[jdk21u-dev] RFR: 8272364: Parallel GC adaptive size policy may shrink the heap below MinHeapSize

[Daniel Hu]

On Wed, 12 Nov 2025 08:16:59 GMT, Goetz Lindenmaier <goetz at openjdk.org> wrote:

>> Clean backports. Fixes miscalculation of Parallal GC ergonomics; followup backports to fix related gtest (JDK-8333674 is trivial and backported so that JDK-8331675 is clean). New test passes with change, fails without. Changes pass GHA and internal Amazon pipelines: jtreg tier1-4 on platforms linux x64, aarch64, aarch32; alpine x64, aarch64; mac aarch64, x64; windows x64.
>
> Hi @cost0much 
> Why do you merge this into one PR?  If it is clean, it would be much better to have three PRs.
> You can do dependent PRs, i.e. for the second make a branch on top of the first, and in the PR 
> use the git branch of the first RP to compare to.
> 
> Also, please give a reason why you want to backport [JDK-8272364](https://bugs.openjdk.org/browse/JDK-8272364)

@GoeLin Ah, that's my bad. I assumed that given the subsequent test failures of [JDK-8272364](https://bugs.openjdk.org/browse/JDK-8272364), an atomic merge would be preferred. I can re-open with multiple PRs if that's better.

The justification for [JDK-8272364](https://bugs.openjdk.org/browse/JDK-8272364) is that the current behavior is incorrect, and the issue was significant enough to warrant a backport to 22. Moreover, the risk is comparatively low for hotspot backports, given the lack of bugtail (besides some failing tests). The only major risk I see is if 21 users are already reliant on the current incorrect behavior; if that's a main consideration, then this backport can be abandoned.

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

PR Comment: https://git.openjdk.org/jdk21u-dev/pull/2419#issuecomment-3524715204