L1 Calo Software meeting at Birmingham on 1 March 2001.
-------------------------------------------------------

The agenda was:

Overview of software status and plans............Murrough

TDAQ news from recent ATLAS week.................Murrough

Slice and subsystem test configurations..........Norman
- who needs what and when?.......................Discussion
- CPUs: audit, local/remote booting

Software for the slice and subsystem tests
- Test vectors...................................Bill
- Run controllers................................Murrough
- Databases......................................Murrough
- Calibration/setup..............................Murrough
- Readout........................................Bruce
- Monitoring/simulation..........................Steve

HDMC related issues
- UK/HD/Mainz/Stk experience/improvements........?
- HDMC in the DAQ................................Bruce?
- Future support?................................NN

Experience with new VME PC at QMW................Murrough

AOB



Software Status and Plans
-------------------------

[slides]

Murrough started by giving an overview of the agenda and a brief summary
of our main activities in the recent past. He also presented a breakdown
of the software tasks involved in preparing for the slice tests. With some
very rough estimates of the time these may take and the people available
to do them, it is clear that we are short of effort.

Oliver stated that he will be joining the project officially in June
when he finishes his PhD. He expects to support HDMC.



News from ATLAS TDAQ Week
-------------------------

[slides]

Murrough then gave a short summary of the TDAQ aspects of the recent
ATLAS week, concentrating on some useful private meetings. From the
official meetings it is clear that most other TDAQ software projects
are also short of people.



Slice and Subsystem Test Configurations
---------------------------------------

[slides]

Norman presented his thoughts on this subject. We need to be thorough,
which implies work on test vectors. He was reluctant to present diagrams
of configurations as we will have to adapt to changing timescale and
should not be tempted to make the software rely on module appearing in
the ideal order.

To aid flexible testing, it would be useful if the software could
determine the connectivity of the system from the hardware itself,
rather than relying on databases to specify it.



Software for the Slice Tests:
-----------------------------

(A) Test Vectors
----------------

[slides]

Bill described recent work at Birmingham related to producing test
vectors. There have been two strands so far: Alan has used the offline
simulation to generate test vectors for testing the algorithm FPGA
code. Secondly Bill has written standalone programs to generate sets
of test vectors from a concise and flexible command language. So far
these programs have only considered CPM slice data and ROI packets
into and out of the ROD. They will need to be extended to cope with
other modules. Some addition intelligence to identify errors in test
results compared with the input test vectors is also desirable.



(B) Run Control, Databases, Calibrations
----------------------------------------

[slides]

Murrough presented his recent work on run control and related topics.
He outlined the ATLAS run control framework and list the requirements
of our own run controllers, displaying a proposed controller hierarchy
suitable for our system. This must provide the necessary synchronisation
(eg of TTC commands) with actions in the rest of the system.

In addition to the run controllers, we also need to specify several
databases (hardware configuration, trigger menu and calibrations). Here
work has only just started. Ralf mentioned that he also has some ideas
on the trigger menu.

Lastly Murrough mentioned that he has started listing all our calibration
and test procedures as a first step in ensuring they can be managed by
the run control: either as single runs or sets of runs. Here we need
more communication with the calorimeters to converge on preferred
calibration scenarios.



(C) Readout
-----------

[slides]

Bruce described his work and thoughts on readout - how he is managing
the ROD tests in the absence, so far, of any readout offering from the
CERN DAQ group.



(D) Monitoring and Simulation
-----------------------------

URL to Steves slides?[slides]
[slides]

Steve discussed issues related to monitoring. He noted that even the
slice test system is an order of magnitude larger than the previous
demonstrator setup. We need comparably better tools for presentation
and filtering of the extra information. Some components already exist
at least in initial versions, eg offline simulation, event dump, etc
but need to be integrated and improved.



HDMC Related Issues
-------------------

In addition to some bug fixes (eg in ModuleView) it is clear that we need
some new developments in HDMC in particular relating to its use within
the DAQ. We did have some discussions previously with Cornelius, but did
not reach any conclusion before Cornelius left.



Experience with Concurrent CPU
------------------------------

[slides]

Murrough briefly described his experience with the Concurrent VME PC at
QMW. Since the last joint meeting both Hannappel and VMELinux drivers
have been tried and incorporated into HDMC. The UK will buy three more
of these boards for the slice tests.



AOB
---

At some point the comprehensive use cases document prepared by the TGC
muon trigger group was mentioned. This is available
here.



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

The next joint software meeting will be on Thursday 28 June in Mainz.
Meanwhile, the next UK software meeting will be on Tuesday 20 March at RAL.