With version 1.5.20 in Ubuntu Hardy, artifacts are left on the screen by applications that use cario for rendering. And example is AWN, which leaves artifacts on the screen after resizing. The problem is difficult to reproduce - it seems to only happen every once in a while - I happened twice after upgrading (I unfortunately took no screenshots) and has not happened since. I've noticed this in only in AWN so far, and other apps seem unaffected, but that could be solely because of the low frequency of the bug.
Are you confident that the bug is unique to cairo 1.5.20? Did you not see the bug under similar usage with previous cairo versions? And in that case, do you know which cairo versions you had been using previously? We are just about to relabel 1.5.20 as our big 1.6.0 release, so it would be good to know if there's a recent regression here. Thanks for any input you can provide. -Carl
The bug seems to also have appeared in 1.5.18 (due to reports from others) but I never upgraded to that version (it was pulled from the repos). The previous version was probably 1.5.16, but I don't think its possible to be certain. In any case, before the upgrade, the bug did not show up. I can't be certain that this is a bug in cairo itself, but it seems likely due to the reports of major regressions in 1.5.18 - perhaps some sneaked into 1.5.20. Since filing the bug report, I've been unable to replicate it again. I'll provide any help I can, don't want 1.6 to have this bug :) (In reply to comment #0) > With version 1.5.20 in Ubuntu Hardy, artifacts are left on the screen by > applications that use cario for rendering. And example is AWN, which leaves > artifacts on the screen after resizing. > > The problem is difficult to reproduce - it seems to only happen every once in a > while - I happened twice after upgrading (I unfortunately took no screenshots) > and has not happened since. I've noticed this in only in AWN so far, and other > apps seem unaffected, but that could be solely because of the low frequency of > the bug. > (In reply to comment #1) > Are you confident that the bug is unique to cairo 1.5.20? > > Did you not see the bug under similar usage with previous cairo versions? > > And in that case, do you know which cairo versions you had been using > previously? > > We are just about to relabel 1.5.20 as our big 1.6.0 release, so it would > be good to know if there's a recent regression here. > > Thanks for any input you can provide. > > -Carl >
(In reply to comment #2) > Since filing the bug report, I've been unable to replicate it again. > > I'll provide any help I can, don't want 1.6 to have this bug :) I don't want the bug to be there either, but if we can't reproduce the bug, I don't know what else we can do. If we can find a way to replicate the bug, (even with some low probability), then it should be very easy to bisect down to the buggy commit. There really aren't all that many changes from 1.5.16 to 1.5.20. Thanks again, -Carl
Reproduced! Seems to happen randomly, but it finally did happen - the attached screenshot shows major artifacts in AWN (bottom of screen) which uses cairo for drawing. This effect went away shortly - I had just enough time to take a screenshot.
Created attachment 15825 [details] Screenshot of problem Demonstrates the problem. Note bottom of screenshot.
What graphics card? What version of the X server?
xserver-xorg 1:7.3+10ubuntu8 (/var/lib/apt/lists/us.archive.ubuntu.com_ubuntu_dists_hardy_main_binary-i386_Packages) (/var/lib/dpkg/status) graphics card: nVidia 7600
Looking at the screenshot, I'm none the wiser. I can't tell whether the ugly stippling effect is intentional or not. If not, the bug could be in any of the pixman (or driver) composition routines. What is the current status of this bug?
No duplicates and no activity. Hopefully the bug has been resolved, and in all probability was due to the binary driver.
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.