May 7, 2001
M. K. Brewer

DAQ Scheduler
=============

I made a major reorganization of John Monnier's DAQ scheduler program.
I broke the state machine into three parts.

A top level state machine now looks at what the DAQ scheduler is being
asked to do.  When a switch is made from quadrant dump mode to
scanning mode or vice versa, the top level state machine loads a new
program into the Altera FPGA and calls either do_quadrant() or
do_scans() to take data in either mode.

A second state machine in do_scans() manages the fringe data taking.
I fully implemented interrupt handling and DAC based fringe scanning
in do_scans().

At third state machine manages the data spooler.  This keys off of the
scanning state machine to store the data and flush the data buffers.

I didn't do much to the quadrant dumping routine other than fully
implement interrupt handling.  Ettore and I tested the new quadrant
dumping routine before I left and it works as before.

The fringe scanning routine has yet to be tested.  Ettore will do so
when the FPGA software is ready.

CB-coms, Serial Board, GPS
==========================

The serial board drivers now load into ppc0 on boot.  It is no longer
necessary to use serial_install().  The CB-Com server was tested and
works fine.  The beam shutters, however, are pretty unreliable.
Sometimes they open, sometimes they don't.  This seems to be a problem
with weak solenoids.  The GPS clock software works well.  I would
suggest purchasing an additional GPS 200 clock in case the current one
gets hit by lightning (They're relatively cheap).