Dr. J. Van Sckalkwyk (external)
Tue, 17 Jan 1995 07:55:43 SAT
> Note to JVS: this is no more "Low-level". Join/fork can be just
> *so different* according to the underlying hardware, and this is never
> a cheap operation; which does not mean it cannot be a basic one in the
> HLL, as the dynamic compiler can remove parallelism when too costly (or done
> inside CPU pipelines), or inline it to whatever is best suited.
As previously (and rather tediously, somewhat wheel-spinning-ly),
argued, we need a common "LLL" or "platform" that is machine-
independent. If our forkjoin is deep to this level then we should
probably not be talking about it as it is machine-specific. If
however, it is part of the common LLL (as I would hope) we should
clearly distinguish between _structural_ considerations and ones of
optimization! Bye, JVS.