RFR: 8372816: New test sun/security/provider/acvp/ML_DSA_Intrinsic_Test.java succeeds in case of error

[Martin Doerr]

On Tue, 2 Dec 2025 10:47:29 GMT, Arno Zeller <azeller at openjdk.org> wrote:

>> - Stop catching exception to make test able to fail
>> - Remove java-only test variation that cannot fail
>> - Reduce fuzz repeat count
>
> The change looks good, but you might consider using jdk.test.lib.RandomFactory. It allows to get and set a seed that is also printed at every run to make a failure easily reproducible.
> 
> And as the test now fails on aarch64 perhaps it should be excluded there and a bug should be opened - or is the behavior expected on aarch64 ?

> The integration of this change is unacceptable. The known aarch64 issue should have been resolved before integration, not deferred to another bug. If reviewers (@ArnoZeller @TheRealMDoerr) cannot verify other architectures, then the PR should not be approved unless the submitter is explicitly required to validate on all relevant platforms first.

That was a bit too quick. Sorry for that. We should add the test to the problem list if we can't fix it quickly.

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

PR Comment: https://git.openjdk.org/jdk/pull/28584#issuecomment-3611114149