[cpp-threads] Asynchronous Function Proposal

[Lawrence Crowl]

On 6/1/09, Peter Dimov <pdimov at mmltd.net> wrote:
> I think that an enum parameter may be better as a policy because
> it would allow the choice to be made at runtime.

I can certainly live with that.  It wasn't my first choice because
of the overhead, but we made the same decision in atomics and can
reasonably do so here.

> In principle, there is no need to remove is_ready; it can be made
> to always return false for the separate thread case until wait
> is called. A dedicated implementation can even make it work.

My intent was to produce something that could reasonably be
implemented in terms of the exisiting mechanisms.  Even in response
to national body comments, too much last-minute invention can
cause trouble.

-- 
Lawrence Crowl