Hi Sergey,
That looks good to me.
Thanks,
Chris
On Jul 10, 2012, at 6:27 AM, Sergey Bylokhov wrote:
> Hi, Chris.
> Thanks for review. Can you please look into the new version:
> http://cr.openjdk.java.net/~serb/7181438/webrev.01/
> in this version OGLContext_SetExtraAlpha was removed and GL_ALP
Hi, Chris.
Thanks for review. Can you please look into the new version:
http://cr.openjdk.java.net/~serb/7181438/webrev.01/
in this version OGLContext_SetExtraAlpha was removed and GL_ALPHA_BIAS
changed to 1.0f.
I'll create additional CR to track changes from 7125723&7181438.
10.07.2012 03:16,
Hi Sergey,
> Probably OGLBlitSwToTexture assumes that both of sw and texture should be
> opaque or non opaque.
That is correct. OGLBlitSwToTexture is only used to upload a system memory
surface (a BufferedImage surface) into an OpenGL texture for the purposes of
managed images. So, prior to
Hi, Chris.
I am sure, that those changes should be reverted back or at least we
should rethink.
Note, that when I paint something from the buffered image to the
surface, I'll get correct image. Because OGLBlitSwToSurface, that is
used in this case, will block alpha for opaque surface. But whe
> P.S. Good to see some activity on the OGL pipeline. I don't get paid to
> worry about these things anymore though, so perhaps I should just crawl back
> in my hole now :)
But if you wanted to be I'm sure we could arrange it ^^
Hi Sergey,
Do you happen to know the intent behind this original change from macosx-port?
http://hg.openjdk.java.net/macosx-port/macosx-port/jdk/rev/50b8fef5df7d
The commit message for that changeset didn't have much information; it just
said:
"Fixing AWT components painted over in white"
Perha
Also note that CR 7177173 and 7124244 depends from this one.
Typo fixed. Should be:
Also note that CR *7124513* and 7124244 depends from this one.
--
Best regards, Sergey.
Hi Everyone,
Please review the fix for 7181438.
This bug was introduced in 7u4b11, when macosx-port was integrated to
jdku7 by this changeset:
http://hg.openjdk.java.net/jdk7u/jdk7u-dev/jdk/rev/9dfe50f456be
Note that now we always set format to GL_RGBA. So opaque surface has an
alpha channel.
