Click on the image for a postscript file.....
This object also handles the attachment of tokens to events and the return of tokens to the tokenQue on occurence of an aborted or completed event. Currently the number of tokens initially placed in the queue is an input parameter, so that up to 4095 tokens can be used during a run. When the tokenQue becomes empty a busy condition is set so that L0 triggers are suppressed until a token has been added back to the que. The methodology presently described in the L1/L2 Trigger Architecture document where tokens are added from the L1QueManager for accepeted L1 events has not been implemented (see Future Additions). The number of tokens in the que is recorded at intervals selected in the input. Statistical results and distributions of the number of tokens in the queue are also recorded.
The broadcast of results to the appropriate detectors is based on the trigActionWord, which again is generated in the tcu and updated at Levels 1-4 based on the analysis result (i.e. ABORT, ACCEPT). Cable delays of 220 ns have been included for the broadcast to the various detectors TPC,EMC, etc.
Upon accept of a L0 trigger the event is added to the trigTypeFifo, where the Level 1 Queue Manager checks for L0 triggered events.
Back to the Object Model J.P. Whitfield, Carnegie Mellon 4/20/95