[jdk21u-dev] RFR: 8348402: PerfDataManager stalls shutdown for 1ms

[Aleksey Shipilev]

On Wed, 6 Aug 2025 07:33:03 GMT, Goetz Lindenmaier <goetz at openjdk.org> wrote:

> I backport this for parity with 21.0.9-oracle.
> 
> src/hotspot/share/runtime/objectMonitor.cpp
> Resolved as comments in context are different.
> 
> src/hotspot/share/runtime/perfData.hpp
> Copyright
> 
> src/hotspot/share/runtime/synchronizer.cpp
> Resolved #include.

The backport itself looks good.
 
But I have a small doubt we should do it for JDK 21. Catching the critical section for FutileWakeups increment might be not worth it. In mainline, we have deleted these after replacing these PerfData events with related JFR events (https://bugs.openjdk.org/browse/JDK-8348829). 

But it is a minor risk, so I would not be against it as well.

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

Marked as reviewed by shade (Reviewer).

PR Review: https://git.openjdk.org/jdk21u-dev/pull/2058#pullrequestreview-3092408320