Bug 82835 - [NV92] GeForce 8800 GS VDPAU h264 decoding hang
Summary: [NV92] GeForce 8800 GS VDPAU h264 decoding hang
Status: NEEDINFO
Alias: None
Product: xorg
Classification: Unclassified
Component: Driver/nouveau (show other bugs)
Version: git
Hardware: x86 (IA32) Linux (All)
: medium normal
Assignee: Nouveau Project
QA Contact: Xorg Project Team
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2014-08-20 03:55 UTC by Andrew Randrianasulu
Modified: 2017-11-25 12:24 UTC (History)
3 users (show)

See Also:
i915 platform:
i915 features:


Attachments
dmesg from 3.17.0-rc1 kernel, nouveau.debug=debug (93.77 KB, text/plain)
2014-08-20 03:55 UTC, Andrew Randrianasulu
no flags Details
X log (35.81 KB, text/plain)
2014-08-20 03:55 UTC, Andrew Randrianasulu
no flags Details
dmesg with new mesa (98.60 KB, text/plain)
2014-09-02 16:25 UTC, Andrew Randrianasulu
no flags Details
dmesg again, after some system use (but ends up with usual hang with mplayer) (130.38 KB, text/plain)
2014-09-03 01:58 UTC, Andrew Randrianasulu
no flags Details
mesa's config.log (128.18 KB, text/plain)
2014-09-03 01:59 UTC, Andrew Randrianasulu
no flags Details
hwtest stdout log (5.95 KB, text/plain)
2015-10-19 14:56 UTC, Andrew Randrianasulu
no flags Details

Note You need to log in before you can comment on or make changes to this bug.
Description Andrew Randrianasulu 2014-08-20 03:55:35 UTC
Created attachment 104933 [details]
dmesg from 3.17.0-rc1 kernel, nouveau.debug=debug

So, I  followed  instructions  at http://nouveau.freedesktop.org/wiki/VideoAcceleration/ and tried vdpau/h264 acceleration for  my  card. It  hanged  happily...

Kernel: both self-compiled 3.14.9 (64-bit) and 3.17.0-rc1 (32-bit) failed.
Mesa: currently OpenGL version string: 3.0 Mesa 10.3.0-devel (git-a4359bc)
libdrm: git d9ce09de12131fa8c5ff2d7c2dbd6b61d8a5fd14 (" exynos: Use symbol visibility.")
xf86-video-nouveau: 16c885ce9c7bebcfa1252a5d500956dd5f551043 (" Fix event handling on multi-x-screen configs.")
Xserver: 1.12.4 (from 1.12 branch)

+ MPlayer SVN-r37116 (32  bit, like  full userspace), compiled  with gcc 4.7.3

Command  line:
 mplayer ~/botva/vid/MMC_14_may_2011_four_pilot_whales.mp4 -quiet -vo vdpau -vc ffh264vdpau -fs (it  hangs  without 'fs', too)

Same  video plays  if  VDPAU  used  only  for  displaying.

vdpauinfo:
 vdpauinfo
display: :0   screen: 0
API version: 1
Information string: G3DVL VDPAU Driver Shared Library version 1.0

Video surface:

name   width height types
-------------------------------------------
420     8192  8192  NV12 YV12
422     8192  8192  UYVY YUYV
444     8192  8192  Y8U8V8A8 V8U8Y8A8

Decoder capabilities:

name               level macbs width height
-------------------------------------------
MPEG1                 0 16384  2048  2048
MPEG2_SIMPLE          3 16384  2048  2048
MPEG2_MAIN            3 16384  2048  2048
H264_BASELINE        41 16384  2048  2048
H264_MAIN            41 16384  2048  2048
H264_HIGH            41 16384  2048  2048

Output surface:

name              width height nat types
----------------------------------------------------
B8G8R8A8          8192  8192    y  NV12 YV12 UYVY YUYV Y8U8V8A8 V8U8Y8A8
R8G8B8A8          8192  8192    y  NV12 YV12 UYVY YUYV Y8U8V8A8 V8U8Y8A8
R10G10B10A2       8192  8192    y  NV12 YV12 UYVY YUYV Y8U8V8A8 V8U8Y8A8
B10G10R10A2       8192  8192    y  NV12 YV12 UYVY YUYV Y8U8V8A8 V8U8Y8A8

Bitmap surface:

name              width height
------------------------------
B8G8R8A8          8192  8192
R8G8B8A8          8192  8192
R10G10B10A2       8192  8192
B10G10R10A2       8192  8192
A8                8192  8192

Video mixer:

feature name                    sup
------------------------------------
DEINTERLACE_TEMPORAL             y
DEINTERLACE_TEMPORAL_SPATIAL     -
INVERSE_TELECINE                 -
NOISE_REDUCTION                  y
SHARPNESS                        y
LUMA_KEY                         -
HIGH QUALITY SCALING - L1        -
HIGH QUALITY SCALING - L2        -
HIGH QUALITY SCALING - L3        -
HIGH QUALITY SCALING - L4        -
HIGH QUALITY SCALING - L5        -
HIGH QUALITY SCALING - L6        -
HIGH QUALITY SCALING - L7        -
HIGH QUALITY SCALING - L8        -
HIGH QUALITY SCALING - L9        -

parameter name                  sup      min      max
-----------------------------------------------------
VIDEO_SURFACE_WIDTH              y        48     2048
VIDEO_SURFACE_HEIGHT             y        48     2048
CHROMA_TYPE                      y
LAYERS                           y         0        4

attribute name                  sup      min      max
-----------------------------------------------------
BACKGROUND_COLOR                 y
CSC_MATRIX                       y
NOISE_REDUCTION_LEVEL            y      0.00     1.00
SHARPNESS_LEVEL                  y     -1.00     1.00
LUMA_KEY_MIN_LUMA                y
LUMA_KEY_MAX_LUMA                y
Comment 1 Andrew Randrianasulu 2014-08-20 03:55:59 UTC
Created attachment 104934 [details]
X log
Comment 2 Ilia Mirkin 2014-08-20 04:04:25 UTC
From the log:

