Docker Swarm Scheduling Performance with node.id Constraint

[GBisi]

If I created a service with docker swarm specifying e.g. node.id==node42 as a constraint what would happen from the point of view of the scheduling process? The process would be performed entirely taking into account the constraint or, having specified a constraint that forces the choice to a single node, docker only takes care of verifying if the choice is admissible "skipping" then the whole scheduling process?