[Bug 84958] [BDW] Screen stop at certain frame when running Synmark2 cases while FPS is updating

[bugzilla-daemon at freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=84958

Eero Tamminen <eero.t.tamminen at intel.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |eero.t.tamminen at intel.com
          Component|Drivers/DRI/i965            |Driver/intel
            Version|unspecified                 |git
           Assignee|idr at freedesktop.org         |chris at chris-wilson.co.uk
            Product|Mesa                        |xorg
         QA Contact|intel-3d-bugs at lists.freedes |intel-gfx-bugs at lists.freede
                   |ktop.org                    |sktop.org

--- Comment #1 from Eero Tamminen <eero.t.tamminen at intel.com> ---
Issue is not visible with latest kernel & Mesa on top of Ubuntu 14.04 X server
(1.15.1) and X video driver (2.99.910), but updating X server and X video
driver makes it visible (or having latest X & X intel video driver with old
Mesa & kernel).
-> seems like X issue, moving.  

It's BDW specific and has been there at least from Spring.

The issue is that although tests render correctly and report fine FPS numbers,
those rendered frames aren't shown on screen.  If SynMark2 validates the window
content (reads & compares it against correct version), that frame shows on
display just fine.  I.e. one cannot use test-automation to detect the issue, it
needs to be verified from screen externally/manually.

Only the first (= validated/read) frame in test is visible after context
creation.  Tests themselves haven't stopped as they report good FPS values when
test completes.  In SynMark context re-creation test one can see several
frames, but that test re-creates the context on every frame.

In which tests this happens, can change after reboot, and when 3D stack is
updated, so it could be (context creation?) timing related.

Wendy, didn't you see this also in some other test than SynMark (Lightmark
2008?).

-- 
You are receiving this mail because:
You are the QA Contact for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.freedesktop.org/archives/intel-gfx-bugs/attachments/20141015/6bdda595/attachment.html>