RFR: 8305915: java/awt/Frame/FrameLocation/FrameLocation.java fails with "The frame location is wrong!"

[Alexey Ivanov]

On Thu, 18 Sep 2025 14:51:46 GMT, Tejesh R <tr at openjdk.org> wrote:

> The test passed on CI machines with multiple test runs. Few stabilization fix has been made to make the test more robust.

> > What was the reason why the test failed?
> 
> Actually, the test hasn't failed in CI for mainline since 2023.
> 
> The exception that was thrown was "The frame location is wrong! Current: 250, 287; expected: 250, 250" was most likely addressed by [this fix](https://github.com/openjdk/jdk/commit/632a3c56e0626b4c4f79c8cb3d2ae312668d63fc#diff-2c406926284602be40cad012fcd1f92ca66e770427e760ced53c5e609f827ac4R1372-R1374) (resolved in 2023), in which the top inset is 37.
> 
> ```java
> case MUTTER_WM:
>     res = new Insets(37, 0, 0, 0);
>     break;
> ```

This makes me more confident that using `invokeAndWait` is wrong.

Would it be possible to temporarily revert this fix and verify if the test starts to fail again? If it's confirmed, this bug should be closed as duplicate of [JDK-8305825](https://bugs.openjdk.org/browse/JDK-8305825 "getBounds API returns wrong value resulting in multiple Regression Test Failures on Ubuntu 23.04").

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

PR Comment: https://git.openjdk.org/jdk/pull/27366#issuecomment-3311610248