RFR: 8079267: [TEST_BUG] Test java/awt/Frame/MiscUndecorated/RepaintTest.java fails

[Alexander Zvegintsev]

On Thu, 28 Oct 2021 19:30:45 GMT, Sergey Bylokhov <serb at openjdk.org> wrote:

>> This test does not fail on Windows and does fail on Linux. 
>> 
>> After de-iconification we expect to see the button focused:
>> ![image](https://user-images.githubusercontent.com/77687766/139171852-54413dba-d58a-46d4-ab83-59d93868e287.png)
>> 
>> Actual look:
>> ![image](https://user-images.githubusercontent.com/77687766/139171775-80e64e58-c49d-489a-8fe0-fb0640fc3be9.png)
>> 
>> Button loses focus on `frame.to Front()` call, after its removal test passes. There is an [open issue](https://bugs.openjdk.java.net/browse/JDK-7156130) for this.
>> 
>> 
>> Some cleanup was also made. Testing is green for all platforms.
>> 
>> 
>> I am unable to reproduce [8266244 / macosx-aarch64](https://bugs.openjdk.java.net/browse/JDK-8266244), however it is not removed from the problem list due to its intermittent nature which needs more time to investigate.
>
> test/jdk/ProblemList.txt line 264:
> 
>> 262: java/awt/Component/SetEnabledPerformance/SetEnabledPerformance.java 8165863 macosx-all
>> 263: java/awt/Clipboard/HTMLTransferTest/HTMLTransferTest.java 8017454 macosx-all
>> 264: java/awt/Frame/MiscUndecorated/RepaintTest.java 8266244 macosx-aarch64
> 
> What is the reason for the failure on macOS aarch64?

It seems that it is the same that described by @prsadhuk in #6140 - cursor presence on screenshot:

![image](https://user-images.githubusercontent.com/77687766/139330827-ab54cfbe-2053-4f9a-91cd-75aee76c300e.png)
![image](https://user-images.githubusercontent.com/77687766/139330844-13d44fb0-e72f-465d-90dd-ce706505b4ee.png)

However it did fail for me only for original test before modifications, and didn't fail after, but I think it still need more investigation under a separate bug.

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

PR: https://git.openjdk.java.net/jdk/pull/6157