We ran xtest suite on a x86-64 machine, using i810 driver. We found some cases failed due to the background pixel value ~0L. Xlib9/XDrawArc 91 Xlib9/XDrawArcs 101 Xlib9/XDrawImageString16 19 Xlib9/XDrawImageString 19 Xlib9/XDrawLine 80 Xlib9/XDrawLines 92 Xlib9/XDrawRectangle 72 Xlib9/XDrawRectangles 74 Xlib9/XDrawSegments 81 Xlib9/XFillArc 38 Xlib9/XFillArcs 40 Xlib9/XFillPolygon 43 Xlib9/XFillRectangle 37 Xlib9/XFillRectangles 38 Xlib9/XPutImage 32 These cases create a stipple with data 0, set the fill style to FillOpaqueStippled and specify the background pixel value ~0L by XSetBackground call. So it will draw the graphic (such as arc, lines) with background pixel value ~0L. But the result is this background pixel value ~0L doesn't take effect. All pixel is drawn with pixel value 0. If these cases run with vesa driver, they are passed. If we change the stipple and fill style and use foreground pixel value ~0L to draw these graphic, it is OK.
I highly suspect that these are XAA bugs rather than driver bugs. Can you try another driver that uses XAA and confirm whether you see them with another driver ?
Yes, itt looks like XAA bug. The cases pass if XAA disabled.
Sorry about the phenomenal bug spam, guys. Adding xorg-team@ to the QA contact so bugs don't get lost in future.
The bug priority was upgraded (P2->high) with the bugzilla configuration change. I'm Changing the priority back to the normal one. Sorry for the spam.
And the cases also pass with EXA.
Just pushed Hong Liu's change for the xaaGC.c fix. Can you try again ? If it works can work make this bug depend on 8888 so it gets into the 7.3 release.
Fixed in xserver master 9adea807038b64292403ede982075fe1dcfd4c9a.
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.