Understood, thanks for the very clear answer!
Hi @graven sorry to jump in, I understand the general issue. These are geometry related variables, and it is virtually impossible that they will change whilst the algorithm is running, hence why it was originally implemented (in the Muon Monitor that runs at the pit) like so. Or do you mean that the sole operation of a condition update can mess up the way these are filled in when running multithreaded? (I ask because if so, I'll make sure to also update the other Muon Monitor)
Lorenzo Paolucci (f50cdafc) at 21 Mar 14:43
Merge branch 'master' into lopaoluc_MuonMonitoring
... and 678 more commits
We use std::arrays
to contain some of our Gaudi histograms (since we need to plot the same variable over different regions of the detector), as well as caching detector properties in them (again, over the different regions of the detector)
How dare you tell me what to do @lhcbdpa