[Bug 84971] New: Display switching off.

[bugzilla-daemon at freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=84971

            Bug ID: 84971
           Summary: Display switching off.
           Product: xorg
           Version: unspecified
          Hardware: x86-64 (AMD64)
                OS: Linux (All)
            Status: NEW
          Severity: major
          Priority: medium
         Component: Driver/intel
          Assignee: chris at chris-wilson.co.uk
          Reporter: alexsecret at hotmail.com
        QA Contact: intel-gfx-bugs at lists.freedesktop.org

System environment:
-- chipset: 965GM
-- system architecture: 64-bit
-- xf86-video-intel: 2.99.911
-- xserver: 1.15.1
-- mesa: 10.2.2
-- libdrm: 2.4.54
-- kernel: 3.13.0-37-generic
-- Linux distribution: Xubuntu 14.04 LTS
-- Machine or mobo model: Lenovo Thinkpad R61 laptop


Hi,

I am using Xubuntu 14.04 64-bits and Chrome as my main browser.  The computer
is a Lenovo Thinkpad R61 laptop with an Intel 965GM video card inside. 
Although everything was going fine when I was using Xubuntu 12.04, now, with
14.04, I am facing a problem that I am not able to find either a solution or a
successful workaround for.

When I'm trying to watch a flash video, most of the time it works fine but
randomly, the playback will pause right after it starts and 3-4 seconds later
the screen will go black.  By saying black, I mean it's switched off, not
simply blank.  It doesn't go back on unless I restart the laptop.  Everything
else keeps working fine when this happens, and I can even hear the sound of the
video playing.  I can press Ctrl-W and close the Chrome window or Ctrl-Alt-F1
and go to TTY1 mode, logon and turn the system off normally using the poweroff
command.

Sometimes, when I see the video pause like that and rush to close its window
before the screen goes black, I can avoid it from happening.  If I attempt to
play the same video again right after that, it plays fine and, most of the
time, many more videos play just fine afterwards.  Till it happens again for
some unknown reason so far and the story goes on...

The only app that causes this on a standard basis is glmark2 on a specific test
it runs, which is called "[ideas] speed=duration:".  All previous tests are ok.
 I don't know about the tests after this specific one of course.  I never get
to test them.

Additionally, Chrome has caused this problem two more times and both are not
flash video related.  It happened for the first time when I clicked on the
"Download Chrome" link on the google page at the point where it usually brings
up a separate white window where the file links are and the second time when I
clicked on a messenger contact in hotmail to see what happens now they brought
msn back.  Never got to find out eventually.  :D

The point is that I've clicked the "Download Chrome" link tens of times after
that and it always worked fine.  This issue is completely random.

It happens with both the Intel drivers installed with the system and the latest
one I installed from Intel using 1.0.6 installer.  I also tried switching
AccelMethod from SNA (the default) to UXA which seemed to save others from this
issue but in my case it didn't.

As you understand, this is quite serious and it has become even more serious
after the latest Chrome update to version 38 since it happens more often.

This is the beginning of the log entries concerning this, in syslog:
Oct 13 03:41:39 Lenovo kernel: [41804.816060] [drm] stuck on render ring
Oct 13 03:41:39 Lenovo kernel: [41804.816069] [drm] GPU crash dump saved to
/sys/class/drm/card0/error
Oct 13 03:41:39 Lenovo kernel: [41804.816071] [drm] GPU hangs can indicate a
bug anywhere in the entire gfx stack, including userspace.
Oct 13 03:41:39 Lenovo kernel: [41804.816079] [drm] Please file a _new_ bug
report on bugs.freedesktop.org against DRI -> DRM/Intel
Oct 13 03:41:39 Lenovo kernel: [41804.816080] [drm] drm/i915 developers can
then reassign to the right component if it's not a kernel issue.
Oct 13 03:41:39 Lenovo kernel: [41804.816081] [drm] The gpu crash dump is
required to analyze gpu hangs, so please always attach it.
Oct 13 03:41:39 Lenovo kernel: [41804.817061] [drm:i915_set_reset_status]
*ERROR* render ring hung inside bo (0xb99e000 ctx 0) at 0xb99f804
Oct 13 03:41:40 Lenovo kernel: [41805.332085] [drm:i915_reset] *ERROR* Failed
to reset chip.
Oct 13 03:41:40 Lenovo kernel: [41805.332791] ------------[ cut here
]------------
Oct 13 03:41:40 Lenovo kernel: [41805.332879] WARNING: CPU: 1 PID: 1834 at
/build/buildd/linux-3.13.0/drivers/gpu/drm/i915/intel_display.c:922
assert_pll+0x68/0x70 [i915]()
Oct 13 03:41:40 Lenovo kernel: [41805.332884] PLL state assertion failure
(expected on, current off)

The file (/sys/class/drm/card0/error) it's referring to, is empty.  Nothing is
actually saved in there.



I'd appreciate it if you could help on this one.  Thanks a lot everyone.  ;)

-- 
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/20141013/72bf624b/attachment-0001.html>