So one successful round and w can stale forever? Seriously, I don't even know what to say, this is obviously not good.

This needs to happen for GetAvalancheVoteForBlock too.

I see, it is about to use a member. LGTM.

It doesn't looks like it is using any state of the object, so why should this be a member method of that object? That doesn't make sense.

This count cannot be simply retrieved using size() because it is a multimap like structure.

I don't understand what problem this is solving. First, this is using low free gizmo that seem completely unecessary - aren't VoteReccord owned by locks to begin with?

No, compare the address.

ProofComparator does complex computation, when it was before just comparing pointers. This solves a real problem, but the solution is not good.

I don't think this patch helps with anything.

This is good, I left a few comments on ways the code could be improved upon.

This is obviously unsound. How can this make consistent conclusion with the rest of the network?

As per our discussion, I think the main problem with this is that, while the existing code indeed looks funny, not in a good way, it's not clear what the correct structure actually is. I think we should put this on hold untill we actually have a a problem that would benefit from restructuring this code, then we'll have a better idea of what it should look like.

They are all adjective, I don't see any of this is consistent, and in fact, Invalidated imply that they were made invalid at some point in the past, which may not be the case at all (they might always have been invalid, for instance, and therefore, never invalidated).

There is pretty much the same code stamped 3 times in a row now. This is clearly an indication that it is time to stop piling up and go to the design board.

Also, the test plan doesn't test the change.

Tests are failing, back on your queue.

Fix typo

This is O(n^2)

Clearing my queue.

Clearing my queue.

Remove code moved by mistake.

Where is this tested?

The behavior changed, and yet no test changed. This indicates to me the current behavior is not tested appropriately. This is the first step that needs to be taken here.

The registration time is now stored for each UTXO, I don't think this really makes sense.