Upstream DAQ Tech meeting

Thursday 4 Nov 2021, 14:00 → 15:00 Europe/Zurich

 

https://cern.zoom.us/j/68022868314

 

1 Current sprint

dunedaq-2.8.2
- updates to readout for new data formats
- timing opmon issues ?

dunedaq-2.10.0
- TPG integration
- Consolidation & refactoring
- SNB writer update

Stretch goals :
- RC integration
- Ethernet readout

Speakers: Jim Brooke, Roland Sipos (CERN)

Current sprints

• dunedaq-2.8.2
  • nobody had anything to report
• dunedaq-2.10.0
  • TP integration
    • Timing issues with a register related to ZCU102 ?  Fixed by hand.
    • Successful FELIX build (1 link/SLR) that meets timing. But locks up after 1 sec, now debugging
    • Goal for 2.10 is for working readout (no crashes etc) in dunedaq with TP inclusive FW
    • Problems in SW when we have 3 TP hits in a frame. Related to default destructor, investigating.
  • SNB write (zero copy) and refactoring.
  • Ethernet readout
    • Roland & Florian to allocate some time in the near future for this.

 

Integration

• CERN NP
  • Timing and readout is working nicely for both CBs
• Bristol
  • Recording data in 2.8.0 in Bristol lab working as basis for validation tools
  • Xilinx U50 on order for testing ethernet readout with TPs
    • Also need 100Gb NIC - any suggestions for the card?

 

Round table

• Timing
  • MIB will commission ipBus control over front panel, then backplane
  • Then move onto test with bugfixed daughterboard
  • Planning when/how we will introduce uTCA crate (+MIB+FIB) at CERN NP, in early 2022.
  • Still outstanding FW tasks (GPS, redundancy)
  • Software on track.  Outstanding task is RC integration (JB will organise meeting).
• Readout FW
  • nothing to report
• Readout SW
  • nothing to report
• TP FW
  • WIB2 : broken dependencies in data reception fixed (KM), testing underway (KK). Once this is tested will move on to testing the new repacker, and then whole data router testing.
  • Config : found some bugs in simulation, and fixed them. Only working with 1 link for now, will try multiple links next.  Then integrate into ZCU102 project to examine resource use.
• TP SW
  • Stitching algorithm implemented (IH), needs validation. Might make 2.10 but if not, can be factorised and released later.
  • KK working on validation tools.  Need to decide in which repo this code belongs.

 

2 Integration

Integration status and upcoming tasks at:
- Neutrino Platform
- CERN DT-DI lab
- Bristol
- UPenn
- ICEBERG

Speakers: Jim Brooke, Roland Sipos (CERN)

3 Round table

Looking beyond the current sprint, at planned development, upcoming issues/challenges etc.

4 AOB