RFR: 8343704: Bad GC parallelism with processing Cleaner queues [v7]

[Aleksey Shipilev]

On Thu, 14 Nov 2024 14:50:21 GMT, Axel Boldt-Christmas <aboldtch at openjdk.org> wrote:

> I was however thinking that it would only be pooled asymmetrically as some type of hystereses. So you pool when you remove a node (switch the head) and keep it far an arbitrary amount of removals. So it would only really waste memory for cleaners that have this behaviour that they keep adding and removing cleanable around a NODE_CAPACITY boundary.

I really do not want to make an improvement that is more complicated than it needs to be :) As it stands, current thing improves performance across the board, so chasing minor inefficiencies looks like venturing into diminishing returns territory. We can do this as the follow-up, if you want to explore that. I don't see clearly how this would work, and I would prefer to hold off more advanced heuristics in favor of simplicity here.

-------------

PR Comment: https://git.openjdk.org/jdk/pull/22043#issuecomment-2476972428