nouveau D[    PBSP][0000:05:00.0] Loading firmware to address: 0x17824000
nouveau D[   PFIFO][0000:05:00.0][0xc000826f][f5f05e40] attached PBSP context
nouveau D[     PVP][0000:05:00.0] reset
nouveau D[     PVP][0000:05:00.0] Loading firmware to address: 0x177d4000
nouveau D[   PFIFO][0000:05:00.0][0xc000826f][f5c0bf00] attached PVP context
nouveau E[  PGRAPH][0000:05:00.0] TRAP_PROP - TP 0 - RT_FAULT - Address 0020404c40
nouveau E[  PGRAPH][0000:05:00.0] TRAP_PROP - TP 0 - e0c: 00000000, e18: 00000000, e1c: 00400010, e20: 00001100, e24: 00030000
nouveau E[  PGRAPH][0000:05:00.0] TRAP_PROP - TP 1 - RT_FAULT - Address 0020404000
nouveau E[  PGRAPH][0000:05:00.0] TRAP_PROP - TP 1 - e0c: 00000000, e18: 00000000, e1c: 00340000, e20: 00001100, e24: 00030000
nouveau E[  PGRAPH][0000:05:00.0] TRAP_PROP - TP 2 - RT_FAULT - Address 0020404040
nouveau E[  PGRAPH][0000:05:00.0] TRAP_PROP - TP 2 - e0c: 00000000, e18: 00000000, e1c: 00340010, e20: 00001100, e24: 00030000
nouveau E[  PGRAPH][0000:05:00.0] TRAP_PROP - TP 3 - RT_FAULT - Address 0020404080
nouveau E[  PGRAPH][0000:05:00.0] TRAP_PROP - TP 3 - e0c: 00000000, e18: 00000000, e1c: 00340020, e20: 00001100, e24: 00030000
nouveau E[  PGRAPH][0000:05:00.0] TRAP_PROP - TP 4 - RT_FAULT - Address 00204040c0
nouveau E[  PGRAPH][0000:05:00.0] TRAP_PROP - TP 4 - e0c: 00000000, e18: 00000000, e1c: 00340030, e20: 00001100, e24: 00030000
nouveau E[  PGRAPH][0000:05:00.0] TRAP_PROP - TP 5 - RT_FAULT - Address 0020404c00
nouveau E[  PGRAPH][0000:05:00.0] TRAP_PROP - TP 5 - e0c: 00000000, e18: 00000000, e1c: 00400000, e20: 00001100, e24: 00030000
nouveau E[  PGRAPH][0000:05:00.0] ch 4 [0x00178fb000 mplayer[7033]] subc 3 class 0x8297 mthd 0x1b0c data 0x1000f010
nouveau E[     PFB][0000:05:00.0] trapped write at 0x00204040c0 on channel 0x000178fb [mplayer[7033]] PGRAPH/PROP/RT0 reason: PAGE_NOT_PRESENT
nouveau D[   PFIFO][0000:05:00.0][0xc000826f][f5f05e40] detached PBSP context
nouveau D[   PFIFO][0000:05:00.0][0xc000826f][f5c0bf00] detached PVP context
nouveau E[     PFB][0000:05:00.0] trapped write at 0x002050c6c0 on channel 0x000178fb [mplayer[7033]] PGRAPH/PROP/RT0 reason: PAGE_NOT_PRESENT


And then things go downhill. I wonder why it detaches the PBSP/PVP contexts, that's a bit odd.
Comment 3 Ilia Mirkin 2014-09-01 22:57:50 UTC
I made a couple of fixes in git... they *might* affect your situation (although TBH I'd be a little surprised). Can you test it out with a mesa from git?

Note that LD_LIBRARY_PATH has no effect on which libvdpau_nouveau.so is loaded... make sure to set VDPAU_DRIVER_PATH (and have a new enough libvdpau that supports that... 0.8 iirc). Or install a libvdpau into the same prefix that the mesa install goes into, then LD_LIBRARY_PATH should do what you expect.
Comment 4 Andrew Randrianasulu 2014-09-02 16:24:27 UTC
No, sadly hang still here  with mesa git-021e84f and nouveau kernel up to b0ae082dcb4304eb2443b3319b89fb5cb9202a5a ("therm/nv84+: do not expose non-calibrated internal temp sensor")
Comment 5 Andrew Randrianasulu 2014-09-02 16:25:11 UTC
Created attachment 105618 [details]
dmesg with new  mesa
Comment 6 Ilia Mirkin 2014-09-02 16:55:42 UTC
Hm, just noticed this (same as last time, but I guess I wasn't paying enough attention):

nouveau E[  PGRAPH][0000:05:00.0] ch 4 [0x00178fb000 mplayer[1645]] subc 3 class 0x8297 mthd 0x1b0c data 0x1000f010
nouveau E[     PFB][0000:05:00.0] trapped write at 0x00204050c0 on channel 0x000178fb [mplayer[1645]] PGRAPH/PROP/RT0 reason: PAGE_NOT_PRESENT

method 1b0c == QUERY_GET. However the data is 0x1000f010, not 0xf010, which means it's not coming from the video stuff or the screen fence stuff, but rather is a PIPE_QUERY_GPU_FINISHED query. Which is interesting, because nothing appears to actually use that PIPE_QUERY type...

