RFR: 8341715: PPC64: ObjectMonitor::_owner should be reset unconditionally in nmethod unlocking

[Richard Reingruber]

On Mon, 14 Oct 2024 13:14:52 GMT, Richard Reingruber <rrich at openjdk.org> wrote:

> This removes the `ObjectMonitor::_owner` check when a nmethod unlocks an inflated monitor on ppc64.
> Monitor operations by nmethods are guaranteed to be balanced (see JBS-item for a reference) therefore the check is redundant. Other platforms don't have it either.
> 
> I've removed the assertion that the unlocking thread owns the monitor again because it won't work with vthread monitor support in the loom repository.
> 
> The fix passed our CI testing with `LockingMode` set to `LM_LEGACY`
> Tier1-4 of hotspot and jdk. All of Langtools and jaxp. Renaissance Suite and SAP specific tests.
> Testing was done on the main platforms and also on Linux/PPC64le and AIX.
> 
> The test runtime/logging/MonitorInflationTest.java failed on all platforms. Apparently it has issues with `LM_LEGACY`.

Thanks!

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

PR Comment: https://git.openjdk.org/jdk/pull/21494#issuecomment-2418759878