RFR: 8339730: Windows regression after removing ObjectMonitor Responsible

[David Holmes]

On Fri, 4 Oct 2024 13:19:49 GMT, Fredrik Bredberg <fbredberg at openjdk.org> wrote:

> In order to mitigate the regression seen on windows I added calls to enable hi res timer resolution around the call to WaitForSingleObject, when doing infinite parking. Got some good results, like these:
> 
> DaCapo23-tomcat-large 87.11%
> DaCapo23-spring-large 38.25%
> Renaissance-Reactors 10.34%
> DaCapo-h2-large 10.00%
> 
> Tested ok on tier1-7 on windows-x64

I've asked @fbredber to hold off on this while we try to run more performance checks. We need to find a Windows machine that always enables high-res timers (e.g. my laptop does this) and show that it 
a) has no regression with the Responsible thread removal; and
b) has no regression with this patch applied on top of that

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

PR Comment: https://git.openjdk.org/jdk/pull/21357#issuecomment-2415510891