Could you stick a assert(0) right before

      nv50_query_get(push, q, 0, 0x1000f010);

in nv50_query.c and see if it triggers (make sure to use a debug build... or make it something nastier than assert), and get a backtrace from gdb? (Or actually just set a breakpoint on that line and see if gdb hits it...)
Comment 7 Andrew Randrianasulu 2014-09-02 23:51:04 UTC
(In reply to comment #6)
> Hm, just noticed this (same as last time, but I guess I wasn't paying enough
> attention):
> 
> nouveau E[  PGRAPH][0000:05:00.0] ch 4 [0x00178fb000 mplayer[1645]] subc 3
> class 0x8297 mthd 0x1b0c data 0x1000f010
> nouveau E[     PFB][0000:05:00.0] trapped write at 0x00204050c0 on channel
> 0x000178fb [mplayer[1645]] PGRAPH/PROP/RT0 reason: PAGE_NOT_PRESENT
> 
> method 1b0c == QUERY_GET. However the data is 0x1000f010, not 0xf010, which
> means it's not coming from the video stuff or the screen fence stuff, but
> rather is a PIPE_QUERY_GPU_FINISHED query. Which is interesting, because
> nothing appears to actually use that PIPE_QUERY type...
> 
> Could you stick a assert(0) right before
> 
>       nv50_query_get(push, q, 0, 0x1000f010);
> 
> in nv50_query.c and see if it triggers (make sure to use a debug build... or
> make it something nastier than assert), and get a backtrace from gdb? (Or
> actually just set a breakpoint on that line and see if gdb hits it...)

No, I tried  to add assert :

diff --git a/src/gallium/drivers/nouveau/nv50/nv50_query.c b/src/gallium/drivers/nouveau/nv50/nv50_query.c
index a373dc6..23f5eaf 100644
--- a/src/gallium/drivers/nouveau/nv50/nv50_query.c
+++ b/src/gallium/drivers/nouveau/nv50/nv50_query.c
@@ -222,6 +222,7 @@ nv50_query_end(struct pipe_context *pipe, struct pipe_query *pq)
       break;
    case PIPE_QUERY_GPU_FINISHED:
       q->sequence++;
+      assert(0);
       nv50_query_get(push, q, 0, 0x1000f010);
       break;
    case NVA0_QUERY_STREAM_OUTPUT_BUFFER_OFFSET:

on top of mesa-git-454aab4 and it doesn't trigger  during mplayer's run (resulted in usual hang).
Comment 8 Ilia Mirkin 2014-09-03 00:03:48 UTC
(In reply to comment #7)
> (In reply to comment #6)
> > Hm, just noticed this (same as last time, but I guess I wasn't paying enough
> > attention):
> > 
> > nouveau E[  PGRAPH][0000:05:00.0] ch 4 [0x00178fb000 mplayer[1645]] subc 3
> > class 0x8297 mthd 0x1b0c data 0x1000f010
> > nouveau E[     PFB][0000:05:00.0] trapped write at 0x00204050c0 on channel
> > 0x000178fb [mplayer[1645]] PGRAPH/PROP/RT0 reason: PAGE_NOT_PRESENT
> > 
> > method 1b0c == QUERY_GET. However the data is 0x1000f010, not 0xf010, which
> > means it's not coming from the video stuff or the screen fence stuff, but
> > rather is a PIPE_QUERY_GPU_FINISHED query. Which is interesting, because
> > nothing appears to actually use that PIPE_QUERY type...
> > 
> > Could you stick a assert(0) right before
> > 
> >       nv50_query_get(push, q, 0, 0x1000f010);
> > 
> > in nv50_query.c and see if it triggers (make sure to use a debug build... or
> > make it something nastier than assert), and get a backtrace from gdb? (Or
> > actually just set a breakpoint on that line and see if gdb hits it...)
> 
> No, I tried  to add assert :
> 
> diff --git a/src/gallium/drivers/nouveau/nv50/nv50_query.c
> b/src/gallium/drivers/nouveau/nv50/nv50_query.c
> index a373dc6..23f5eaf 100644
> --- a/src/gallium/drivers/nouveau/nv50/nv50_query.c
> +++ b/src/gallium/drivers/nouveau/nv50/nv50_query.c
> @@ -222,6 +222,7 @@ nv50_query_end(struct pipe_context *pipe, struct
> pipe_query *pq)
>        break;
>     case PIPE_QUERY_GPU_FINISHED:
>        q->sequence++;
> +      assert(0);
>        nv50_query_get(push, q, 0, 0x1000f010);
>        break;
>     case NVA0_QUERY_STREAM_OUTPUT_BUFFER_OFFSET:
> 
> on top of mesa-git-454aab4 and it doesn't trigger  during mplayer's run
> (resulted in usual hang).

Can you confirm that you built this with --enable-debug? And you saw the 0x1000f010 value in your dmesg?
Comment 9 Andrew Randrianasulu 2014-09-03 01:57:05 UTC
Yes, I can confirm new dmesg still contain mentioned  value:

root@slax:~# cat nv92_dmesg_new_1.log | grep 0x1000f010
nouveau E[  PGRAPH][0000:05:00.0] ch 4 [0x00178fb000 mplayer[4559]] subc 3 class 0x8297 mthd 0x1b0c data 0x1000f010
nouveau E[  PGRAPH][0000:05:00.0] ch 4 [0x00178fb000 mplayer[4559]] subc 3 class 0x8297 mthd 0x1b0c data 0x1000f010
root@slax:~#
Comment 10 Andrew Randrianasulu 2014-09-03 01:58:27 UTC
Created attachment 105647 [details]
dmesg again, after some system use (but ends up with usual hang with mplayer)
Comment 11 Andrew Randrianasulu 2014-09-03 01:59:34 UTC
Created attachment 105648 [details]
mesa's config.log
Comment 12 Ilia Mirkin 2014-09-03 03:35:16 UTC
(In reply to comment #9)
> Yes, I can confirm new dmesg still contain mentioned  value:
> 
> root@slax:~# cat nv92_dmesg_new_1.log | grep 0x1000f010
> nouveau E[  PGRAPH][0000:05:00.0] ch 4 [0x00178fb000 mplayer[4559]] subc 3
> class 0x8297 mthd 0x1b0c data 0x1000f010
> nouveau E[  PGRAPH][0000:05:00.0] ch 4 [0x00178fb000 mplayer[4559]] subc 3
> class 0x8297 mthd 0x1b0c data 0x1000f010
> root@slax:~#

