RFR: 8340166: [REDO] CDS: Trim down minimum GC region alignment

Aleksey Shipilev shade at openjdk.org
Mon Sep 16 11:10:08 UTC 2024


On Mon, 16 Sep 2024 09:10:45 GMT, Aleksey Shipilev <shade at openjdk.org> wrote:

> Now that [JDK-8338912](https://bugs.openjdk.org/browse/JDK-8338912) is in, we can try to redo [JDK-8337828](https://bugs.openjdk.org/browse/JDK-8337828). See the rationale for this change in [JDK-8337828](https://bugs.openjdk.org/browse/JDK-8337828) PR: https://github.com/openjdk/jdk/pull/20469. This change is identical to that PR.
> 
> Additional testing:
>  - [x] Linux x86_64 server fastdebug, `runtime/cds`
>  - [x] Linux x86_64 server fastdebug, `runtime/cds` `-XX:-UseCompressedOops` (failed during the original attempt to do this)
>  - [x] Linux x86_64 server fastdebug, `runtime/cds` `-XX:+UseShenandoahGC`
>  - [x] Linux x86_64 server fastdebug, `runtime/cds` `-XX:+UseShenandoahGC -XX:-UseCompressedOops`

I'll appreciate if anyone from Oracle could run this through their testing pipeline to avoid a second breakage. [The previous attempt broke Tier4](https://bugs.openjdk.org/browse/JDK-8338753) (however that is defined in Oracle testing).

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

PR Comment: https://git.openjdk.org/jdk/pull/21011#issuecomment-2352616954


More information about the hotspot-runtime-dev mailing list