1 Introduction
This section intent to validate the software production.
2 Bugs to solve
- DIF does'nt run RNG test a second time (after a soft reset).
- Sending configuration sometime loose the last word and push it into the next configuration.
- Reading data show empty chips lines.
- Reading data show strange BCids.
3 ASU configuration
We have to validate a configuration file extracted from the /data/DB/ch_28_26_37_32_36.sqlite database available on all machine in Lyon.
This database is browsable using the Sqlite manager Firerfox's plugin.
Here we have such a file: acqilc@lyoac26:/data/online/config/sc_lyon_1.txt.
Gains are about:
4 Read-out chip's data
Tested on Scientific Linux 5.5:
- With one LDA and manual CCC
# ./exemple polcaldaq.conf -i1000 -t 100 -d 500000 -m4 -S2 -U10
- With one LDA and automatic CCC
# tools/utDevice polcaldaq.conf -i1000 -t 100 -d 500000 -m4 -S2 -U10
- With two LDA and automatic CCC
# tools/utDaq -i1000 -t 100 -d 500000 -m4 -S2 -U10
Note that we can capture the ethernet flow and re-inject it slower:
- Capture raw packets
# run/config polcaldaq.conf -i1000 -t 100 -d 500000 -m4 -S2 -U10
# tcpdump -xx -s 1024 -i eth0 ether host 5e:70:0c:d2:5c:d6 -w dump.raw
... ^C
- Run the libLDA without configuration:
# ./exemple polcaldaq.conf -p eth0 -n
- Re-inject the packets:
Note that eth0 cable must be plugged to a switch, but the LDA must be unplugged to that switch.
The
-Z parameter is the mac address of the host which run the libLDA.
# ./pcat -I eth0 -M 5e:70:0c:d2:5c:d6 -Z 5c:26:0a:28:73:0a -m replay -i dump.raw
|