Yi Lin

@petrochenkov Declaring `pub` doesn't help. I assume it is because `pub` does not mean they are accessible from native, thus their symbols can still be safely removed.

Hi @smarr We use ReBench to run Java micro benchmarks as one of the checks in our CI. I will update to the latest revision and let you know. It...

> # Things that shouldn't have been exported > The following are about plans: > > * `plan::Plan`: Contains `dyn` methods of a plan. A VM shouldn't directly interact with...

binding-refs OPENJDK_BINDING_REF=mimalloc-ms-support JIKESRVM_BINDING_REF=mimalloc-ms-support V8_BINDING_REF=update-pr-643

@wks @wenyuzhao I have requested reviews from you two. Before you do a comprehensive review, could you let me know if I should get this PR split into a few...

I don't think this is a good idea. MMTk is a garbage collector, and it is MMTk's responsibility to determine which objects are alive, and which are dead in a...

We had some discussion about this: * Instead of directly 'declaring an object dead', the API should be clearing the valid object bit. * The old 'alloc bit' is now...

> I am OK with doing both (1) clearing the VO bit, and (2) clearing the object header. If you have to do (2), does that mean (1) is not...

> The issue with (2) is that it is not sound w.r.t. the "valid object" semantics. Some random memory may return a false positive. The fact that (2) is still...

I suggested using a special policy for buffers. I think it does not conflict with Steve's description about the disjoint objects. It is indeed one way to implement disjoint objects....