<jmaier@midamerica.net> (John Maier) writes:

> So an app written with threading will show up as one process?
> 
> If I am understanding you correctly, The NetBSD implementation will be a LWP
> with kernel tie ins to keep calls that block [i.e. connect(), read()] from
> hanging the other threads?

That's approximately correct. 

> If this is true, then how might this be moved to a multiprocessor
> architecture model?  I realize that LWP threading has typically superior
> performance to HWP, where number crunching is *not* involved, but it won't
> take advantage of a multiprocessor system.

As Bill mentioned, there seem to be some nomenclature problems here.

Anyway, for the gory details on how this is going to work
(approximately), you should read:

"Scheduler Activations: Effective Kernel Support for the User-Level
Management of Parallelism". THOMAS E. ANDERSON, BRIAN N. BERSHAD,
EDWARD D. LAZOWSKA, and HENRY M. LEVY (University of Washington)
Proceedings of the 19th SOSP pp. 95-109 (1991)

http://www.acm.org/pubs/articles/journals/tocs/1992-10-1/p53-anderson/p53-anderson.pdf

        - Nathan