Hi, I am seeing the issue on a machine with the following configuration:
- Kubuntu linux 15.10 64 bit
X.Org X Server 1.17.2
Intel driver 2.99.917 (+git 20150808 as shipped by ubuntu)
- Intel haswell laptop
Intel Corporation Crystal Well Integrated Graphics Controller (rev 08)
The issue seems to be triggered by using Libreoffice 5.0.3 impress for a sufficiently long time. It is visible in the slide editing mode, not in presentation.
When you start using libreoffice impress everything is fine.
After some use, the screen stops updating. For instance, you press page down and the slide you are working on does not change. Or you click on an object handle and the latter does not appear as selected. However, as soon as you move the mouse, the screen gets updated.
Note that the issue does not happen immediately. You need to use Libreoffice on a complex document for about 1 hour before seeing the issue.
It is unclear to me if this is an issue with Libreoffice or with the Intel driver. However, from googling I see that in the people ware complaining about something very similar happening with firefox.
The biggest question is whether the whole screen updates or only the region around the mouse?
In the last day, we tracked down a refleak on a fence used for synchronising DRI clients that was causing memory exhaustion. As you also have a slow issue, it may be related. So it is worth grabbing xf86-video-intel.git and giving a try, and please attach your Xorg.0.log when doing so.
I would say the whole screen, or at least a very large area thereof, since it involves a whole slide being edited on LibO impress. I'll try to see what happens to the clock on the screen screen.
firstname.lastname@example.org, if this is reproducible with http://cdimage.ubuntu.com/daily-live/current/, it will help immensely if you filed a new report with Ubuntu by ensuring you have the package xdiagnose installed, and that you click the Yes button for attaching additional debugging information running the following from a terminal:
Also, please feel free to subscribe me to it.
For more on why this is helpful, please see https://wiki.ubuntu.com/ReportingBugs.
Unfortunately, I cannot test right now on a live image, since I am under pressure with work and the bug does not reveal immediately. I'll do it when possible.
For the current time, I can say that switching to the updated drivers in the OIBAF PPA seems to have fixed (cannot be 100% sure, but I am somehow confident) this odd graphics behavior. This is consistent with Chris' suggestion.
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct.