You are not logged in.

#1 2014-04-03 11:57:31

zer0def
Member
Registered: 2014-04-03
Posts: 1

[SELF-SOLVED] Issues with initializing second graphics card

Running a rig with a GTX750 and GT640 rev2 on nvidia 334.21.

Attempts at running deviceQuery sample end up in:

[zer0def@arch-dev deviceQuery]$ ./deviceQuery
./deviceQuery Starting...

 CUDA Device Query (Runtime API) version (CUDART static linking)

cudaGetDeviceCount returned 10
-> invalid device ordinal
Result = FAIL

Utilities return:

[zer0def@arch-dev ~]$ nvidia-smi
Unable to determine the device handle for GPU 0000:02:00.0: Unknown Error
[zer0def@arch-dev ~]$ nvidia-debugdump -l
Found 2 NVIDIA devices
        Device ID:              0
        Device name:            GeForce GTX 750   (*PrimaryCard)
        GPU internal ID:        GPU-df106a72-c61f-ea5d-fe8e-2dacf0b48f4c

Error: nvmlDeviceGetHandleByIndex(): Unknown Error
FAILED to get details on GPU (0x1): Unknown Error

Funnily enough, in the nvidia-debugdump listing example, only the device in the first PCIe slot is detected (already tried swapping cards between slots).

[zer0def@arch-dev ~]$ sudo lshw -c display
  *-display
       description: VGA compatible controller
       product: NVIDIA Corporation
       vendor: NVIDIA Corporation
       physical id: 0
       bus info: pci@0000:01:00.0
       version: a2
       width: 64 bits
       clock: 33MHz
       capabilities: pm msi pciexpress vga_controller bus_master cap_list rom
       configuration: driver=nvidia latency=0
       resources: irq:18 memory:fc000000-fcffffff memory:c0000000-cfffffff memory:d6000000-d7ffffff ioport:b800(size=128) memory:fbf80000-fbffffff
  *-display
       description: VGA compatible controller
       product: GK208 [GeForce GT 640 Rev. 2]
       vendor: NVIDIA Corporation
       physical id: 0
       bus info: pci@0000:02:00.0
       version: a1
       width: 64 bits
       clock: 33MHz
       capabilities: pm msi pciexpress vga_controller bus_master cap_list rom
       configuration: driver=nvidia latency=0
       resources: irq:19 memory:fd000000-fdffffff memory:d8000000-dfffffff ioport:c800(size=128) memory:fe980000-fe9fffff

On every attempt to run CUDA device query, dmesg returns:

[  483.509263] nvidia 0000:01:00.0: irq 44 for MSI/MSI-X
[  484.279823] NVRM: Your system is not currently configured to drive a VGA cons
ole
[  484.279830] NVRM: on the primary VGA device. The NVIDIA Linux graphics driver
[  484.279832] NVRM: requires the use of a text-mode VGA console. Use of other c
onsole
[  484.279834] NVRM: drivers including, but not limited to, vesafb, may result i
n
[  484.279835] NVRM: corruption and stability problems, and is not supported.
[  484.282174] nvidia 0000:02:00.0: irq 45 for MSI/MSI-X
[  485.059831] NVRM: RmInitAdapter failed! (0x26:0xffffffff:1177)
[  485.059937] NVRM: rm_init_adapter failed for device bearing minor number 1
[  485.059993] NVRM: nvidia_frontend_open: minor 1, module->open() failed, error -5

Seperately, these cards run just fine. Now I'm wondering, whether this could be a hardware, driver or my brain's issue. I'd appreciate any help with this one.

Apparently, in my particular instance it most likely was an old PCIe bus (whether it was incompatible or simply unable to deliver enough power, is unknown at this point).

Last edited by zer0def (2014-06-03 22:53:00)

Offline

#2 2014-06-03 22:41:34

jkd
Member
Registered: 2014-06-03
Posts: 1

Re: [SELF-SOLVED] Issues with initializing second graphics card

Hello there,
I am also getting exactly same error. Please help ... !!
In my system I have 4 GPUs sometimes it shows address of 1 and other times it shows addresses of 2 GPUs and others simply says unable to find the handle.

My gpus are Geforce GTX 590
Ubuntu 14.04 LTs

CUDA 6
device driver NVIDIA 331.79

Thanks in advance.
JKD

Offline

Board footer

Powered by FluxBB