On Sat, 10 Oct 2026 07:30:56 GMT, Yunbo Zhang <[email protected]> wrote:

>> Please review this small test fix.
>> 
>> With `-XX:ActiveProcessorCount=1`, `HeapDumpParallelTest` incorrectly 
>> expects serial dumping, even though it configures two GC workers. The heap 
>> dump succeeds, but the test fails because it rejects the parallel dump 
>> output.
>> 
>> **Fix:**
>> 
>> 1. Identify the target VM's GC from its startup log, including builds that 
>> select Serial GC ergonomically.
>> 2. Remove the processor-count check, since two workers can perform a 
>> parallel dump even on a single CPU.
>> 3. Request two threads explicitly for the compressed parallel dump case.
>> 
>> **Testing:**
>> 
>> - Passed with G1, Parallel, Serial, Epsilon, ZGC, and Shenandoah, each with 
>> `ActiveProcessorCount=1` and `2`.
>> - `serviceability/dcmd/gc`: 13 passed, 1 skipped, 0 failed.
>> 
>> 
>> 
>> ---------
>> - [x] I confirm that I make this contribution in accordance with the 
>> [OpenJDK Interim AI Policy](https://openjdk.org/legal/ai).
>
> Yunbo Zhang has updated the pull request incrementally with one additional 
> commit since the last revision:
> 
>   8384612: Test serviceability/dcmd/gc/HeapDumpParallelTest.java failed on a 
> single core machine

Thank you for fixing.

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

Marked as reviewed by lmesnik (Reviewer).

PR Review: https://git.openjdk.org/jdk/pull/33212#pullrequestreview-5498337549

Reply via email to