On Thu, 17 Nov 2022 15:51:50 GMT, Alexander Zvegintsev <[email protected]> 
wrote:

> This test needs more stabilization:
> 
> - The test can be easily hang by replacing <br> iteration on 
> `keycharHash.elements()` with `keycharHash.forEach((k, v) -> 
> punchCtrlKey(robot, v));` 
> I do not see any reason why we need a synchronized `Hashtable`, so switched 
> to `HashMap`
> Looks like it might be a reason of original test hanging.
> - `keyRelease`, `dispose` calls wrapped with finally block
> - Just to be safe added timeout `@run main/timeout=600` ( none of the CI test 
> runs took too long thought)
> 
> CI testing looks good on all platforms.

It was a deadlock on the table access, due to some internals of Hashtable this 
should cure it.
Although I wonder if that wouldn't be a bug in the collections classes ?

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

Marked as reviewed by prr (Reviewer).

PR: https://git.openjdk.org/jdk/pull/11214

Reply via email to