Re: RFR: 8386085: Livelock in AbstractQueuedSyncronizer.cleanQueue() when multiple threads do tryAcquire() with a short timeout and no permits available [v3]

[Kirill Shirokov]

On Wed, 10 Jun 2026 16:03:36 GMT, Viktor Klang <vkl...@openjdk.org> wrote:

>> I think one test suffices, adding documentation that if it ever fails, to 
>> further investigate the other classes.
>
> Thanks for the input, Doug. I have added some verbiage to that effect for the 
> regression test.

As a person that worked in dev & QA, I would still rather create an additional 
test for AQLS (if somebody would modify AQLS alone, we are covered). It looks 
like it should be easy: there is already 
`AbstractQueuedLongSynchronizerTest.Mutex` class with `tryAcquireNanos()`. Just 
my opinion.

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

PR Review Comment: https://git.openjdk.org/jdk/pull/31436#discussion_r3390748277