[Nouveau] nouveau errors failure lockups

Thu Sep 26 16:43:19 PDT 2013

On 15/09/13 10:38, P NIKOLIC wrote:
> Hi .
> 
Hi Pete

> For some time now i have had problems with video on this system  .
> 
> from dmesg iget the following 
> 
>  6.408406] nouveau  [  DEVICE][0000:01:00.0] BOOT0  : 0x0a5000a2
> [    6.408411] nouveau  [  DEVICE][0000:01:00.0] Chipset: GT216 (NVA5)
> [    6.408414] nouveau  [  DEVICE][0000:01:00.0] Family : NV50
> [    6.409206] nouveau  [   VBIOS][0000:01:00.0] checking PRAMIN for image...
> [    6.513939] nouveau  [   VBIOS][0000:01:00.0] ... appears to be valid
> [    6.513941] nouveau  [   VBIOS][0000:01:00.0] using image from PRAMIN
> [    6.514101] nouveau  [   VBIOS][0000:01:00.0] BIT signature found
> [    6.514104] nouveau  [   VBIOS][0000:01:00.0] version 70.16.39.00.0f
> [    6.514469] nouveau  [     PFB][0000:01:00.0] RAM type: DDR2
> [    6.514471] nouveau  [     PFB][0000:01:00.0] RAM size: 1024 MiB
> [    6.514473] nouveau  [     PFB][0000:01:00.0]    ZCOMP: 1920 tags
> 
> 
>  7.188561] nouveau  [     DRM] VRAM: 1024 MiB
> [    7.188567] nouveau  [     DRM] GART: 512 MiB
> [    7.188575] nouveau  [     DRM] TMDS table version 2.0
> [    7.188577] nouveau  [     DRM] DCB version 4.0
> [    7.188581] nouveau  [     DRM] DCB outp 00: 01000302 00020030
> [    7.188583] nouveau  [     DRM] DCB outp 01: 02000300 00000000
> [    7.188586] nouveau  [     DRM] DCB outp 02: 01032310 00000000
> [    7.188587] nouveau  [     DRM] DCB outp 03: 02021362 00020010
> [    7.188589] nouveau  [     DRM] DCB conn 00: 00001030
> [    7.188592] nouveau  [     DRM] DCB conn 01: 00002161
> [    7.188594] nouveau  [     DRM] DCB conn 02: 00000200
> [    7.197040] [drm] Supports vblank timestamp caching Rev 1 (10.10.2010).
> [    7.197042] [drm] No driver support for vblank timestamp query.
> [    7.197188] nouveau  [     DRM] 3 available performance level(s)
> [    7.197191] nouveau  [     DRM] 0: core 135MHz shader 270MHz memory 135MHz
>  voltage 850mV [    7.197194] nouveau  [     DRM] 1: core 405MHz shader 810MHz
>  memory 324MHz voltage 900mV [    7.197197] nouveau  [     DRM] 3: core 625MHz
>  shader 1360MHz memory 400MHz voltage 1000mV [    7.197200] nouveau  [     DRM] c:
>  core 405MHz shader 810MHz memory 324MHz voltage 900mV fanspeed 32% [    7.205654]
>  nouveau  [     DRM] MM: using COPY for buffer copies [    7.267852] nouveau
>  [     DRM] allocated 1280x1024 fb: 0x70000, bo ffff8800c277e000 [    7.267939]
>  fbcon: nouveaufb (fb0) is primary device
> 
> 
>  730.397758] nouveau E[   PFIFO][0000:01:00.0] DMA_PUSHER - ch 2 [X[274]] get
>  0x00200294bc put 0x0020029508 ib_get 0x00000255 ib_put 0x00000256 state 0x80004290
>  (err: INVALID_CMD) push 0x00400040 [  730.397937] nouveau E[  PGRAPH][0000:01:00.0]
>  DATA_ERROR INVALID_VALUE [  730.397946] nouveau E[  PGRAPH][0000:01:00.0]
>  DATA_ERROR [  730.397955] nouveau E[  PGRAPH][0000:01:00.0] ch 2 [0x003fb3a000
