[Sbs_daq] [EXTERNAL] Bank numbering guideline for SBS

[Robert Michaels]

And that limit is because apparently the bank number is encoded in 16 bits.
I have a super simple analyzer (based on Dave Abbott's work) that prints out
data from a bank
adaq1: ~adaq/quickAna/quickAna.c  (see README)

banknum = (bankhead>>16)&0xffff; // bank number; where bankhead is a data buffer location

This is the way we usually think of banks.  (but I think there are other kinds of "banks" in CODA
and EVIO, it's a little confusing).  The banks I'm talking about are created in the readout with

 BANKOPEN(250,BT_UI4,blockLevel);
         // insert code here that does the readout ...
 BANKCLOSE;

The example above is for bank 250 which, following Alexandre's rule, is for the
FADC model 250.   A typical DAQ system has around ~10 crates with ~5 banks in them.
The bank numbers should be unique within a crate.

Bob
________________________________
From: Sbs_daq <sbs_daq-bounces at jlab.org> on behalf of David Flay <flay at jlab.org>
Sent: Tuesday, July 6, 2021 10:54 AM
To: Alexandre Camsonne <camsonne at jlab.org>; sbs_daq at jlab.org <sbs_daq at jlab.org>
Subject: Re: [Sbs_daq] [EXTERNAL] Bank numbering guideline for SBS

Hi Alex,

In the analyzer, ~/hana_decode/Decoder.h, we have:

namespace Decoder {
...
static const int MAXBANK = (1<<16)-1; // = 65535
...
}

Best regards,
David

----------------------------------------------------------------
David Flay, Ph. D.
Hall A/C Staff Scientist
Thomas Jefferson National Accelerator Facility
----------------------------------------------------------------


________________________________
From: Sbs_daq <sbs_daq-bounces at jlab.org> on behalf of Alexandre Camsonne <camsonne at jlab.org>
Sent: Tuesday, July 6, 2021 9:43 AM
To: sbs_daq at jlab.org <sbs_daq at jlab.org>
Subject: [Sbs_daq] [EXTERNAL] Bank numbering guideline for SBS

Hi all,

the bank number can be a random number, for simplification I would
like people use the same number for bank number as the model number in
the analyzer database, so by default the decoder would first try to
look for the model corresponding to the bank number and if it does not
seem to correspond it could use a model number from the cratemap.

So if you read out list has a random bank number, could you make sure
it is changed to match the model number so when we start taking data
it will be consistent for all the detector subsystems.

Does anyone know what is maximum limit on bank number value ?

Thank you,

Alexandre
_______________________________________________
Sbs_daq mailing list
Sbs_daq at jlab.org
https://mailman.jlab.org/mailman/listinfo/sbs_daq
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.jlab.org/pipermail/sbs_daq/attachments/20210706/91ea0137/attachment.html>