On 6/23/06, Martijn van Oosterhout <[EMAIL PROTECTED]> wrote:
> Well, I can't see the problem. There no docs publically available on
> the chip and the code isn't commented either. Given there's only some
> dirty windows under certain circumstances, I''m not that motivated to
> fix it.

Gah, I just can't leave this alone. OTOH, I think I've nearly sussed
it. I noticed that the problem has a far more visible effect: if you
move the window so it's off the right of the screen, some of the
drawing wraps around to the left. Now, I'm almost certain the hardware
rendering engine is smart enough to not do that. And in particular,
certain things like circles and text, which I know are not hardware
accelerated, always wrap. This places the fault squarely on the
software renderer.

What I think is happening is that there is disagreement on whether the
pClipRects field in __DRIdrawablePrivateRec is in screen coordinates
or window coordinates. And the reason it only happens sometimes to
polygons is because only sometimes is the program enabling an option
that causes the code to fallback to software.

Currently the pClipRects are in screen coordinates, but I *think* the
clipping code is assuming they're in window coordinates, which what
causes the clipping to fail. So, could someone please tell me: what
should those fields be relative to?

Thanks in advance,
-- 
Martijn van Oosterhout <[EMAIL PROTECTED]> http://svana.org/kleptog/

Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Mesa3d-dev mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/mesa3d-dev

Reply via email to