>  X[274]] subc 2 class 0x502d mthd 0x0284 data 0xffffffff [  730.397982] nouveau
>  E[  PGRAPH][0000:01:00.0] DATA_ERROR INVALID_VALUE [  730.397987] nouveau
>  E[  PGRAPH][0000:01:00.0]  DATA_ERROR [  730.397999] nouveau
>  E[  PGRAPH][0000:01:00.0] ch 2 [0x003fb3a000 X[274]] subc 2 class 0x502d mthd
>  0x0288 data 0xffffffff [  730.398023] nouveau E[  PGRAPH][0000:01:00.0] DATA_ERROR
>  INVALID_VALUE [  730.398027] nouveau E[  PGRAPH][0000:01:00.0]  DATA_ERROR
>  [  730.398033] nouveau E[  PGRAPH][0000:01:00.0] ch 2 [0x003fb3a000 X[274]] subc 2
>  class 0x502d mthd 0x028c data 0xffffffff [ 1468.048083] nouveau
>  E[   PFIFO][0000:01:00.0] DMA_PUSHER - ch 2 [X[274]] get 0x00200333f0 put
>  0x002003343c ib_get 0x00000098 ib_put 0x00000099 state 0x80004290 (err:
>  INVALID_CMD) push 0x00400040 [ 1468.048262] nouveau E[  PGRAPH][0000:01:00.0]
>  DATA_ERROR INVALID_VALUE [ 1468.048270] nouveau E[  PGRAPH][0000:01:00.0]
>  DATA_ERROR [ 1468.048279] nouveau E[  PGRAPH][0000:01:00.0] ch 2 [0x003fb3a000
>  X[274]] subc 2 class 0x502d mthd 0x0280 data 0xffffffff [ 1468.048306] nouveau
>  E[  PGRAPH][0000:01:00.0] DATA_ERROR INVALID_VALUE [ 1468.048311] nouveau
>  E[  PGRAPH][0000:01:00.0]  DATA_ERROR [ 1468.048319] nouveau
>  E[  PGRAPH][0000:01:00.0] ch 2 [0x003fb3a000 X[274]] subc 2 class 0x502d mthd
>  0x0284 data 0xffffffff [ 1468.048339] nouveau E[  PGRAPH][0000:01:00.0] DATA_ERROR
>  INVALID_VALUE [ 1468.048349] nouveau E[  PGRAPH][0000:01:00.0]  DATA_ERROR
>  [ 1468.048355] nouveau E[  PGRAPH][0000:01:00.0] ch 2 [0x003fb3a000 X[274]] subc 2
>  class 0x502d mthd 0x0288 data 0xffffffff [ 1468.048376] nouveau
>  E[  PGRAPH][0000:01:00.0] DATA_ERROR INVALID_VALUE [ 1468.048385] nouveau
>  E[  PGRAPH][0000:01:00.0]  DATA_ERROR [ 1468.048391] nouveau
>  E[  PGRAPH][0000:01:00.0] ch 2 [0x003fb3a000 X[274]] subc 2 class 0x502d mthd
>  0x028c data 0xffffffff [ 1488.832401] nouveau E[   PFIFO][0000:01:00.0] DMA_PUSHER
>  - ch 2 [X[274]] get 0x002003226c put 0x0020032324 ib_get 0x000001ca ib_put
>  0x000001cb state 0x80000020 (err: INVALID_CMD) push 0x00400040
> 
> 
> this leads to a completely locked system the onoy sucsessful way out is a hard reset
> 
pm-suspend may help as well :P

But on the serious side, the issue (according to the log provided) is
that the command buffer(pushbuf) is getting corrupted thus with a bit
more time hell breaks loose.

Do you have a reliable way to reproduce the issue ?
> the system is 
>  Linux 7-of-9 3.10.10-1-ARCH #1 SMP PREEMPT Fri Aug 30 11:30:06 CEST 2013 x86_64
> GNU/Linux
> 
> AMD Phenome quad core  4Gb ram  GT216 video card 1Gb  
> latest Arch linux fully updated KDE 4.11.1   
> 
> This has been happening for the last 6 kernel upgrades  
> 
If that is truly the case I would recommend bisecting the issue :)

To keep a track of all the info please file a bug [1]

Cheers
Emil

[1] http://nouveau.freedesktop.org/wiki/Bugs/

> Thanks for any insight 
> 
> 
> Pete .
> 
> 