RFR: 8322859: Parallel: Move transform_stack_chunk [v2]

[Thomas Schatzl]

On Tue, 2 Jan 2024 17:34:59 GMT, Albert Mingkun Yang <ayang at openjdk.org> wrote:

>> Simple moving the `transform_stack_chunk` call to match its counterpart in G1.
>
> Albert Mingkun Yang has updated the pull request incrementally with one additional commit since the last revision:
> 
>   review

src/hotspot/share/gc/parallel/psPromotionManager.inline.hpp line 256:

> 254: 
> 255:   // Parallel GC claims with a release - so other threads might access this object
> 256:   // after claiming and they should see the "completed" object.

Doesn't this comment describe the exact reason why the `ContinuationGCSupport::transform_stack_chunk` method has been placed here? Has other code been verified to not access the object (or at least these parts of the object) any more? I.e. this seems to be exactly the same issue as https://github.com/openjdk/jdk/pull/4751, requiring some evidence/description that this movement is okay (i.e. only the thread that does the CAS accesses the body of the oop).

At least this comment is misplaced if not wrong here now.

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

PR Review Comment: https://git.openjdk.org/jdk/pull/17227#discussion_r1445966321