This change improves the performance of calling an event with 0 handlers by about 10% with no other changes.
Since we have to access the list eventually anyway, we can cut out some unnecessary work by returning the handlers from the cache directly, instead of fetching the HandlerList for no reason.
This also improves the performance of Event::hasHandlers() by about 40%, which is pretty significant (120 ns -> 80 ns).
this particular optimisation became possible thanks to changes in 4.19.
I observed that the allocation of Event objects and calling ->call() was costing us a significant percentage of the time taken in PlayerAuthInputPacket handlers. This change produces a measurable 2 microsecond reduction in overhead for PlayerAuthInputPacket handling when players are not moving (10.7 -> 8.7 microseconds). On a server with 200 players, this translates into a 1% reduction in CPU load for PlayerAuthInputPacket alone. It will also benefit other packets, but not to the extent that PlayerAuthInputPacket benefits.
This change significantly reduces the amount of work done by event handlers. Instead of traversing all of the priorities and event parent chain multiple times, we reduce event handlers down to a simple list, which doesn't require any logic to iterate over.
Previously, calling an event with lots of parents costed more than an event which directly descended from Event.
In addition, we had to do a lot of usually useless work to check all priorities, when in practice, only NORMAL will be used in almost all cases.
This change makes it more cost effective to implement the feature suggested by #5678; however, it will still require additional changes.
this gives a somewhat better overview of events, particularly if many plugins are subscribed to the same costly event (e.g. PlayerMoveEvent).
In addition, it allows us to see the frequency that events are occurring.