RFR (S) 8038939: Some options related to RTM locking optimization works inconsistently

Filipp Zhinkin filipp.zhinkin at oracle.com
Wed Apr 9 20:45:12 UTC 2014


Hi Vladimir,

On 09.04.2014 01:39, Vladimir Kozlov wrote:
> http://cr.openjdk.java.net/~kvn/8038939/webrev
> https://bugs.openjdk.java.net/browse/JDK-8038939
>
> Switch UseRTMXendForLockBusy flag ON by default and change code to 
> address issues described in the bug report.
>
> Jump to inflated locking when inflated lock is detected without 
> starting a new transaction regardless to RTMXendForLockBusy value.
>
> Now it retries RTM locking on lock busy condition by default.
>
> Tested with the test from bug report.
>
> Filipp, could you run your other tests with new patch?
Yes, no issues found.

Thanks,
Filipp.
>
> Thanks,
> Vladimir



More information about the hotspot-compiler-dev mailing list