[OpenJDK 2D-Dev] [10] Review Request JDK - 8164811 : [hidpi]Tests fail with OpenGL Rendering

Pankaj Bansal pankaj.b.bansal at oracle.com
Tue Nov 7 07:20:59 UTC 2017


Hi Sergey,

<< This version will creates back-buffer every time the paint/update is fail, but before the fix it was created only once before painting.
I am assuming you are talking about changes in UpdateWindow function in TranslucentPainter.
I think it is doing exactly what was happening before the fix. It called getBackBuffers and saved it in bb initially and then,  If the update fails, "done" will be false and it was calling getBackBuffers (true) to save the buffer in bb everytime. So it was calling getBackBuffers() everytime the update fails. Also the createBackBuffers function is creating backbuffer only once and then just returning it if the width and height remains same. So when I call update(getBackBuffers(false)), it will just return the backBuffers, not create it. So this change has added a call to getBackBuffers, but without it, I will have to add a check to see if the Image is BufferedImage or VolatileImage to decide to transform the Graphics or not.

Please let me know if you are talking about something else or please elaborate a bit more about the issue.

Regards,
Pankaj Bansal


-----Original Message-----
From: Sergey Bylokhov 
Sent: Tuesday, November 7, 2017 2:26 AM
To: Pankaj Bansal; Philip Race; Prem Balakrishnan
Cc: swing-dev at openjdk.java.net; 2d-dev at openjdk.java.net
Subject: Re: [OpenJDK 2D-Dev] [10] Review Request JDK - 8164811 : [hidpi]Tests fail with OpenGL Rendering

Hi, Pankaj.
This version will creates back-buffer every time the paint/update is fail, but before the fix it was created only once before painting.

On 06/11/2017 04:23, Pankaj Bansal wrote:
> I tested this on a machine which supports D3D pipeline. It fails with uiScale > 1.0 with -Dsun.java2d.d3d=false. With just scale being set, it passes as D3D is used as default pipeline on this machine.
> 
> Webrev:
> http://cr.openjdk.java.net/~pbansal/8164811/webrev.02/


--
Best regards, Sergey.


More information about the 2d-dev mailing list