RFR: 8338103: Stabilize and open source a Swing OGL ButtonResizeTest [v4]

[Phil Race]

On Sat, 24 Aug 2024 06:30:44 GMT, Manukumar V S <mvs at openjdk.org> wrote:

>> test/jdk/javax/swing/JButton/SwingButtonResizeTestWithOpenGL.java line 54:
>> 
>>> 52:  * a JButton.  A file image of the component will be saved before and after
>>> 53:  * window resize. The test passes if both button images are the same.
>>> 54:  * @run main/othervm -Dsun.java2d.opengl=True -Dsun.java2d.opengl.fbobject=false SwingButtonResizeTestWithOpenGL
>> 
>> Does this really work well on Windows ?  Commonly OpenGL tests fail on Windows.
>> Ah I see in the bug report you say you tested on all platforms OK.
>> 
>> Also is there any reason we can't test this with additional pipelines on specific platforms ?
>> The behaviours being tested are Java behaviours so it ought to pass in other cases too.
>
> @prrace I’m not sure I understand what you meant by 'testing this with additional pipelines on specific platforms.' Are you suggesting that we should test this not only with -Dsun.java2d.opengl=True -Dsun.java2d.opengl.fbobject=false, but also with other parameters?

Pipeline. You can't easily know what will be chosen on a test system, so you need to do things like
@run main/othervm -Dsun.java2d.xrender=false
@run main/othervm -Dsun.java2d.xrender=true
@run main/othervm -Dsun.java2d.d3d=false
@run main/othervm -Dsun.java2d.d3d=true

In some cases, nothing will change but if the test is cheap and fast, nothing lost

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

PR Review Comment: https://git.openjdk.org/jdk/pull/20532#discussion_r1732176864