[Bug 26023] New: Multi-monitor issues with RV710 (HD4350)

[bugzilla-daemon at freedesktop.org]

http://bugs.freedesktop.org/show_bug.cgi?id=26023

           Summary: Multi-monitor issues with RV710 (HD4350)
           Product: xorg
           Version: 7.4
          Platform: x86-64 (AMD64)
        OS/Version: Linux (All)
            Status: NEW
          Severity: normal
          Priority: medium
         Component: Driver/Radeon
        AssignedTo: xorg-driver-ati at lists.x.org
        ReportedBy: gtdev at spearhead.de
         QAContact: xorg-team at lists.x.org


Created an attachment (id=32607)
 --> (http://bugs.freedesktop.org/attachment.cgi?id=32607)
Server log with "stock" driver

Driver version: 6.12.4 (Debian unstable packages)
Kernel: Linux 2.6.32.3 (stock)

This may be related to http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=550637

The issue is that with DRI enabled and two monitors connected (HDMI and DVI)
(no matter whether running mirrored or some separate layout), drawing
operations anywhere on the screen cause short-lived corruption. It looks like
partial lines are replaced with uninitialized memory. This happens in KDE and
Gnome, with or without acceleration, as long as DRI is not disabled in the
xorg.conf. It seems like the effect is most severe on pixel rows where some
changes occur.

I tried the setting DisplayPriority option to HIGH as mentioned in the Debian
bug report, but that did no good.

The effect does not seem to occur in current git master, but there I get
different issues, making the whole thing unusable. It seems that screen sizes
get mixed up, so that the screen doesn't get drawn properly. I have a 1920x1200
display in landscape mode, and a 1600x1200 display in portrait to the right of
it. When setting that up with the xrandr tool or the Gnome display tool, the
left monitor has a picture that seems to be only 1200 pixels wide stuck to the
left, while the right of the screen is black, but the mouse pointer moves into
the area. At some point, the driver appeared to forget about the right monitor
altogether, displaying only a background image and a functional mouse pointer.
Windows on that screen still seem to be there, since the pointer changes
shapes. During some attemps, I also experienced serious mouse pointer
corruption.

None of those effects occur when only physically connecting a single monitor;
Both the 6.12.4 and the git master drivers work fine then.


-- 
Configure bugmail: http://bugs.freedesktop.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.