Level 1 Calo Software Meeting at RAL on Tuesday 23 November 1999.
-----------------------------------------------------------------

This is a brief summary of what we discussed during this fairly
informal meeting.

Present: Murrough, Norman, Eric, Kithsiri, Scott, Tara, Reg, Bruce.



TTCvi Software
--------------

Kithsiri presented his latest UML diagrams for TTC software which have
changed since they were shown in Stockholm. After some discussion we
veered back towards his earlier ideas. We werent clear about the best
approach. Kithsiri will start to implement one scheme. To avoid being
dependent on one particular suite of VME access & register classes, he
should hide it in another class and encapsulate information about
VME base addresses and crates etc in yet another class. He will try to
use Software through Pictures (StP) to draw diagrams and generate the
skeleton code as another useful part of our learning exercises.


Configuration Database
----------------------

Eric showed a first draft of the list of items we will need to store
in the configuration database. Some omissions were corrected. He will
turn his slides into a document with the next week or two.


CVS Updates
-----------

Tara has updated our CVS repository with the latest versions of DAQ
and Diagnostic code. The DAQ C libraries are still missing - to be
rectified soon.


HDMC
----

Tara reported on his attempts to install and run the Heidelberg software.
He was able to install it on the HP systems and the departmental Linux
machines. But we are having many problems getting it to compile on our
LynxOS systems.
[PS After the meeting it was successfully installed on the new Linux
machine in the lab. We were also able fairly quickly to produce the code
required to access VME bus on our MVME167 LynxOS systems. However we are
still not able to compile it.... the LynxOS problems may be intractable.
Our 68K LynxOS 2.4.0 systems have an older compiler than the PPC based
LynxOS 2.4.0 systems and lack some of the STL support].


ROD & DAQ
---------

Norman raised the issue of planning for DAQ for our ROD and the integration
with DAQ -1 components. He is worried that we will want a working DAQ system
continuously and should consider adding DAQ -1 components to that, while
carrying on with work on a new DAQ system (which will not be in any working
state for some time).

He is also concerned that the HDMC software lacks some useful features that
we have grown accustomed to in the diagnostics: eg compact grouping of many
registers in one window with the ability to perform some actions on all of
them at once (eg zero, etc).

He would like us (me!) to consider adding block read/write network access
to the existing diagnostics so that we can continue using it without having
to run it on the painfully slow LynxOS systems.
[PS Measurements by Bruce & Kithsiri since the meeting suggest that the
performance would be adequate].

Questions (still) to be answered:
- how do we proceed (in the UK) with diagnostics & VME access?
- how do we proceed with DAQ development?
- how do we see HDMC and DAQ -1 convergence?


New Linux systems
-----------------

Bruce has installed Linux on the new PC in the lab. Some packages (eg ssh)
remain to be installed.

A PC has been ordered by Birmingham but has not yet arrived. The Bham HP
system has been unstable recently so installing Linux may have to wait
until their system manager sorts out the HP problems.


Next meeting
------------

Date not yet fixed. We should try to set up a video conference. To make
discussions with Cornelius et al most useful, we should prepare the ground
in advance: eg questions about HDMC, integration with DAQ -1, etc.


Actions
-------

Eric: produce config database document (first draft); distribute among
      ourselves (AT1SOFT list?); then to other level 1 colleagues
      
General: investigate HDMC on UK LynxOS systems vs UK diagnostics via network