Search/Navigation:
Related:
SLAC/EPP/HPS Public
Jefferson Lab/Hall B/HPS Run Wiki
S30XL-LESA/LDMX
Basic info:
Before starting working make sure that the heatsink of the FEB is facing upward and the fan is pointing to it. If the FEB overheats it will reset itself (you will see all the leds halfway up).
Start working/developing:
Setup the general environment:
source /u1/hps/setup_env.sh
Start the conda environment:
conda activate rogue-hps-dev-ben
Run the software from this folder.
cd /u1/hps/server/heavy-photon-daq/software/scripts python SvtDaqGui.py --env SLACSM
The env command line is to tell which network configuration to load. The network configuration that gets loaded is stored in python/hps/constants.py
Check the FebLinkStatus. In the RCE test, you should see FebFpga[0] and check the Link[0] state. If False, Read the variables again (click Read). If now True, means we are talking to the Feb. After the link is established one can load the configuration.
Load the configuration for the system configuration (back end: ATCA (advanced telecommunication computing architecture) and front end (FEB+Hybrid+APVs)) . Click on the HpsSvtDaqRoot tab and load settings
/u1/hps/daq/heavy-photon-daq/software/config/rce-test.yml
There is a copy of this configuration file in /u1/hps/server/heavy-photon-daq/software/config/
In Variables tab navigate to the FebArray→FebCore→FebConfig and HybridPwrEn the 0 and 1 (the power supply will not be able to sustain 3 hybrids, but should be OK for 2).
After turning on the Hybrids, load the configuration again to be sure you are sending the right config to them.
Navigate to "Commands" tab and in PcieTiDtmArray click the sequence:
1) ApvClkAlign (to set all the APVs to the same phase)
2) ApvReset101. (reset and start to wait for a trigger signal)
With the hybrids configured and synced.
1) Go to HpsSvtDaqRoot tab→Browse and then set the name of the file ( it will set the output name with data and time automatically)
2) Cick on Open. You will see File Open → True when ready.
3) Set run Rate to something reasonable (10Hz should be OK)
4) Click on Run State and select "Running". When successfully running you'll see Run Count going up.
5) When done with local data taking, first set Run State to "Stopped" and Close the file.
Configuration files used to configure rogue.
The 2019 CODA (central DAQ software at jLab) configuration files for HPS DAQ are stored on jLab machines at
/usr/clas12/release/1.4.0/parms/trigger/HPS/Run2019/
The general configuration for the DAQ usually ends in .trg, the relevant block for SVT is under the #SVT Config block in one of those files.
This should point to something like ..../svt/svt_config.cnf, where the configuration should be pointed.
To access this information, one needs to ssh to the clonfarm machines. In particular, clonfarm2 and clonfarm3 are the SVT DAQ machines.
Setup of RCE bootloader for diskless booting: (based on this page)
ssh (or minicom via usb) to DTM (192.168.2.221)
minicom bay<bay>.<RCE> (bay is 0-3 <4 is the DTM> and RCE is 0 or 2 <NOT 1>)
From rdusr219 (with SDK env): cob_rce_reset 192.168.1.254/2/<bay>/<RCE>
Interrupt boot sequence by pressing any key in minicom
Apply the default u-boot environment:
zynq-uboot> env default -a
zynq-uboot> setenv modeboot nfsboot
zynq-uboot> setenv bootdelay 3
zynq-uboot> saveenv
zynq-uboot> reset
In order to down/upload to github, set up a tunnel between rhel6-64 and rdusr219. See Proto-DUNE: How to setup a temporary tunnel for github for a recipe.
The HPS RCEs boot in "diskless mode" via NFS. For info about this, see RCE Diskless Node (NFS).
RCE network names:
data DPMs: eth1_test_dpm<1-6>
control DPM: eth1_test_dpm7
DTM: eth1_test_dtm1
TO DO:
Upgrade sw to rogue, requires minor tweaks to fw (Ben)
Update to new config format used by rogue (Cameron and Omar)
Switch to JLab supported TI interface to use event builder supported at JLab and respin RTM (Ryan)
Build sw to take event blocks from data DPM fw and build EVIO blocks for CODA (JJ)
Install CODA at SLAC (Cameron and Omar)
Write EPICS sw (Cameron and Omar)