the libstdcxx_so7-branch. It's kind 
of crufty now.

I think it might make more sense to re-branch off of current mainline 
(to libstdcxx_so_7_1-branch?), and then apply the valid bits from the 
old branch, and then start to work.

Thoughts?

Another other idea would be to use the cxx0x-branch, and convert that 
libstdc++ to be so7.

We are going to have to figure out how to do divergence well. I'm not 
quite sure what the best policy will be. Right now there are

gcc-4_1-branch
gcc-4_2-branch
gcc
gcc via the cxx0x-branch
libstdcxx_so_7-branch

branches that libstdc++ maintainers have to deal with. I'm just trying 
to figure out where all this fits in, and how to position libstdc++ for 
divergent work over the next 1-2 years.

Thoughts?

-benjamin