• Main INDEX
  • Monthly INDEX
  • PREV
  • NEXT

    User name R. Michaels

    Log entry time 13:56:29 on April30,2003

    Entry number 99192

    Followups:

    keyword=Final flush of ring buffer implemented

    On the R-arm DAQ, I implemented a final flushing of the ring buffer scaler data
    for end-of-run. This appears as event type 100 near the end of the CODA file,
    and it has the same format as the ROC10/11 data, including the ring buffer.
    I only did this on R-arm, not L-arm yet (low priority).

    This solves the very minor problem that the ring buffer, which got flushed only
    every synch event, did not get flushed at end. It was typically 0.1 sec of data.

    However it was not really a problem because:
    1) The data gets flushed by another routine at end of run, packaged by helicity
    and recorded in three places (see below); and
    2) The online packaging which this relies on is working ok.

    The three places where end-run scalers appear are:
    1) Scaler history file ~/scaler/scaler_history.dat
    2) halog
    3) The last event type 140 in CODA data file.

    It was verified that the new event 100 data agrees with the above three info's.