Bug 70566

Summary: [NVD9] hard lock-ups entire system at random moments
Product: xorg Reporter: Ronald <ronald645>
Component: Driver/nouveauAssignee: Nouveau Project <nouveau>
Status: RESOLVED INVALID QA Contact: Xorg Project Team <xorg-team>
Severity: normal    
Priority: medium    
Version: git   
Hardware: Other   
OS: All   
Whiteboard:
i915 platform: i915 features:

Description Ronald 2013-10-17 07:37:47 UTC 
My card is a GF119 (NVD9). A relatively new card on an aged board (Asus A8N-E).

It seems to lock-up at random moments, Magic SysRq does nothing. Only a hard reboot is possible.

Sometimes it happens when the contents of the screen is changing, sometimes not. I found it to have locked after an idle period a couple of times.

I'm currently testing the nvidia proprietary blob after swapping out RAM (which was not it). This all started after I swapped this card for another broken nVidia card. But I figured it might be coincidence.

If the nvidia proprietary driver proofs to not suffer from this issue. I will try to test the nouveau driver with as many features disabled (e.g. PCIE) and see if that helps.

Any suggestions are really welcome.

Card specs below.

16-10-2013 20:58 .info ::: nouveau  [  DEVICE][0000:01:00.0] BOOT0  : 0x0d90a0a1
16-10-2013 20:58 .info ::: nouveau  [  DEVICE][0000:01:00.0] Chipset: GF119 (NVD9)
16-10-2013 20:58 .info ::: nouveau  [  DEVICE][0000:01:00.0] Family : NVD0
16-10-2013 20:58 .info ::: nouveau  [   VBIOS][0000:01:00.0] checking PRAMIN for image...
16-10-2013 20:58 .info ::: nouveau  [   VBIOS][0000:01:00.0] ... appears to be valid
16-10-2013 20:58 .info ::: nouveau  [   VBIOS][0000:01:00.0] using image from PRAMIN
16-10-2013 20:58 .info ::: nouveau  [   VBIOS][0000:01:00.0] BIT signature found
16-10-2013 20:58 .info ::: nouveau  [   VBIOS][0000:01:00.0] version 75.19.55.00.02
16-10-2013 20:58 .debug ::: nouveau 0000:01:00.0: irq 44 for MSI/MSI-X
16-10-2013 20:58 .info ::: nouveau  [     PMC][0000:01:00.0] MSI interrupts enabled
16-10-2013 20:58 .info ::: nouveau  [     PFB][0000:01:00.0] RAM type: DDR3
16-10-2013 20:58 .info ::: nouveau  [     PFB][0000:01:00.0] RAM size: 1024 MiB
16-10-2013 20:58 .info ::: nouveau  [     PFB][0000:01:00.0]    ZCOMP: 0 tags
16-10-2013 20:58 .info ::: nouveau  [  PTHERM][0000:01:00.0] FAN control: PWM
16-10-2013 20:58 .info ::: nouveau  [  PTHERM][0000:01:00.0] fan management: disabled
16-10-2013 20:58 .info ::: nouveau  [  PTHERM][0000:01:00.0] internal sensor: yes
16-10-2013 20:58 .info ::: [TTM] Zone  kernel: Available graphics memory: 509984 kiB
16-10-2013 20:58 .info ::: [TTM] Initializing pool allocator
16-10-2013 20:58 .info ::: [TTM] Initializing DMA pool allocator
16-10-2013 20:58 .info ::: nouveau  [     DRM] VRAM: 1024 MiB
16-10-2013 20:58 .info ::: nouveau  [     DRM] GART: 1048576 MiB
16-10-2013 20:58 .info ::: nouveau  [     DRM] TMDS table version 2.0
16-10-2013 20:58 .info ::: nouveau  [     DRM] DCB version 4.0
16-10-2013 20:58 .info ::: nouveau  [     DRM] DCB outp 00: 02000300 00000000
16-10-2013 20:58 .info ::: nouveau  [     DRM] DCB outp 01: 01000302 00020030
16-10-2013 20:58 .info ::: nouveau  [     DRM] DCB outp 02: 02011362 00020010
16-10-2013 20:58 .info ::: nouveau  [     DRM] DCB outp 03: 04022310 00000000
16-10-2013 20:58 .info ::: nouveau  [     DRM] DCB conn 00: 00001030
16-10-2013 20:58 .info ::: nouveau  [     DRM] DCB conn 01: 00002161
16-10-2013 20:58 .info ::: nouveau  [     DRM] DCB conn 02: 00000200
16-10-2013 20:58 .info ::: [drm] Supports vblank timestamp caching Rev 1 (10.10.2010).
16-10-2013 20:58 .info ::: [drm] No driver support for vblank timestamp query.
16-10-2013 20:58 .info ::: nouveau  [     DRM] 2 available performance level(s)
16-10-2013 20:58 .info ::: nouveau  [     DRM] 1: core 270MHz shader 540MHz memory 405MHz voltage 900mV
16-10-2013 20:58 .info ::: nouveau  [     DRM] 3: core 810MHz shader 1620MHz memory 500MHz voltage 1110mV
16-10-2013 20:58 .info ::: nouveau  [     DRM] c: core 270MHz shader 540MHz memory 405MHz voltage 900mV fanspeed 40%
16-10-2013 20:58 .info ::: nouveau  [     DRM] MM: using COPY0 for buffer copies
16-10-2013 20:58 .info ::: nouveau  [     DRM] allocated 1280x1024 fb: 0x60000, bo ffff88003dc5c000
16-10-2013 20:58 .info ::: fbcon: nouveaufb (fb0) is primary device
Comment 1 Ilia Mirkin 2013-10-17 19:49:40 UTC 
Did this ever work properly, or has it always failed? (Or is this too new of a setup?)

First order of business would be to try disabling MSI -- that will be off for the final 3.12 release (although will be making a come-back for 3.13 in all likelihood). Try booting with nouveau.config=NvMSI=0. [I assume you're using a 3.12-rc based on the messages you gave, but having a full dmesg is always preferable.]

Would it be possible to set up a serial console or netconsole to see if there are any further messages?
Comment 2 Ronald 2013-10-18 08:11:14 UTC 
This never worked, unfortunately. I got the card a few months ago.

When I got the card, MSI was disabled in nouveau. It then locked up as well. I was hoping this would fix it.

I know netconsole, I think I tried but that gave no results.

Will attach laptop and try again when I have some time.

But I'm testing the nvidia blob first to rule out a broken card, is that ok?
Comment 3 Ilia Mirkin 2015-11-07 08:04:42 UTC 
Does this still happen with recent kernels?
Comment 4 Ronald 2015-11-07 16:29:07 UTC 
The mobo died a while ago. So the problem is solved.

Thank you for your effort.

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.