This is the mail archive of the
java-patches@gcc.gnu.org
mailing list for the Java project.
Re: [gui][patch] fixes to buffered image painting
- From: Ziga Mahkovec <ziga dot mahkovec at klika dot si>
- To: graydon hoare <graydon at redhat dot com>
- Cc: java-patches at gcc dot gnu dot org
- Date: Tue, 07 Dec 2004 03:30:30 +0100
- Subject: Re: [gui][patch] fixes to buffered image painting
- References: <41B4FD1B.3080106@redhat.com>
On Mon, 2004-12-06 at 19:45 -0500, graydon hoare wrote:
> this patch, which I've committed to java-gui-branch, fixes the
> BufferedImage Graphics2D paint performance problem Ziga reported last
> week. with this change, Graphics2D objects built on BufferedImages will
> construct a client-side cairo context which paints directly into the
> java int[] inside the image's Raster, accessed through JNI. the speed
> now ranges between "the same" and "twice as fast" as sun's
> BufferedImage-target painting (on gcj). it might be slower on systems in
> which JNI array access involves copying arrays.
This is great news! Just as I've implemented SVG rendering as a
workaround :)
I updated jhbuild and the patch works great (and fast too, faster than
IBM's JRE). I didn't see any text though ([1]), but I'm guessing this
is due to some pango errors I got during jhbuild. I'll test some more
tomorrow.
> unfortunately since there's still some incomplete parts of
> javax.imageio, we can't quite run bootchart yet. but it should be alive
> within the next week or two.
I have since switched to a different PNG encoder ([2]) but will
definitely give imageio a try once it's ready.
[1] http://www.klika.si/ziga/bootchart/images/bootchart-gcj.png
[2] http://catcode.com/pngencoder/
Thanks,
--
Ziga