On Fri, 11 Sep 2026 17:34:48 GMT, Coleen Phillimore <[email protected]> wrote:

>> Remove the upcall to addClass during class loading.  The comment says it's 
>> only so GC can keep classes alive while the class loader is alive.  We have 
>> other ways to do that. There were some JVMTI tests in the past that failed 
>> without this vector but today seems to be only one test.  Maybe there's some 
>> code that has a dependency on this in heap walking.
>> Tested tier1-6
>> 
>> ---------
>> - [x] I confirm that I make this contribution in accordance with the 
>> [OpenJDK Interim AI Policy](https://openjdk.org/legal/ai).
>
> Coleen Phillimore has updated the pull request incrementally with one 
> additional commit since the last revision:
> 
>   Use Alan's suggested spec changes.

Just to confirm, I think the current proposal is now:

1. JVMTI_HEAP_REFERENCE_OTHER will be specified for both references and heap 
roots.
2. A reference for each (non-hidden) class defined by a class loader will be 
reported as an OTHER reference. We expect this should be transparent to most 
heap walking agents but there is some risk that a ref kind of OTHER will be 
treated as a heap root so the agent won't look at the referrer tag (or referrer 
class tag).
3. Legacy IterateOverReachableObjects is degraded, no ref will be synthesized.

-------------

PR Comment: https://git.openjdk.org/jdk/pull/32519#issuecomment-5653866263

Reply via email to