[OpenJDK 2D-Dev] request for review: 7188093

Vadim Pakhnushev vadim.pakhnushev at oracle.com
Thu Sep 20 20:36:20 UTC 2012

I copied original test .java and .html files, renamed and adjusted them 
for this case. Is it enough for jtreg?

Automated test will require taking screenshot and comparing results, I'm 
not sure if this mechanism works good.
If it is, I could develop automated test, Andrew mentioned that there is 
some helper class which provides this functionality.


On 21.09.2012 0:28, Phil Race wrote:
> Ideally it should go under this bug id but its not the end of the 
> world if we
> push it under another one.
> If you can provide it soon in jtreg form we can do it as part of this 
> fix.
> Ideally it should be automated.
> -phil.
> On 9/20/2012 7:51 AM, Vadim Pakhnushev wrote:
>> Phil,
>> I'm not sure about original test case, it uses cow.gif file. 
>> Moreover, it's not useful for this particular bug at all.
>> I can open my version of the test which clearly shows the bug.
>> I suppose I should post a webrev for this same CR with added test?
>> Thanks,
>> Vadim
>> On 18.09.2012 22:32, Phil Race wrote:
>>> Approved (2nd reviewer).
>>> Vadim: can you also see if we can 'open' the regression test that 
>>> identified this bug.
>>> -phi.
>>> On 9/18/2012 10:37 AM, Jim Graham wrote:
>>>> This looks good.  OpenGL and D3D punt under the same circumstances...
>>>>         ...jim
>>>> On 9/18/2012 1:16 AM, Clemens Eisserer wrote:
>>>>> Hello,
>>>>> Please review my fix for bug 7188093, located at:
>>>>> http://cr.openjdk.java.net/~ceisserer/7188093/webrev.01/
>>>>> Billinear interpolation is the highest quality interpolation 
>>>>> supported
>>>>> by xrender, so the xrender-pipeline has to fall back software 
>>>>> loops in
>>>>> case bicubic interpolation is requested.
>>>>> Nimbus only uses billinear interpolation, however I have no idea how
>>>>> much this affects other real-world applications out there.
>>>>> Thank you in advance, Clemens

More information about the 2d-dev mailing list