RFR: 8213531: Test javax/swing/border/TestTitledBorderLeak.java fails [v2]

Sergey Bylokhov serb at openjdk.java.net
Thu Apr 28 22:43:39 UTC 2022


On Thu, 28 Apr 2022 13:34:31 GMT, Prasanta Sadhukhan <psadhukhan at openjdk.org> wrote:

>> Test was failing in linux citing `java.lang.RuntimeException: Expected Total TitledBorder to be freed : 10 Freed 9 `
>> As per the fix done in JDK-8204963 http://hg.openjdk.java.net/jdk/jdk/rev/cd7d2f9154fd
>> there was no platform specific code done for the fix and logs in `TitledBorder.installPropertyChangeListeners` shows it was called 10 times for the test but `CleanerFactory.cleaner().register` action was only called 9 times in linux causing it to fail.
>> 
>> Modified the test to not show the frame which cause the problem to go away and also it can still be used as 8204963 regression test as it still fails without the fix and pass with it. Modified test has passed in all platforms for several iterations.
>> 
>> Also removed the deprecated `System.runFinalization`
>
> Prasanta Sadhukhan has updated the pull request incrementally with one additional commit since the last revision:
> 
>   Test fix

> Modified the test to not show the frame which cause the problem to go away

Does it mean that setVisible(true) cause some memory leak since it prevents the window to be disposed?

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

PR: https://git.openjdk.java.net/jdk/pull/8450



More information about the client-libs-dev mailing list