OK, and just to confirm... you're *sure* your new libvdpau_nouveau.so is being used? You can check by doing e.g. "strace -f -e open vdpauinfo" to see which file it's opening. Judging by your config log, it should be reading /usr/X11R7/lib/vdpau/libvdpau_nouveau.so.
Comment 13 Andrew Randrianasulu 2014-09-03 06:50:09 UTC
(In reply to comment #12)
> (In reply to comment #9)
> > Yes, I can confirm new dmesg still contain mentioned  value:
> > 
> > root@slax:~# cat nv92_dmesg_new_1.log | grep 0x1000f010
> > nouveau E[  PGRAPH][0000:05:00.0] ch 4 [0x00178fb000 mplayer[4559]] subc 3
> > class 0x8297 mthd 0x1b0c data 0x1000f010
> > nouveau E[  PGRAPH][0000:05:00.0] ch 4 [0x00178fb000 mplayer[4559]] subc 3
> > class 0x8297 mthd 0x1b0c data 0x1000f010
> > root@slax:~#
> 
> OK, and just to confirm... you're *sure* your new libvdpau_nouveau.so is
> being used? You can check by doing e.g. "strace -f -e open vdpauinfo" to see
> which file it's opening. Judging by your config log, it should be reading
> /usr/X11R7/lib/vdpau/libvdpau_nouveau.so.

guest@slax:~/New_hdd$ strace -f -e open vdpauinfo
open("/etc/ld.so.cache", O_RDONLY|O_CLOEXEC) = 3
open("/usr/X11R7/lib/libX11.so.6", O_RDONLY|O_CLOEXEC) = 3
open("/usr/X11R7/lib/libvdpau.so.1", O_RDONLY|O_CLOEXEC) = 3
open("/usr/lib/libstdc++.so.6", O_RDONLY|O_CLOEXEC) = 3
open("/lib/libm.so.6", O_RDONLY|O_CLOEXEC) = 3
open("/usr/lib/libgcc_s.so.1", O_RDONLY|O_CLOEXEC) = 3
open("/lib/libc.so.6", O_RDONLY|O_CLOEXEC) = 3
open("/usr/X11R7/lib/libxcb.so.1", O_RDONLY|O_CLOEXEC) = 3
open("/usr/X11R7/lib/libXau.so.6", O_RDONLY|O_CLOEXEC) = 3
open("/usr/X11R7/lib/libXdmcp.so.6", O_RDONLY|O_CLOEXEC) = 3
open("/lib/libdl.so.2", O_RDONLY|O_CLOEXEC) = 3
open("/usr/X11R7/lib/libXext.so.6", O_RDONLY|O_CLOEXEC) = 3
open("/home/guest/.Xauthority", O_RDONLY) = 4
display: :0   screen: 0
open("/usr/X11R7/lib/vdpau/libvdpau_nouveau.so.1", O_RDONLY|O_CLOEXEC) = 4
open("/etc/ld.so.cache", O_RDONLY|O_CLOEXEC) = 4
open("/usr/X11R7/lib/libX11-xcb.so.1", O_RDONLY|O_CLOEXEC) = 4
open("/usr/X11R7/lib/libxcb-dri2.so.0", O_RDONLY|O_CLOEXEC) = 4
open("/usr/lib/libexpat.so.1", O_RDONLY|O_CLOEXEC) = 4
open("/usr/X11R7/lib/libdrm_nouveau.so.2", O_RDONLY|O_CLOEXEC) = 4
open("/usr/lib/libz.so.1", O_RDONLY|O_CLOEXEC) = 4
open("/lib/libpthread.so.0", O_RDONLY|O_CLOEXEC) = 4
open("/usr/lib/libncurses.so.5", O_RDONLY|O_CLOEXEC) = 4
open("/usr/lib/libelf.so.0", O_RDONLY|O_CLOEXEC) = 4
open("/usr/X11R7/lib/libdrm_radeon.so.1", O_RDONLY|O_CLOEXEC) = 4
open("/usr/X11R7/lib/libdrm.so.2", O_RDONLY|O_CLOEXEC) = 4
open("/lib/librt.so.1", O_RDONLY|O_CLOEXEC) = 4
open("/usr/lib/libLLVM-3.4.so", O_RDONLY|O_CLOEXEC) = 4
open("/dev/dri/card0", O_RDWR)          = 4
open("/etc/ld.so.cache", O_RDONLY|O_CLOEXEC) = 5
open("/lib/tls/i686/sse2/libudev.so.1", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory)
open("/lib/tls/i686/libudev.so.1", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory)
open("/lib/tls/sse2/libudev.so.1", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory)
open("/lib/tls/libudev.so.1", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory)
open("/lib/i686/sse2/libudev.so.1", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory)
open("/lib/i686/libudev.so.1", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory)
open("/lib/sse2/libudev.so.1", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory)
open("/lib/libudev.so.1", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory)
open("/usr/lib/tls/i686/sse2/libudev.so.1", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory)
open("/usr/lib/tls/i686/libudev.so.1", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory)
open("/usr/lib/tls/sse2/libudev.so.1", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory)
open("/usr/lib/tls/libudev.so.1", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory)
open("/usr/lib/i686/sse2/libudev.so.1", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory)
open("/usr/lib/i686/libudev.so.1", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory)
open("/usr/lib/sse2/libudev.so.1", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory)
open("/usr/lib/libudev.so.1", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory)
open("/etc/ld.so.cache", O_RDONLY|O_CLOEXEC) = 5
open("/lib/libudev.so.0", O_RDONLY|O_CLOEXEC) = 5
open("/etc/udev/udev.conf", O_RDONLY|O_LARGEFILE) = 5
open("/sys/devices/pci0000:00/0000:00:0e.0/0000:05:00.0/uevent", O_RDONLY|O_LARGEFILE) = 5
open("/etc/ld.so.cache", O_RDONLY|O_CLOEXEC) = 6
open("/usr/lib/libtxc_dxtn.so", O_RDONLY|O_CLOEXEC) = 6
open("/sys/devices/system/cpu/online", O_RDONLY|O_CLOEXEC) = 6
API version: 1
Information string: G3DVL VDPAU Driver Shared Library version 1.0
[snip]
Comment 14 ppine 2014-10-12 11:51:36 UTC
(In reply to Andrew Randrianasulu from comment #10)
> Created attachment 105647 [details]
> dmesg again, after some system use (but ends up with usual hang with mplayer)

