[OpenJDK 2D-Dev] [9] Review Request: JDK-8158524: Test file added for VolatileImage -vs- BufferedImage output comparison

[Sergey Bylokhov]

Looks fine.

On 16.08.16 11:11, Prahalad Kumar Narayanan wrote:
> Hello Everyone
>
> This is a follow-up to my webrev.
>
> Thanks to Sergey for his feedback.
> I 've incorporated the feedback and changes are now available for review
>           Link: http:// http://cr.openjdk.java.net/~pnarayanan/8158524/webrev.02/
>
> Summary on changes from previous revision:
>        . '@key headful' has been added to the jtreg comments
>        . Swing UI components are invoked on EDT
>
> Kindly review the changes and provide your feedback at your convenience
>
> Thank you
> Have a good day
>
> Prahalad N.
>
>
> -----Original Message-----
> From: Sergey Bylokhov
> Sent: Friday, August 12, 2016 6:39 PM
> To: Prahalad Kumar Narayanan; Prasanta Sadhukhan; 2d-dev at openjdk.java.net
> Subject: Re: [OpenJDK 2D-Dev] [9] Review Request: JDK-8158524: Test file added for VolatileImage -vs- BufferedImage output comparison
>
> On 12.08.16 16:06, Sergey Bylokhov wrote:
>> Looks fine.
>> Please add @key heedful jtreg tag before the push.
>
> Ouch I just realize that the test use the Swing components on non-EDT.
>
>>
>> On 11.08.16 13:22, Prahalad Kumar Narayanan wrote:
>>> Hello Everyone
>>>
>>> First, Thanks to Prasanta for his feedback.
>>> I 've incorporated the feedback in the code and changes are available
>>> for review.
>>>       Review Link:
>>> http://cr.openjdk.java.net/~pnarayanan/8158524/webrev.01/
>>>
>>> Quick summary on changes from previous revision:
>>>       . Wild card imports have been replaced with specific imports
>>>       . @override annotation has been added to the paint() method
>>>       . RuntimeException is thrown instead of Error though both would
>>> report regression failure.
>>>       . Image loading has been corrected with proper filePath.
>>>       . With regard to image and licensing issue:
>>>                 . I checked the image properties. There is no
>>> 'author/copyright' information with this image.
>>>                 . The file matches when compared (binary) with the
>>> similar .gif image in test/java/awt/print/PrinterJob/
>>>
>>> Kindly review the changes at your convenience and share your views.
>>>
>>> Thank you
>>> Have a good day
>>>
>>> Prahalad N.
>>>
>>>
>>> -----Original Message-----
>>> From: Prasanta Sadhukhan
>>> Sent: Thursday, August 11, 2016 1:02 PM
>>> To: Prahalad Kumar Narayanan; 2d-dev at openjdk.java.net
>>> Cc: Philip Race; Sergey Bylokhov
>>> Subject: Re: [OpenJDK 2D-Dev] [9] Review Request: JDK-8158524: Test
>>> file added for VolatileImage -vs- BufferedImage output comparison
>>>
>>> You need to use @Override annotation for paint() method. Also, you
>>> can use specific imports rather than package imports.
>>> Also, I guess it's better to throw RuntimeException than Error for
>>> failure, not sure how jtreg will react to it.
>>> Lastly, I am not sure if you can use duke.gif image which maynot be
>>> licensed although we have similar duke.gif in java/awt/print/PrinterJob.
>>>
>>> Regards
>>> Prasanta
>>> On 8/11/2016 11:30 AM, Prahalad Kumar Narayanan wrote:
>>>> Hello Everyone
>>>>
>>>> Good day to you.
>>>>
>>>> I 'm planning to add a test file into jdk/test/ repository.
>>>> The webrev changes for the test file is presented herewith.
>>>>             Link:
>>>> http://cr.openjdk.java.net/~pnarayanan/8158524/webrev.00/
>>>>
>>>> Brief on the test & its objectives:
>>>>          . As we know, VolatileImage utilizes accelerated rendering
>>>> using GPU while BufferedImage utilizes software rendering
>>>>          . The objective of this test-case is to check whether the
>>>> rendered 2d primitives appear same on VolatileImage and BufferedImage
>>>>          . In rare cases, the rendering on VolatileImage might
>>>> differ by few pixels from BufferedImage due to D3D/OpenGL driver 's
>>>> handling of primitives
>>>>                   . Java2D pipelines for D3D/OpenGL APIs use fudge
>>>> factors to fine tune rendering calls so that output matches with
>>>> output of BufferedImage.
>>>>                   . Thus the test case will help in checking for
>>>> consistent rendering across pipelines and also in identifying
>>>> scenarios where pipelines need to be fine-tuned.
>>>>
>>>> Kindly review the test file and share your views at your convenience
>>>>
>>>> Thank you
>>>> Have a good day
>>>>
>>>> Prahalad N.
>>>
>>
>>
>
>
> --
> Best regards, Sergey.
>


-- 
Best regards, Sergey.