Skip to content

MM Digitization - Make the subtools thread safe

Johannes Junggeburth requested to merge jojungge/athena:ATLASRECTS-7101 into 22.0

Hi everybody,

as I've mentioned in !54871 (merged) the MM digitization is not really thread-safe. That's going to be changed in this MR plus few other things.

  • First commit is a complete clang-format of the package. Please ignore these changes and only take a look at the subsequent ones.
  • MM_StripsResponseSimulation: Outsource all data members that are configured during the initialization into a Config struct object. Which is a const data member to ensure that's never touched. Remove also all vectors that are used during the processing of the simulation by local Cache objects
  • MM_ElectronicsResponseSimulation repeat the same exercise here. Declare all methods to be const. Draw random numbers from CLHEPP instead of TF1.
  • MM_DigitizationTool remove the functionality to dump the debug TTree. It's no longer needed and deprecated. Outsource all variables that might change during the process to a function local cache

Addresses ATLASRECTS-7101/ATLASRECTS-7107

Edited by Johannes Junggeburth

Merge request reports