RFR: 8338103: Stabilize and open source a Swing OGL ButtonResizeTest

[Manukumar V S]

On Fri, 9 Aug 2024 18:45:18 GMT, Manukumar V S <mvs at openjdk.org> wrote:

> This PR creates a new test by stabilising and open sourcing a closed test.
> This test verifies that the OpenGL pipeline does not create artifacts with swing components after window is zoomed to maximum size and then resized back to normal.
> This test is run twice, with and without the flags "-Dsun.java2d.opengl=True -Dsun.java2d.opengl.fbobject=false" .
> 
> This is tested(15 times per platform) in all the available mach5 headful platforms and found to be stable.

> The test is failing on my windows machine with an error - _java.lang.RuntimeException: Button renderings are different after window resize, num of Diff Pixels=1576_ Images are uploaded for reference.
> 
> ![image1](https://private-user-images.githubusercontent.com/94159358/359016384-349f9300-25ba-4402-9668-d9e89321b23c.png?jwt=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.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.pgGc3C1vDtfnl_Zqx6wDFW7jkzsC-AjfGYdRTK3Op-s) ![image2](https://private-user-images.githubusercontent.com/94159358/359016448-c25fb95e-1f80-4762-9388-1b799a0184e8.png?jwt=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJpc
 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.8rwB9KPKg3Hpi_diZ7Rrm7BLbsVocU9j6Dn1F8j0mUM)

@TejeshR13 I have run it multiple times on different windows machines(Windows 2022, 10, 11) in mach5 and it passed everywhere and visibly I don't see any differences in the images attached by you. Can you try with the updated test which is attached to the bug(it will generate two diff images if the test fails), so that we will come to know what exactly is the difference between the images.

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

PR Comment: https://git.openjdk.org/jdk/pull/20532#issuecomment-2296041767