Re: MPS: Bytecode stack

["Mattias Engdegård" <mattias.engdegard@gmail.com>]

6 maj 2024 kl. 15.35 skrev Gerd Möllmann <gerd.moellmann@gmail.com>:

> Or, wait a moment... I'm wondering about what happens when the client
> continues to exec byte code while we are scanning... The root is not
> like objects in MPS memory to which we get granted exclusive access. Or
> so I think ATM. But roots can have a barrier applied to them, I think.
> MPS_RM_PROT. I have to read a bit.

Yes, that's basically what I was wondering and the reason for the synthetic little example.

Essentially, what context is the scanning code running in? The mutator thread? If so, as an async signal handler, or in a more controlled context? If in a separate thread, does the mutator keep running? How did we get hold of the mutator stack and registers? And so on.