[Bug 79843] New: i855gm: Display corruption in presence of graphics memory pressure

[bugzilla-daemon at freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=79843

          Priority: medium
            Bug ID: 79843
                CC: intel-gfx-bugs at lists.freedesktop.org
          Assignee: intel-gfx-bugs at lists.freedesktop.org
           Summary: i855gm: Display corruption in presence of graphics
                    memory pressure
        QA Contact: intel-gfx-bugs at lists.freedesktop.org
          Severity: normal
    Classification: Unclassified
                OS: Linux (All)
          Reporter: bonbons67 at internet.lu
          Hardware: x86 (IA32)
            Status: NEW
           Version: XOrg CVS
         Component: DRM/Intel
           Product: DRI

As determined with the help if Chris, on my i855GM system graphics corruption
seems to be triggered by i915 moving objects to/from GPU memory/aperture.

The corruption is visible either as:
- ghost-screenshot of some previously displayed surface, usually reflowed to
different width
- on some GTK applications, on scroll down by 1 line all the rest of the
surface gets repainted with repetitions of new line (common on hexchat/xchat,
but also mail view pane of claws-mail or pidgin message window). Scroll-up is
not affected.


History of theses corruptions: exist since a long time, show up mostly when
compositing (in Enlightenment, maybe due to fallback to software rendering as E
wants shaders for OpenGL acceleration)

Latest software stack used for testing:
- Kernel "drm-intel-nightly: 2014y-06m-02d-19h-49m-25s" on top of 3.15-rc8
- x11-libs/libdrm-2.4.52
- xf86-video-intel GIT from 31 may 2014

See also mail exchange with Chris:
  Message-ID: <20140427223402.572010a9 at neptune.home>
which ended with:
  "Hmm. They do suggest that your system is under memory pressure - it has
to move objects in and out the GPU address space regularly - but they
don't suggest that you are out of memory."

-- 
You are receiving this mail because:
You are the QA Contact for the bug.
You are on the CC list for the bug.
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.freedesktop.org/archives/intel-gfx-bugs/attachments/20140609/44cf5c57/attachment.html>