finalize() without drop()?

[arjennienhuis]

The only way I can find to "gracefully" stop another thread that is blocking on C_WaitForSlotEvent is to call C_Finalize.

Pkcs11::finalize() is a no-op and Pkcs11Impl::finalize() is only called on drop(). I cannot drop the library if it is being used in another thread.

Is it ok to let Pkcs11::finalize() call Pkcs11Impl::finalize()?

We also need to fix Pkcs11::is_initialized() but that is a separate issue.

[wiktor-k]

Hmm... Yeah, it looks problematic. I'm not the author of this code but I assume the intention is to allow multiple threads access through Arc to the Impl struct and finalize when all threads are done.

And your point is that you want to call finalize earlier? The problematic part is, the pkcs11 struct assumes that what it holds is an initialized object before finalization. Would it be possible for your code to get all threads to drop their references? That'd deterministically call finalize when the last reference is dropped.

[hug-dev]

The problematic part is, the pkcs11 struct assumes that what it holds is an initialized object before finalization.

Agreed, if we were to change Pkcs11::finalize to only take a reference like the other functions and call Pkcs11Impl::finalize it would mean that users would still be able to use the other PKCS11 function and get the NotInitialized error everytime. I guess this is not a big safety problem but reducing those kind of errors is maybe still very nice to have?

Tell us if you are still having issues for this and if there are no other ways we can maybe think of something!