[OpenJDK 2D-Dev]  request for review: 7124245: [lcms] ColorConvertOp to color space CS_GRAY apparently converts orange to 244, 244, 0
james.graham at oracle.com
Thu Jan 3 01:16:31 UTC 2013
Why the move to source base of 1.5? I have been trying to keep as much
of J2DBench still running on 1.2 as I can so that we can do
apple-to-apple comparisons across releases of the same benchmark code.
Were the 1.5-isms something you could have written with 1.2 language
idioms? (I think I saw an enum in there, was that the reason to up the
language level to 1.5?)
Basically, for the ability to use enums, did we just give up the ability
to run the benchmark on 1.2 -> 1.4? Did we weigh those pros/cons?
On 12/21/2012 9:10 AM, Andrew Brygin wrote:
> Hello Jennifer and Phil,
> could you please review a fix for CR 8005402?
> Bug: http://bugs.sun.com/view_bug.do?bug_id=8005402
> Webrev: http://cr.openjdk.java.net/~bae/8005402/webrev.00/
> This fix extends J2DBench with benchmarks for color management:
> - ColorConvertOp performance for images and rasters.
> - rendering speed for images with non-sRGB color spaces.
> - performance of access to profile data.
> Please take a look.
More information about the 2d-dev