Skip to content

WIP: Test for using ReadHandle for the conditions

Hello @fwinkl ,

As you can see, this MR is much simpler. I am not applying yet the condition, but I am fetching and std::cout'ing it, so, this MR will never be non-WIP as it is. However, I am not sure about the scheduling of the algorithm (the conditions one, which actually produces a WriteHandle instead of a WriteCondHandle..) at every event (we have to measure the time of that running - most of the optimisations I made to the tool in the past are presently here, but I am never sure of the details). Also, in each loadCollection, only the instantiation of the read handler is taking more than 0.15 ms on average, which is completely non-acceptable for the L2 Calo egamma algorithm. Maybe we could push the offline friends to have a better solution. They will also gain from that. This is quite weird..

Cheers, Denis

Merge request reports