Turn calls of __traverse__ into no-ops for unsendable pyclass if on the wrong thread

[adamreichold]

Adds a "threadsafe" variant of PyCell::try_borrow which will fail instead of panicking if called on the wrong thread and use it in call_traverse to turn GC traversals of unsendable pyclasses into no-ops if on the wrong thread.

This can imply leaking the underlying resource if the originator thread has already exited so that the GC will never run there again, but it does avoid hard aborts as we cannot raise an exception from within call_traverse.

Closes #3688

[codspeed-hq]

CodSpeed Performance Report

Merging #3689 will not alter performance

_{Comparing unsendable-threadsafe-traverse (4dc6c16) with main (65f25d4)}

🎉 Hooray! pytest-codspeed just leveled up to 2.2.0!

A heads-up, this is a breaking change and it might affect your current performance baseline a bit. But here's the exciting part - it's packed with new, cool features and promises improved result stability 🥳!
Curious about what's new? Visit our releases page to delve into all the awesome details about this new version.

Summary

✅ 78 untouched benchmarks

[adamreichold]

@JRRudy1 Could you use a dependency specification like

pyo3 = { git = "https://github.com/PyO3/pyo3.git", branch = "unsendable-threadsafe-traverse" }

and verify that this would work for your use case? Thanks.

[davidhewitt]

Thanks, nice! Looks straightforward enough.

I wonder if there's a way to add a test for this. I suspect the most reasonable option, if we felt it worth it, would be to add a test which manually calls the traverse slot?

[adamreichold]

I wonder if there's a way to add a test for this. I suspect the most reasonable option, if we felt it worth it, would be to add a test which manually calls the traverse slot?

Will give it a try.

I also noticed that this should be mentioned in the guide...

[davidhewitt]

Looks great to me, thanks, great test 👍