On Tue, 16 Dec 2025 07:09:55 GMT, SilenceZheng66 <[email protected]> wrote:
> Since my JVM's version is 25.191-b12 ... I thought it means JDK25... My apologies. > Now I just curious about is it a bug when people use configurations as > following and could raise full GC problems and heap can't expand. I think that behavior is definitely undesirable. Whether or not it's a bug can be subjective. I believe this behavior has been fixed/improved in later releases of JDK. At least the following two tickets are relevant. JDK-8328744: Parallel: Parallel GC throws OOM before heap is fully expanded Use max old-gen capacity to decide whether to start a "proactive" full-gc, instead of young-gc. JDK-8338977: Parallel: Improve heap resizing heuristics In `PSParallelCompact::summary_phase`, old-gen is expanded via `ParallelScavengeHeap::heap()->old_gen()->try_expand_till_size` even with `-XX:-UseAdaptiveSizePolicy`. I wonder if you can try your benchmark(s) with newer JDKs and check if the problem persists. ------------- PR Review Comment: https://git.openjdk.org/jdk/pull/25000#discussion_r2622829786
