Bruno Kolenbrander

> very rarely use linked lists As does everyone else 😆 So this wouldn't be very useful. My main concern is that it doesn't do what people expect. If you...

Do we have metrics showing the performance differences between this and the current implementation? I'd be uncomfortable with this if there was not a noticeable performance benefit. > So if...

To support things like subinterpreters and "no gil" pyclasses must be `Sync`, so we might be requiring that at some point anyway.

To make subinterpreters and no-gil work we'd need to get rid of GilProtected too

First of all I very much dislike having a custom threadpool in pyo3, let alone one written with unsafe. Personally I don't feel qualified to review it and convince myself...

Finally I do want to thank you for all the effort you and the reviewers have put into this. I really appreciate the commitment whether or not we end up...

> I don't know what I can do for the nightly pipeline error 🤷‍♂️ > > ``` > error: non-local `impl` definition, they should be avoided as they go against...

You can also wait for someone to send a PR (probably me tomorrow) and rebase after it is merged.

> One straightforward option is to just add a default implementation of __hash__. This is a breaking change, however. Why is it a breaking change? Another option is to require...

I think part of this would be to have an optional `state: &State` argument in functions and methods that passes in some user defined (part of) the module state, so...