RFR: 8287876: The recently de-problemlisted TestTitledBorderLeak test is unstable
Sergey Bylokhov
serb at openjdk.java.net
Tue Jun 7 11:40:11 UTC 2022
On Tue, 7 Jun 2022 04:15:49 GMT, Sergey Bylokhov <serb at openjdk.org> wrote:
> This is an update of the test which was de-problemlisted [here](https://github.com/openjdk/jdk/pull/8450).
>
> Looks like the initial root cause of the issue was related to the last disposed frame which for some reason was not deleted from the memory, for that issue, the [next](https://bugs.openjdk.org/browse/JDK-8287707) bug was reported.
> But that change also replaced the usage of weak references by the Runtime.getRuntime().freeMemory() which is not stable enough to verify the leak - its return values vary a lot.
>
> This change returns the usage of weak references and tests only one instance of TitledBorder which is enough to verify the old [bug](https://bugs.openjdk.org/browse/JDK-8204963).
I have found that the simpler version of the test can be used to verify that old bug. the patch is updated.
-------------
PR: https://git.openjdk.java.net/jdk/pull/9051
More information about the client-libs-dev
mailing list