ben-wpan: production and test process

[Werner Almesberger]

The process for flashing of atusb and post-SMT testing of atben and
atusb is now in a state where I think it should survive real-life
use. The (quite simple) test scripts live under

http://projects.qi-hardware.com/index.php/p/ben-wpan/source/tree/master/prod/

They rely heavily on the user-space tools under

http://projects.qi-hardware.com/index.php/p/ben-wpan/source/tree/master/tools

The process setup, its execution, and fault analysis are documented
in

http://downloads.qi-hardware.com/people/werner/wpan/prod/index.html

Documentation sources:

http://projects.qi-hardware.com/index.php/p/ben-wpan/source/tree/master/prod/doc/

Note that this is engineering-level documentation. At a factory,
you'd want to something a lot simpler and more streamlined. Since
we don't know yet what gremlins will crawl out of the SMT line, I
kept the analysis high-level and didn't go too deep into the details
of repair work either. (Even though the boards are simple, a "how to
identify anything that could possibly go wrong in atben/atusb and how
to repair it" manual would be a fearsome tome.)

Full disclosure: the tests don't include a "boundary scan" (GPIO) yet,
nor do they include a LED test. LED and I/O lines get exercised by
other tests, so any issues should eventually make themselves known.
There are two lines that don't get fully tested (SLP_TR and IRQ), so
I may still add that GPIO test.

Next steps:

- review of the documentation for readability, consistency, and
  correctness would be welcome

- at least someone at tuxbrain should walk through the full process,
  to make sure everything works and there are no misunderstandings
  (the other owners of ben-wpan prototypes can also run the test
  process - all you need is one atben and one atusb)

And then it would be time to put it all to good use, i.e., when the
boards roll off the SMT line :-)

- Werner