C1060 undetected NVIDIA driver cannot see C1060

Accelerated Computing CUDA CUDA Programming and Performance
1

I just put a C1060 on an MSI K9A2 Platinum MB, in PCIE slot 4. I have a Quadro FX 1500 in PCIE slot 1. lspci shows:
.
.
.
01:00.0 VGA compatible controller: nVidia Corporation G71 [Quadro FX 1500] (rev a1)
…
03:00.0 3D controller: nVidia Corporation Device 05e7 (rev a1)

But running nvidia-xconfig --enable-all-gpus --multigpu=On --probe-all-gpus --query-gpu-info --sli=Off I get:
Number of GPUs: 1

GPU #0:
Name : Quadro FX 1500
PCI BusID : PCI:1:0:0

…

I am running openSuSE 11.1. I installed driver from rpms:
x11-video-nvidiaG02-180.22-4.1
nvidia-gfxG02-kmp-default-180.22_2.6.27.7_9.1-4.1

I had to create device nvidia1 by hand (c 195 1) as it was not created by OS. Even then, trying to run deviceQuery, I get:
CUDA device count = 1
cudaSafeCall() Runtime API error in file <deviceQuery.cu>, line 60 : initialization error.

which I can tell is attempting to run on the Quadro, as the screen blanks.

How do I get the nvidia driver to see the C1060?

TIA.

D-man

2

Solved.

Removed openSUSE nvidia rpms and ran the nvidia supplied scripts. I guess Novell messed up something.

I had been seeing dmesg messages about nvidiafb or rivafb interfering with the nvidia driver. So I removed those modules. But

/sbin/lspci -d “10de:*” -v -xxx

showed

01:00.0 VGA compatible controller: nVidia Corporation G71 [Quadro FX 1500] (rev a1) (prog-if 00 [V

GA controller])

    Subsystem: nVidia Corporation Device 032c

    Flags: bus master, fast devsel, latency 0, IRQ 18

    Memory at f8000000 (32-bit, non-prefetchable) 

    Memory at c0000000 (64-bit, prefetchable) 

    Memory at f7000000 (64-bit, non-prefetchable) 

    I/O ports at b800 

    [virtual] Expansion ROM at f9ee0000 [disabled] 

    Capabilities: [60] Power Management version 2

    Capabilities: [68] Message Signalled Interrupts: Mask- 64bit+ Count=1/1 Enable-

    Capabilities: [78] Express Endpoint, MSI 00

    Capabilities: [100] Virtual Channel <?>

    Capabilities: [128] Power Budgeting <?>

    Kernel driver in use: nvidia

    Kernel modules: nvidia, nvidiafb

…

03:00.0 3D controller: nVidia Corporation Device 05e7 (rev a1)

    Subsystem: nVidia Corporation Device 066a

    Flags: bus master, fast devsel, latency 0, IRQ 10

    Memory at fd000000 (32-bit, non-prefetchable) 

    Memory at dc000000 (64-bit, prefetchable) 

    Memory at fa000000 (64-bit, non-prefetchable) 

    I/O ports at d800 

    Expansion ROM at fea80000 [disabled] 

    Capabilities: [60] Power Management version 3

    Capabilities: [68] Message Signalled Interrupts: Mask- 64bit+ Count=1/1 Enable-

    Capabilities: [78] Express Endpoint, MSI 00

    Capabilities: [100] Virtual Channel <?>

    Capabilities: [128] Power Budgeting <?>

    Capabilities: [600] Vendor Specific Information <?>

    Kernel modules: nvidiafb
3

Hey!!

I have the same issue!! my tesla C1060 is not showing up neither on nvidia-settings nor on sax2.

Please let me know how did you remove the RPMs, is it via the “software management”? did you also remove their kernel-source?

Thanks!!!

4

I don’t have RPMs and mine still aren’t showing up. So don’t get your hopes up.

5

Finally solved!!

When I chose to remove the old Nvidia driver via “software management”, it automatically forced me to remove its kernels…

Then, I installed the new 185.18.14 driver manually through the command line console using: sh NVIDIA…

Thanks again for the suggestion!!