NVRM: RmInitAdapter failed! (hardware problem?) [SOLVED]

graysky · 2013-10-06 08:16:28

Fairly recently, an older i686 box I have cannot load X but displays the bootup info, and non-graphical login prompt just fine. The video card is an nVidia 8400GS. After googling around for several of the dmesg errors, and not finding much of relevance, I am starting to wonder if the card itself is bad. Anyone with experience is welcomed to chime in with suggestions.

I have tried:
* Moving the card to another PCI slot.
* Both the linux/nvidia and linux-lts/nvidia-lts packages.
* Chapter 8 of the nvidia README recommends several kernel bootline mods, none of which work.

Links
Complete dmesg: http://pastebin.com/J0y8yKT6
Complete /var/log/Xorg.log: http://pastebin.com/iCkq9kM5

Relevant drivers:

% pacman -Qs nvidia
local/libvdpau 0.7-1
    Nvidia VDPAU library
local/nvidia 325.15-8
    NVIDIA drivers for linux
local/nvidia-libgl 325.15-1
    NVIDIA drivers libraries symlinks
local/nvidia-utils 325.15-1
    NVIDIA drivers utilities

Interestingly, I cannot read any info from the card itself which is in stark contrast to any of the other nvidia machines to which I have access:

% cat /proc/driver/nvidia/gpus/0/information 
Model: 		 GeForce 8400 GS
IRQ:   		 16
GPU UUID: 	 GPU-????????-????-????-????-????????????
Video BIOS: 	 ??.??.??.??.??
Bus Type: 	 PCIe
DMA Size: 	 40 bits
DMA Mask: 	 0xffffffffff
Bus Location: 	 0000:02.00.0

Specific errors in dmesg:

[    7.318404] nvidia: module license 'NVIDIA' taints kernel.
[    7.318413] Disabling lock debugging due to kernel taint
[    7.361296] vgaarb: device changed decodes: PCI:0000:02:00.0,olddecodes=io+mem,decodes=none:owns=io+mem
[    7.363264] [drm] Initialized nvidia-drm 0.0.0 20130102 for 0000:02:00.0 on minor 0
[    7.363285] NVRM: loading NVIDIA UNIX x86 Kernel Module  325.15  Wed Jul 31 18:18:15 PDT 2013
[    7.654888] input: ImPS/2 Generic Wheel Mouse as /devices/platform/i8042/serio1/input/input4
[    8.503106] Adding 1023996k swap on /dev/sda3.  Priority:-1 extents:1 across:1023996k FS
[    9.440825] EXT4-fs (sda2): mounted filesystem with ordered data mode. Opts: (null)
[    9.779428] EXT4-fs (sda6): mounted filesystem with ordered data mode. Opts: (null)
[   10.444258] systemd-journald[93]: Received request to flush runtime journal from PID 1
[   65.003814] nvidia 0000:02:00.0: irq 40 for MSI/MSI-X
[   69.922053] NVRM: RmInitAdapter failed! (0x12:0x2b:1830)
[   69.922070] NVRM: rm_init_adapter(0) failed
[   76.328374] nvidia 0000:02:00.0: irq 40 for MSI/MSI-X
[   80.706326] NVRM: RmInitAdapter failed! (0x12:0x2b:1830)
[   80.706343] NVRM: rm_init_adapter(0) failed
[   87.113316] nvidia 0000:02:00.0: irq 40 for MSI/MSI-X
[   91.491315] NVRM: RmInitAdapter failed! (0x12:0x2b:1830)
[   91.491332] NVRM: rm_init_adapter(0) failed
...

% grep EE Xorg.0.log

[   346.147] Initializing built-in extension MIT-SCREEN-SAVER
[   346.220] (EE) Failed to load module "nouveau" (module does not exist, 0)
[   346.222] (EE) Failed to load module "nv" (module does not exist, 0)
[   346.222] (EE) Failed to load module "vesa" (module does not exist, 0)
[   346.223] (EE) Failed to load module "modesetting" (module does not exist, 0)
[   346.223] (EE) Failed to load module "fbdev" (module does not exist, 0)
[   350.605] (EE) NVIDIA(0): The NVIDIA kernel module does not appear to be receiving
[   350.605] (EE) NVIDIA(0):     interrupts generated by the NVIDIA GPU at PCI:2:0:0. 
[   350.605] (EE) NVIDIA(0):     Please see Chapter 8: Common Problems in the README for
[   350.605] (EE) NVIDIA(0):     additional information.
[   350.605] (EE) NVIDIA(0): Failed to initialize the NVIDIA graphics device!
[   350.605] (EE) NVIDIA(0): Failing initialization of X screen 0
[   350.606] (EE) Screen(s) found, but none have a usable configuration.
[   350.606] (EE) 
[   350.606] (EE) no screens found(EE) 
[   350.606] (EE) 
[   350.606] (EE) Please also check the log file at "/var/log/Xorg.0.log" for additional information.
[   350.606] (EE) 
[   351.125] (EE) Server terminated with error (1). Closing log file.

Last edited by graysky (2013-10-06 15:54:40)

graysky · 2013-10-06 15:59:21

After exhausting various options in the BIOS (acting under the belief that something was interfering with the IRQ of the card), I thought I would try installing the nvidia-304-xx driver which ended up solving the problem. As of 06-Oct-2013, the Nvidia web site advises to use the 325.xx or 319.xx driver which is flat out wrong.

In short, using the legacy 304xx driver seems to be required for the 8400GS.

% cat /proc/driver/nvidia/gpus/0/information
Model: 		 GeForce 8400 GS
IRQ:   		 16
GPU UUID: 	 GPU-7f6a5800-9dce-8f7b-a3a8-9bc018f522ef
Video BIOS: 	 62.98.12.00.00
Bus Type: 	 PCIe
DMA Size: 	 40 bits
DMA Mask: 	 0xffffffffff
Bus Location: 	 0000:02.00.0

% pacman -Qs nvidia
local/libvdpau 0.7-1
    Nvidia VDPAU library
local/nvidia-304xx 304.108-7
    NVIDIA drivers for linux, 304xx legacy branch
local/nvidia-304xx-utils 304.108-1
    NVIDIA drivers utilities and libraries, 304xx legacy branch

Last edited by graysky (2013-10-20 09:57:18)

alphe · 2013-10-20 08:35:56

I have the same issue with the 8400GS card.

The last working configuration is
nvidia-319.32-4
linux-3.10.2-1

Following your advise I switched to nvidia-304xx for now and I will wait next version (331) to check it again.

Thanks.

wasutton3 · 2013-12-10 23:08:10

Reporting in. 331 still has the same issue with kernel version 3.12.3-1 . Has there been a bug opened on Nvidia's side? I'm going to see if 304 still works.

Arch Linux

#1 2013-10-06 08:16:28

NVRM: RmInitAdapter failed! (hardware problem?) [SOLVED]

#2 2013-10-06 15:59:21

Re: NVRM: RmInitAdapter failed! (hardware problem?) [SOLVED]

#3 2013-10-20 08:35:56

Re: NVRM: RmInitAdapter failed! (hardware problem?) [SOLVED]

#4 2013-12-10 23:08:10

Re: NVRM: RmInitAdapter failed! (hardware problem?) [SOLVED]

Board footer