All Blocking Primitives need to interact with Safepoint Logic

[smarr]

For the blocking primitives, we need to decide one-by-one what to do with respect to safepoints.

There is always the potential for a deadlock, if they do not unregister before blocking.
Though, there may be performance implications that need to be considered.

This needs to be addressed so that #285 can be merged.

[smarr]

Got fixed with #288.