[ ] better understanding of the fakes explosion for PU tracks.
[ ] start testing the Alpaka modules (first the pixels and the the strips) in the HLT menu. This is needed both for the Alpaka migration itself and to eventually test the pixel+strip tracks and see the effect on the hlt full tracks.
[ ] build an ESProducer to define the CA parameters (especially the one depending on the geometry given the copy them on device);
[ ] allow the CA to get in input the layers we want to use (and their relative cuts) instead of having them hardcoded. These would then go in the ESProducer too.
[ ] estimate the (single) parameter used by the broken line to take into account of the multiple scattering. Marco (Rovere) gave me the document on how they did this average density estimation for pixel. We need to see how to redo this for strip. This shouldn't affect efficiency. But for sure affects resolutions and the kinematics (NB we still don't select tracks on chi2 or pt).
Collecting here the current ongoing tasks.