[OpenJDK 2D-Dev] [9]: RFR JDK-8147077, , IllegalArgumentException thrown by api/java_awt/Component/FlipBufferStrategy/indexTGF_General

[prasanta sadhukhan]

Thanks Jim for review. Since this a JCK bug on linux, I would like to 
push this fix.
I will test on other platforms and pipeline and if any issue found 
similar, I will create separate bug and fix on that id. Will that be ok?

Regards
Prasanta
On 1/23/2016 4:34 AM, Jim Graham wrote:
> That looks good.  While we are at it, do any of the other platforms or 
> pipelines need the same fix applied?
>
>             ...jim
>
> On 1/22/2016 4:01 AM, prasanta sadhukhan wrote:
>> Hi All,
>>
>> Please review a JCK regression fix for JDK-8140530
>> <https://bugs.openjdk.java.net/browse/JDK-8140530>
>>
>> Bug: https://bugs.openjdk.java.net/browse/JDK-8147077
>> webrev: http://cr.openjdk.java.net/~psadhukhan/8147077/webrev.00/
>>
>> JDK-8140530 <https://bugs.openjdk.java.net/browse/JDK-8140530> was fixed
>> to maintain consistency across platforms
>> so that VolatileImage of 0 width,height should result in IAE in linux as
>> it was already causing IAE in windows.
>> Also, non-volatile Image via
>> GraphicsConfiguration.createCompatibleImage(0,0) or a
>> BufferedImage(0,0,imagetype) also was resulting in IAE.
>>
>> But it was found that api/java_awt/Component/FlipBufferStrategy JCK test
>> was causing X11GraphicsConfig.createBackBuffer to create (0,0) 
>> volatileimage
>> resulting in IAE in linux. The same test was passing in windows as it
>> was found that D3DGraphicsConfig#createbackBuffer does not create 0
>> sized VolatileImage rather
>> it resize to (1,1) if it finds a 0 sized VolatileImage is requested.
>> The same fix is applied to X11GraphicsConfig too to make sure back
>> buffer is created with at least (1,1) volatileimage.
>>
>> The issue is marked noreg-jck so no regression test is added.
>>
>> Regards
>> Prasanta