Rendering in geode

[Huang, FrankR]

Jonathan,

	Absoultely, I am using a off-screen framebuffer to expand the 1-pixel solid colour to the mask width*height scale region (do the src*mask operation). I think there is no wrong I use 4*PITCH, and when I use rendercheck to test this part code, it is same with ATI result.
	But it did wrong with Firefox. The link below is the result I got, it is easy to reproduce. Just open the firefox then press the refresh button, then you can see the result on my link.
	https://bugs.freedesktop.org/attachment.cgi?id=36486
	I still can not find the hint on how the black region and discrete line are appearing when I just press the refresh button. What the firefox will do when I press the refresh button? How could I debug this?

Thanks,
Frank

-----Original Message-----
From: xorg-devel-bounces+frankr.huang=amd.com at lists.x.org [mailto:xorg-devel-bounces+frankr.huang=amd.com at lists.x.org] On Behalf Of Jonathan Morton
Sent: 2010?6?24? 20:03
To: Huang, FrankR
Cc: xorg-driver-geode at lists.x.org; xorg-devel at lists.x.org
Subject: RE: Rendering in geode

On Thu, 2010-06-24 at 17:21 +0800, Huang, FrankR wrote:
> The reason I use 4*PITCH is that the src is A8 picture, and the src'
> format is ARGB32, so it is 4*PITCH. I don't know if that is reasonable
> to appoint that pitch directly instead of exaGetPixmapPitch().

Since the source is typically a 1-pixel "solid colour" image in this
situation, the stride of that image would not be directly relevant.  The
stride of the destination would be independent of both source and mask.

Looking very briefly at your code, it seems that you are using a
temporary image to expand the source colour to cover the required area.
You are, of course, in complete control of that temporary image, so you
can use any stride you like that does not cause overlaps.

-- 
------
From: Jonathan Morton
      jonathan.morton at movial.com


_______________________________________________
xorg-devel at lists.x.org: X.Org development
Archives: http://lists.x.org/archives/xorg-devel
Info: http://lists.x.org/mailman/listinfo/xorg-devel