Sergei Zharko authored 10 months ago and Felix Weiglhofer committed 9 months ago

- The auxiliary data objects from unpackers are stored in a temporary data structure AuxDigiUnpack (TODO: Dummy classes for DigiQa, execute the QA inside the Reco::RunUnpacker function)
- The option "--aux-data" is added to force saving the auxiliary data in unpackers (now is used only by std::Unpack)
- STS digi QA: histograms for channel vs. e-link are added for different modules

Dominik Smith authored 10 months ago and Dominik Smith committed 10 months ago

Added a third object to the return types of all unpackers in cbm::algo, which can store optional auxiliary data.

sweep is preceded by a row-wise sweep.

Added a preprocessor flag in trd::Hitfind to switch between row-parallel and module-parallel hit merging.

- Moved HitFactory2D class to dedicated file.
- Parallelized the entire chain "cluster finding -> hit finding -> hit merging"
  by row index. Merging is done in parallel between pairs of neighboring rows,
  using even-odd partitioning (fully supported for 2D, interface supplied for 1D).
- Cleaned up TRD reconstruction classes and fixed several warnings.

2) the mcbm_beam_2024_05_08_nickel setup is added to the CbmMcbmUtils (for RunID > 2917)

Sergei Zharko authored 11 months ago and Sergei Zharko committed 11 months ago

Change of parameter repository hash to support MR CbmSoft/cbmroot_parameter!180

Sergei Zharko authored 11 months ago and Sergey Gorbunov committed 11 months ago

- tracking interfaces for online STS and TRD
- installation rule for the macro/beamtime/mcbm2024 directory

- Add High-water Mark setting on the ZMQ output socket with default value of 1
- Add binary option to set the HWM + help info

- introduce a `ReadoutPars` struct similar to the one used on `tof` to add support for multiple unpack configs
- add new detector as "fake" back-side of the existing diamond detector to avoid need a new geometry
- switch readout parameters based on run ID, switch point hard-coded to run `2941` of 02/05/2024
- add needed hard-coded use of the 2022 version in the "algo offline" tasks (matches TOF/STS choice)