You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Currently propagate_sync function naively synchronizes the actor operation on the sensitive surfaces.
But we would want to synchronize it on the passive surfaces as well in a certain situation, so we can think of adding a config or option to propagator class. Another problem of the current implementation is that there still can be thread divergence in the synchronized actor when the tracks are on the sensitive surfaces but with different mask types. (e.g. one track is in pixel region and the other is in strip region, or one in end cap region and the other in barrel region)
I don't know how much gain we would get out of this, but still interesting to investigate
The text was updated successfully, but these errors were encountered:
See the introductory slides from here
Currently
propagate_sync
function naively synchronizes the actor operation on the sensitive surfaces.But we would want to synchronize it on the passive surfaces as well in a certain situation, so we can think of adding a config or option to
propagator
class. Another problem of the current implementation is that there still can be thread divergence in the synchronized actor when the tracks are on the sensitive surfaces but with different mask types. (e.g. one track is in pixel region and the other is in strip region, or one in end cap region and the other in barrel region)I don't know how much gain we would get out of this, but still interesting to investigate
The text was updated successfully, but these errors were encountered: