Error inserting nvidia nvidia module can't find my Tesla S1070

Accelerated Computing CUDA CUDA Programming and Performance
1

I’m trying to install my Tesla S1070 on my new system and I’m having a problem with the NVIDIA driver. After installing using -k to specify the kernel version I want because using the default option doesn’t install it just dies when attempting the insmod with no such device error, I do a modprobe nvidia and I get the same error message as in the driver install script which is expected. This is the error:

ig ~ # modprobe nvidia
FATAL: Error inserting nvidia (/lib/modules/2.6.32.3-vanilla/kernel/drivers/video/nvidia.ko): No such device

I’ve also tried with the stock CentOS 5.4 kernel 2.6.18-164.10.1.el5 and it yields the same results.

If I do an lspci it shows the device being there:
0b:00.0 3D controller: nVidia Corporation GT200 [Tesla C1060] (rev a1)
0d:00.0 3D controller: nVidia Corporation GT200 [Tesla C1060] (rev a1)
8b:00.0 3D controller: nVidia Corporation GT200 [Tesla C1060] (rev a1)
8d:00.0 3D controller: nVidia Corporation GT200 [Tesla C1060] (rev a1)

This hardware also worked in another system before I transferred it over to the new hardware so I know it is functioning correctly as well. I’ve also tried disabling the nvidia/riva fb module in the kernel 2.6.32.3 as well.

The system also has an nvidia integrated video card that I’ve also disabled. I added an ATI card in an extra pci slot and set the add on card to be the primary card in the BIOS. I thought maybe the module was trying to attach itself to the built in nvidia video card, but it still gives the same results of no such device.

Any help would be greatly appreciated.
nvidia_bug_report.log.gz (47.4 KB)

2

Update to this post. After some more digging around I found this in /var/log/messages

[codebox]Jan 22 15:29:19 ig kernel: nvidia: probe of 0000:0d:00.0 failed with error -1

Jan 22 15:29:20 ig kernel: NVRM: This PCI I/O region assigned to your NVIDIA device is invalid:

Jan 22 15:29:20 ig kernel: NVRM: BAR0 is 0M @ 0x0 (PCI:008b:00.0)

Jan 22 15:29:20 ig kernel: NVRM: The system BIOS may have misconfigured your GPU.

Jan 22 15:29:21 ig kernel: nvidia: probe of 0000:8b:00.0 failed with error -1

Jan 22 15:29:21 ig kernel: NVRM: This PCI I/O region assigned to your NVIDIA device is invalid:

Jan 22 15:29:21 ig kernel: NVRM: BAR0 is 0M @ 0x0 (PCI:008d:00.0)

Jan 22 15:29:21 ig kernel: NVRM: The system BIOS may have misconfigured your GPU.

Jan 22 15:29:22 ig kernel: nvidia: probe of 0000:8d:00.0 failed with error -1

Jan 22 15:29:22 ig kernel: NVRM: The NVIDIA probe routine failed for 4 device(s).

Jan 22 15:29:22 ig kernel: NVRM: None of the NVIDIA graphics adapters were initialized![/codebox]

I’m installing this on a fairly cutting edge system so maybe there is a problem with the Tyan motherboard I’m using? I’m using a Tyan S4985G3NR-E motherboard.