On Thu, 18 Jun 2026 12:57:48 GMT, Oli Gillespie <[email protected]> wrote:

>> @olivergillespie Since this PR doesn't address/resolve the issue it is 
>> linked to, perhaps this change should wait for the fix to the issue?
>
> @viktorklang-ora - thanks for lookin. This PR does address 
> https://bugs.openjdk.org/browse/JDK-8386372, the issue in the title ('add 
> test case'). It doesn't address https://bugs.openjdk.org/browse/JDK-8386278 
> (GC nepotism - mentioned in the description), but that is not a concurrency 
> bug, so it doesn't need to be fixed before the test case is added. The test 
> passes right now.
> 
> The link is that the test case would be useful to ensure a fix for 
> JDK-8386278 (performance issue) does not regress correctness (I created a 
> prototype fix for the nepotism issue, and it was functionally wrong, but no 
> tests caught it until I enabled this test).
> 
> Hope that makes sense.

@olivergillespie Ah, ok, if this PR is a prereq for addressing the other 
(nepotism) issue, would it not make more sense to address the test-change AND 
the "fix" in the same PR, so that if something needs to get backported, we 
don't risk having the test-change being left out from a backport?

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

PR Comment: https://git.openjdk.org/jdk/pull/31462#issuecomment-4742598361

Reply via email to