Same problem here when trying to use vdpau on a G92 card:

[   97.501981] nouveau E[  PGRAPH][0000:01:00.0] TRAP_PROP - TP 0 - RT_FAULT - Address 0020406c40
[   97.501991] nouveau E[  PGRAPH][0000:01:00.0] TRAP_PROP - TP 0 - e0c: 00000000, e18: 00000000, e1c: 00500010, e20: 00001100, e24: 00030000
[   97.502006] nouveau E[  PGRAPH][0000:01:00.0] TRAP_PROP - TP 1 - RT_FAULT - Address 0020405000
[   97.502010] nouveau E[  PGRAPH][0000:01:00.0] TRAP_PROP - TP 1 - e0c: 00000000, e18: 00000000, e1c: 00340000, e20: 00001100, e24: 00030000
[   97.502024] nouveau E[  PGRAPH][0000:01:00.0] TRAP_PROP - TP 2 - RT_FAULT - Address 0020405040
[   97.502028] nouveau E[  PGRAPH][0000:01:00.0] TRAP_PROP - TP 2 - e0c: 00000000, e18: 00000000, e1c: 00340010, e20: 00001100, e24: 00030000
[   97.502041] nouveau E[  PGRAPH][0000:01:00.0] TRAP_PROP - TP 3 - RT_FAULT - Address 0020405080
[   97.502045] nouveau E[  PGRAPH][0000:01:00.0] TRAP_PROP - TP 3 - e0c: 00000000, e18: 00000000, e1c: 00340020, e20: 00001100, e24: 00030000
[   97.502058] nouveau E[  PGRAPH][0000:01:00.0] TRAP_PROP - TP 4 - RT_FAULT - Address 00204050c0
[   97.502062] nouveau E[  PGRAPH][0000:01:00.0] TRAP_PROP - TP 4 - e0c: 00000000, e18: 00000000, e1c: 00340030, e20: 00001100, e24: 00030000
[   97.502075] nouveau E[  PGRAPH][0000:01:00.0] TRAP_PROP - TP 5 - RT_FAULT - Address 0020405c40
[   97.502079] nouveau E[  PGRAPH][0000:01:00.0] TRAP_PROP - TP 5 - e0c: 00000000, e18: 00000000, e1c: 00400010, e20: 00001100, e24: 00030000
[   97.502093] nouveau E[  PGRAPH][0000:01:00.0] TRAP_PROP - TP 6 - RT_FAULT - Address 0020405c80
[   97.502097] nouveau E[  PGRAPH][0000:01:00.0] TRAP_PROP - TP 6 - e0c: 00000000, e18: 00000000, e1c: 00400020, e20: 00001100, e24: 00030000
[   97.502111] nouveau E[  PGRAPH][0000:01:00.0] TRAP_PROP - TP 7 - RT_FAULT - Address 0020406c00
[   97.502115] nouveau E[  PGRAPH][0000:01:00.0] TRAP_PROP - TP 7 - e0c: 00000000, e18: 00000000, e1c: 00500000, e20: 00001100, e24: 00030000
[   97.502120] nouveau E[  PGRAPH][0000:01:00.0] ch 5 [0x001f764000 mplayer[4940]] subc 3 class 0x8297 mthd 0x1b0c data 0x1000f010
[   97.502135] nouveau E[     PFB][0000:01:00.0] trapped write at 0x0020405c40 on channel 0x0001f764 [mplayer[4940]] PGRAPH/PROP/RT0 reason: PAGE_NOT_PRESENT
[   97.524623] nouveau E[  PGRAPH][0000:01:00.0] TRAP_PROP - TP 0 - RT_FAULT - Address 00204f6f40
[   97.524633] nouveau E[  PGRAPH][0000:01:00.0] TRAP_PROP - TP 0 - e0c: 00000000, e18: 00000000, e1c: 0f530018, e20: 00001100, e24: 00030000
[   97.524643] nouveau E[  PGRAPH][0000:01:00.0] TRAP_PROP - TP 1 - RT_FAULT - Address 002040dc40
[   97.524649] nouveau E[  PGRAPH][0000:01:00.0] TRAP_PROP - TP 1 - e0c: 00000000, e18: 00000000, e1c: 00800010, e20: 00001100, e24: 00030000
[   97.524658] nouveau E[  PGRAPH][0000:01:00.0] TRAP_PROP - TP 2 - RT_FAULT - Address 002040c000
[   97.524663] nouveau E[  PGRAPH][0000:01:00.0] TRAP_PROP - TP 2 - e0c: 00000000, e18: 00000000, e1c: 00640000, e20: 00001100, e24: 00030000
[   97.524673] nouveau E[  PGRAPH][0000:01:00.0] TRAP_PROP - TP 3 - RT_FAULT - Address 002040c040
[   97.524677] nouveau E[  PGRAPH][0000:01:00.0] TRAP_PROP - TP 3 - e0c: 00000000, e18: 00000000, e1c: 00640010, e20: 00001100, e24: 00030000
[   97.524691] nouveau E[  PGRAPH][0000:01:00.0] TRAP_PROP - TP 4 - RT_FAULT - Address 002040c080
[   97.524696] nouveau E[  PGRAPH][0000:01:00.0] TRAP_PROP - TP 4 - e0c: 00000000, e18: 00000000, e1c: 00640020, e20: 00001100, e24: 00030000
[   97.524709] nouveau E[  PGRAPH][0000:01:00.0] TRAP_PROP - TP 5 - RT_FAULT - Address 002040c0c0
[   97.524715] nouveau E[  PGRAPH][0000:01:00.0] TRAP_PROP - TP 5 - e0c: 00000000, e18: 00000000, e1c: 00640030, e20: 00001100, e24: 00030000
[   97.524726] nouveau E[  PGRAPH][0000:01:00.0] TRAP_PROP - TP 6 - RT_FAULT - Address 002040cc40
[   97.524730] nouveau E[  PGRAPH][0000:01:00.0] TRAP_PROP - TP 6 - e0c: 00000000, e18: 00000000, e1c: 00700010, e20: 00001100, e24: 00030000
[   97.524742] nouveau E[  PGRAPH][0000:01:00.0] TRAP_PROP - TP 7 - RT_FAULT - Address 002040cc80
[   97.524746] nouveau E[  PGRAPH][0000:01:00.0] TRAP_PROP - TP 7 - e0c: 00000000, e18: 00000000, e1c: 00700020, e20: 00001100, e24: 00030000
[   97.524751] nouveau E[  PGRAPH][0000:01:00.0] ch 5 [0x001f764000 mplayer[4940]] subc 3 class 0x8297 mthd 0x1b0c data 0x1000f010
[   97.524764] nouveau E[     PFB][0000:01:00.0] trapped write at 0x002040c0c0 on channel 0x0001f764 [mplayer[4940]] PGRAPH/PROP/RT0 reason: PAGE_NOT_PRESENT
[   97.547931] nouveau E[  PGRAPH][0000:01:00.0] TRAP_PROP - TP 0 - RT_FAULT - Address 00411a0080
[   97.547940] nouveau E[  PGRAPH][0000:01:00.0] TRAP_PROP - TP 0 - e0c: 00000000, e18: 00000000, e1c: 0ba00020, e20: 00001100, e24: 00030000
[   97.547947] nouveau E[  PGRAPH][0000:01:00.0] TRAP_PROP - TP 1 - RT_FAULT - Address 00411a00c0
[   97.547951] nouveau E[  PGRAPH][0000:01:00.0] TRAP_PROP - TP 1 - e0c: 00000000, e18: 00000000, e1c: 0ba00030, e20: 00001100, e24: 00030000
[   97.547957] nouveau E[  PGRAPH][0000:01:00.0] TRAP_PROP - TP 2 - RT_FAULT - Address 00411a1040
[   97.547961] nouveau E[  PGRAPH][0000:01:00.0] TRAP_PROP - TP 2 - e0c: 00000000, e18: 00000000, e1c: 0bb00010, e20: 00001100, e24: 00030000
[   97.547967] nouveau E[  PGRAPH][0000:01:00.0] TRAP_PROP - TP 3 - RT_FAULT - Address 00411a1080
[   97.547971] nouveau E[  PGRAPH][0000:01:00.0] TRAP_PROP - TP 3 - e0c: 00000000, e18: 00000000, e1c: 0bb00020, e20: 00001100, e24: 00030000
[   97.547977] nouveau E[  PGRAPH][0000:01:00.0] TRAP_PROP - TP 4 - RT_FAULT - Address 00411a10c0
[   97.547981] nouveau E[  PGRAPH][0000:01:00.0] TRAP_PROP - TP 4 - e0c: 00000000, e18: 00000000, e1c: 0bb00030, e20: 00001100, e24: 00030000
[   97.547987] nouveau E[  PGRAPH][0000:01:00.0] TRAP_PROP - TP 5 - RT_FAULT - Address 00411a2040
[   97.547990] nouveau E[  PGRAPH][0000:01:00.0] TRAP_PROP - TP 5 - e0c: 00000000, e18: 00000000, e1c: 0bc00010, e20: 00001100, e24: 00030000
[   97.547997] nouveau E[  PGRAPH][0000:01:00.0] TRAP_PROP - TP 6 - RT_FAULT - Address 00411a0000
[   97.548000] nouveau E[  PGRAPH][0000:01:00.0] TRAP_PROP - TP 6 - e0c: 00000000, e18: 00000000, e1c: 0ba00000, e20: 00001100, e24: 00030000
[   97.548007] nouveau E[  PGRAPH][0000:01:00.0] TRAP_PROP - TP 7 - RT_FAULT - Address 00411a0040
[   97.548010] nouveau E[  PGRAPH][0000:01:00.0] TRAP_PROP - TP 7 - e0c: 00000000, e18: 00000000, e1c: 0ba00010, e20: 00001100, e24: 00030000
[   97.548014] nouveau E[  PGRAPH][0000:01:00.0] ch 5 [0x001f764000 mplayer[4940]] subc 3 class 0x8297 mthd 0x1b0c data 0x1000f010
[   97.548023] nouveau E[     PFB][0000:01:00.0] trapped write at 0x00411a00c0 on channel 0x0001f764 [mplayer[4940]] PGRAPH/PROP/RT0 reason: PAGE_NOT_PRESENT
[   98.012191] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.014850] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.017503] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.019831] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.022160] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.024484] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fc28
