[Bug 73792] New: UVD usage with HD makes X11 quite unstable on Samsung ATIV Book Lite 9 (A4-1450 CPU, KABINI graphics))

[bugzilla-daemon at freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=73792

          Priority: medium
            Bug ID: 73792
          Assignee: dri-devel at lists.freedesktop.org
           Summary: UVD usage with HD makes X11 quite unstable on Samsung
                    ATIV Book Lite 9 (A4-1450 CPU, KABINI graphics))
          Severity: normal
    Classification: Unclassified
                OS: All
          Reporter: rvdb at cryptolab.net
          Hardware: Other
            Status: NEW
           Version: git
         Component: Drivers/Gallium/radeonsi
           Product: Mesa

Created attachment 92387
  --> https://bugs.freedesktop.org/attachment.cgi?id=92387&action=edit
Dmesg output

I have a Smasung ATIV Bool Lite 9 notebook on which I installed a
Slackware-current system which works roughly well except suspend and UVD
playback.
About the later, I run a Kernel 3.12.8 with less-than-a-week-ago git pulls of
libdrm, glamor, mesa, mesa-glu and xfree-video-ati.
Desktop graphic and compositing work well without glitch and no specific lag.
I use libvdpau 0.7 and less-than-a-week-ago git pulls of ffmpeg, x264 and vlc.
vlc is configured to use VDPAU and this the UVD chip in the processor.
When running SD h264 videos, everything works fine and the CPU usage is low
(less than 75% of total W11/KWin/vlc on a GHz core, that's fine with me).
I get issues when trying to run HD h264 videos. The CPU usage is still fine,
the video starts but after some time, the video hangs, the system becomes
unresponsive then the screen turns black, as displaying a black image, it's a
bit less black then when the screen is turned of.
At this point, either I need to reboot or at some point, the screen seems to
turn of, then light up again and resume where I was, though sometimes the video
remains black and I have to move the VLC slider to get some images. Then the
film can play several seconds then the whole pattern occurs again. And so on
until the system does not resume and I must reboot.
I've attached dmesg, syslog, glxinfo et vdpauinfo outputs.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.freedesktop.org/archives/dri-devel/attachments/20140119/757d6ae3/attachment.html>