RFR (S?): 8249676: [REDO] G1 incorrectly limiting young gen size when using the reserve can result in repeated full gcs

Thomas Schatzl thomas.schatzl at oracle.com
Tue Aug 4 07:58:10 UTC 2020


Hi again,

On 24.07.20 12:29, Thomas Schatzl wrote:
> Hi all,
> 
>    I would like to ask for (re-)reviews to re-apply the changes for 
> "JDK-8244603: G1 incorrectly limiting young gen size when using the 
> reserve can result in repeated full gcs".
> 
> This has been backed out because it triggered the unrelated issue 
> JDK-8249192: MonitorInfo stores raw oops across safepoints.
> 
> There are no changes to the webrev from the original.
> 
> CR:
> https://bugs.openjdk.java.net/browse/JDK-8249676
> Webrev:
> http://cr.openjdk.java.net/~tschatzl/8249676/webrev/
> Testing:
> hs-tier1-8 with JDK-8249192
> 
> Original change:
> http://cr.openjdk.java.net/~tschatzl/8244603/webrev.1/
> 
> Thanks,
>    Thomas

   please hold off reviewing for now, some tier8 rerun (getting paranoid 
here ;)) again shows some unexplained crashes that I would like to 
investigate :(

Thanks,
   Thomas



More information about the hotspot-gc-dev mailing list