[   98.027134] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.029456] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.032105] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.034430] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.036757] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.039407] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.042056] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.044706] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.047028] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.049349] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.051676] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000004 FAULT at 0x00fd0c
[   98.053997] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.056647] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.059296] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.061946] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.063939] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.065933] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.068587] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000004 FAULT at 0x00fd0c
[   98.071237] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.073559] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.076208] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.078530] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.080851] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.083173] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.085531] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000004 FAULT at 0x00fd0c
[   98.088181] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.090175] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.092169] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.094490] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.096812] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.099461] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.102116] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000004 FAULT at 0x00fd0c
[   98.104765] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.107415] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.109737] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.112387] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.114709] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.117358] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.119684] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.122009] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.124335] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.126657] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.128979] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.131300] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.133622] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.136278] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.138271] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.140593] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fc28
[   98.142915] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.145564] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.148214] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.150863] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.153518] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.155839] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.157833] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.160483] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.162804] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.165454] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.168104] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.170758] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.173408] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.176057] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.178379] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.180373] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fc28
[   98.183022] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.185677] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000004 FAULT at 0x00fd0c
[   98.187999] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.190323] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.192645] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.195294] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.197616] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fc28
[   98.199937] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.202264] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000004 FAULT at 0x00fd0c
[   98.204914] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.207563] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.210213] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.212534] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.214856] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.216849] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.218848] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000004 FAULT at 0x00fd0c
[   98.221499] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.223497] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.226150] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.228800] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.231450] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.233772] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.236100] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.238422] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.240744] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.243066] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.245716] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.248037] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.250031] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.252686] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000004 FAULT at 0x00fd0c
[   98.255007] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fc28
[   98.257329] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fc28
[   98.259979] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.262628] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.265278] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.267602] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.269928] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fc28
[   98.272578] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.274900] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.277222] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.279543] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.281865] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.284515] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000000 FAULT at 0x00fd0c
[   98.285197] nouveau E[    PBUS][0000:01:00.0] MMIO read of 0x00000004 FAULT at 0x00fd0c
[  117.790892] nouveau E[mplayer[4940]] failed to idle channel 0xcccc0002 [mplayer[4940]]
[  132.790675] nouveau E[mplayer[4940]] failed to idle channel 0xcccc0002 [mplayer[4940]]
[  134.790691] nouveau E[   PFIFO][0000:01:00.0] channel 7 [mplayer[4940]] unload timeout
[  134.790791] nouveau E[     PFB][0000:01:00.0] trapped write at 0x0020312074 on channel 0x0001f6cd [unknown] PFIFO/PFIFO_READ/SEMAPHORE reason: PAGE_NOT_PRESENT
[  134.800392] nouveau E[     PFB][0000:01:00.0] trapped read at 0x0000000000 on channel 0x0001f6cd [unknown] PVP/PVP/0f reason: DMAOBJ_LIMIT
[  165.117942] nouveau E[     DRM] GPU lockup - switching to software fbcon
[  240.385122] nouveau E[Xorg.bin[1541]] failed to idle channel 0xcccc0001 [Xorg.bin[1541]]

