[PATCH, v3] Bringing <mutex> and <condition_variable> up to spec.

[Paolo Carlini]

> particular, I'm afraid the current set of changes would
> essentially broke any target not using gthr-posix (or gthr-single),
> because __gthread_t would remain undefined, still used in gthr.h.

Never mind the false alarm, the patch as-is only changes *comments* in gthr.h ;)

Anyway, my basic points stand: what about the other threads implementations? Who is going to review that? ...

Thanks,
Paolo.