Right now we decode all known the answers in the packet even if we never need to look at them because we can't answer any of the questions in the packet

Decoding is the most expensive operation. In fact it's the most expensive operation on an mostly idle home assistant instance.

If we can defer decoding to only when we need to check for suppression we might save a lot of cpu time.

Need to see how much of the traffic is questions vs answers to see if it's worth the refactor