Then the complete box just freezes up after a while.
Comment 15 Sébastien P. 2015-03-21 17:46:58 UTC
> Same problem here when trying to use vdpau on a G92 card

Same thing on NV96:
Mar 20 23:36:15 fixe kernel: nouveau E[  PGRAPH][0000:02:00.0] TRAP_PROP - TP 0 - RT_FAULT - Address 00204150c0
Mar 20 23:36:15 fixe kernel: nouveau E[  PGRAPH][0000:02:00.0] TRAP_PROP - TP 0 - e0c: 00000000, e18: 00000000, e1c: 00340030, e20: 00001100, e24: 00030000
Mar 20 23:36:15 fixe kernel: nouveau E[  PGRAPH][0000:02:00.0] TRAP_PROP - TP 1 - RT_FAULT - Address 0020415080
Mar 20 23:36:15 fixe kernel: nouveau E[  PGRAPH][0000:02:00.0] TRAP_PROP - TP 1 - e0c: 00000000, e18: 00000000, e1c: 00340020, e20: 00001100, e24: 00030000
Mar 20 23:36:15 fixe kernel: nouveau E[  PGRAPH][0000:02:00.0] ch 5 [0x001f7a8000 mplayer[31053]] subc 3 class 0x8297 mthd 0x1b0c data 0x1000f010
Mar 20 23:36:15 fixe kernel: nouveau E[     PFB][0000:02:00.0] trapped write at 0x0020415080 on channel 0x0001f7a8 [mplayer[31053]] PGRAPH/PROP/RT0 reason: PAGE_NOT_PRESENT
Mar 20 23:36:15 fixe kernel: nouveau E[     PFB][0000:02:00.0] trapped write at 0x00204b7c40 on channel 0x0001f7a8 [mplayer[31053]] PGRAPH/PROP/RT0 reason: PAGE_NOT_PRESENT
Mar 20 23:36:15 fixe kernel: nouveau E[  PGRAPH][0000:02:00.0] TRAP_PROP - TP 0 - RT_FAULT - Address 00204990c0
Mar 20 23:36:15 fixe kernel: nouveau E[  PGRAPH][0000:02:00.0] TRAP_PROP - TP 0 - e0c: 00000000, e18: 00000000, e1c: 08740030, e20: 00001100, e24: 00030000
Mar 20 23:36:15 fixe kernel: nouveau E[  PGRAPH][0000:02:00.0] TRAP_PROP - TP 1 - RT_FAULT - Address 0020499080
Mar 20 23:36:15 fixe kernel: nouveau E[  PGRAPH][0000:02:00.0] TRAP_PROP - TP 1 - e0c: 00000000, e18: 00000000, e1c: 08740020, e20: 00001100, e24: 00030000
Mar 20 23:36:15 fixe kernel: nouveau E[  PGRAPH][0000:02:00.0] ch 5 [0x001f7a8000 mplayer[31053]] subc 3 class 0x8297 mthd 0x1b0c data 0x1000f010
Mar 20 23:36:15 fixe kernel: nouveau E[     PFB][0000:02:00.0] trapped write at 0x0020499080 on channel 0x0001f7a8 [mplayer[31053]] PGRAPH/PROP/RT0 reason: PAGE_NOT_PRESENT
Mar 20 23:36:15 fixe kernel: nouveau E[     PFB][0000:02:00.0] trapped write at 0x0020533c40 on channel 0x0001f7a8 [mplayer[31053]] PGRAPH/PROP/RT0 reason: PAGE_NOT_PRESENT
Mar 20 23:36:15 fixe kernel: nouveau E[  PGRAPH][0000:02:00.0] TRAP_PROP - TP 0 - RT_FAULT - Address 0040840080
Mar 20 23:36:15 fixe kernel: nouveau E[  PGRAPH][0000:02:00.0] TRAP_PROP - TP 0 - e0c: 00000000, e18: 00000000, e1c: 05ac0020, e20: 00001100, e24: 00030000
Mar 20 23:36:15 fixe kernel: nouveau E[  PGRAPH][0000:02:00.0] TRAP_PROP - TP 1 - RT_FAULT - Address 00408400c0
Mar 20 23:36:15 fixe kernel: nouveau E[  PGRAPH][0000:02:00.0] TRAP_PROP - TP 1 - e0c: 00000000, e18: 00000000, e1c: 05ac0030, e20: 00001100, e24: 00030000
Mar 20 23:36:15 fixe kernel: nouveau E[  PGRAPH][0000:02:00.0] ch 5 [0x001f7a8000 mplayer[31053]] subc 3 class 0x8297 mthd 0x1b0c data 0x1000f010
Mar 20 23:36:15 fixe kernel: nouveau E[     PFB][0000:02:00.0] trapped write at 0x00408400c0 on channel 0x0001f7a8 [mplayer[31053]] PGRAPH/PROP/RT0 reason: PAGE_NOT_PRESENT
Mar 20 23:36:15 fixe kernel: nouveau E[     PFB][0000:02:00.0] trapped write at 0x00408df000 on channel 0x0001f7a8 [mplayer[31053]] PGRAPH/PROP/RT0 reason: PAGE_NOT_PRESENT

