RFR: 8256302: releasing oopStorage when deflating allows for faster deleting

[Coleen Phillimore]

On Tue, 29 Nov 2022 21:15:44 GMT, Daniel D. Daugherty <dcubed at openjdk.org> wrote:

>> There's a call from the VMThread ObjectSynchronizer::do_final_audit_and_print_stats() now that can call this.
>
> The VM Thread can cause a deflation cycle to happen when logging is enabled.

The code prior to your change did chk_for_block_req, ie checked for a safepoint.  I was asking above why it did this in this loop since the code just above does a handshake. That is, why would it be necessary? This code is generally run in another thread.
Is the safepoint check the slow part or is the OopStorage release()?  Or both?  Do you have a test to point to that gets behind in freeing ObjectMonitors?

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

PR: https://git.openjdk.org/jdk/pull/11296