Yi Lin

This seems related with full heap GC. If we force full heap GC, this bug will show up.

The problem in https://github.com/mmtk/mmtk-core/pull/429 was resulted by a mistake that introduced more full heap GCs. See https://github.com/mmtk/mmtk-core/pull/429#issuecomment-913264667. I suspect this issue is linked with full heap GCs.

The type `MaybeUninit` is a bit strange. I thought it was mimicking the Rust `MaybeUninit` and we may pass a Rust `MaybeUninit` value to the c++ code. But it seems...

> @qinsoon is it possible to trigger a benchmark github action here? like the ones we did to the mmtk-core PRs? Unfortunately we can't. Those are only set up for...

We probably do not have many assumptions in the code about stacks and how stacks are mapped to threads/mutators. One place @JunmingZhao42 noticed is https://github.com/mmtk/mmtk-core/blob/5f955bcf59b959df1d55e906968dfd14c18c3002/src/global_state.rs#L40 We use the number of...

> > Add a CI check to make sure migration guide is updated if we have API changes. > > This should give a warning instead of failing immediately. Some...

This is what the current form looks like for one API breaking PR, when we hide 'collapsible' sections.  Users need to be able to locate the information they need...

> So there needs to be an option to expand everything (or just the type and method names) and let the developer search for the method name they want to...

> I think we can use Admonish for the TL;DR part, instead. If you mean use admonish to highlight TLDR, that's fine. But don't make TLDR collapsible.

Broken link found: ``` ### Errors in docs/userguide/src/portingguide/howto/nogc.md * [404] [https://docs.mmtk.io/api/mmtk/vm/edge_shape/trait.MemorySlice.html](https://docs.mmtk.io/api/mmtk/vm/edge_shape/trait.MemorySlice.html) | Failed: Network error: Not Found ``` It should be from the last PR that renamed slot.