I have also notice that It happens when the load average of my system is quite hight (compilation, backup).
=> Hang happenned during the playback.

Or when I openned two files too quickly.
=> Hang happenned immedialety.

mplayer 1.2 (20130729)
kernel 3.18.7
mesa 10.3.7
nvidia-firmware 340.32
xf86-video-nouveau 1.0.11
libva-vdpau-driver 0.7.4-r2
libvdpau 0.9
X 1.16.4
Comment 16 Ilia Mirkin 2015-03-21 18:13:43 UTC
(In reply to Sébastien P. from comment #15)
> > Same problem here when trying to use vdpau on a G92 card
> 
> Same thing on NV96:

No, most likely a very different thing on your G96. You're getting some errors. G92's apparently insta-hang when doing anything with vdpau.
Comment 17 Sébastien P. 2015-03-21 18:53:29 UTC
Alright. Logs and result seem quite similar to me.
Do you want a separate bug and more information?
Comment 18 Ilia Mirkin 2015-03-21 19:04:49 UTC
(In reply to Sébastien P. from comment #17)
> Alright. Logs and result seem quite similar to me.
> Do you want a separate bug and more information?

I think your issue is more akin to the one in bug 89572. If you're unsure, file a fresh bug, and we can try to classify it.
Comment 19 Sébastien P. 2015-03-21 19:55:42 UTC
Perhaps.

Theses logs appear when the system hang.
On bug 89572, the “[…] RT_FAULT […]” appears immediately but the system does not hang (did not try to stress the computer to see it would hand).
Comment 20 Andrew Randrianasulu 2015-10-19 14:56:05 UTC
Created attachment 118979 [details]
hwtest stdout log

So, I tried hwtest from envytools (commit 9c8978886d8c9e57ec7e8c167a7b6aa7ab35dc40), of course without nouveau kernel module loaded, as root. Looks like vp2 test failed?
Comment 21 Weedy 2016-04-10 06:51:58 UTC
I have a G92GLM [Quadro FX 3600M].
I use MPV with vdpau and opengl rendering A LOT.

Is this still broken? I want to move off the binary driver but I need this to not be broken. Also I hope freq scaling works.
Comment 22 Andrew Randrianasulu 2016-05-05 19:22:51 UTC
(In reply to Weedy from comment #21)
> I have a G92GLM [Quadro FX 3600M].
> I use MPV with vdpau and opengl rendering A LOT.
> 
> Is this still broken? I want to move off the binary driver but I need this
> to not be broken. Also I hope freq scaling works.

For me it still broken, but you can test and see if it works  for you.
Comment 23 Andrew Randrianasulu 2016-05-05 19:32:47 UTC
Some time ago (~aug, 2015) I did mmiotrace of binary driver to see if it can drive my specific card here  for h264 playback. It worked, so I uploaded xz compressed trace to google drive and forgot about it. Hopefully everyone still can download and decompress it?

https://drive.google.com/file/d/0B84cgTVGGfZaVUZoN1hxNzg4Y00/view?usp=sharing
Comment 24 Andrew Randrianasulu 2017-11-25 12:24:46 UTC
In kernel 4.14 workaround was added for disabling h264 decoding until root cause of this bug will be uncovered and fixed (may be never):

https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?h=v4.14&id=194d68dd051c2dd5ac2b522ae16100e774e8d869

drm/nouveau/bsp/g92: disable by default


Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.