Tesla K40m not recognized, but driver works find for Quadro K420

Accelerated Computing CUDA CUDA Setup and Installation
1

I have a machine with a Quadro K420 and a Tesla K40m. Installation of the driver seems to be OK, I can run examples on the Quadro board. But the Tesla does not seem to be recognized by the module.

I get this from lspci -k

02:00.0 3D controller: NVIDIA Corporation GK110BGL [Tesla K40m] (rev a1)
        Subsystem: NVIDIA Corporation Device 097e
03:00.0 VGA compatible controller: NVIDIA Corporation GK107GL [Quadro K420] (rev a1)
        Subsystem: NVIDIA Corporation Device 1106
        Kernel driver in use: nvidia

And dmesg shows this

[   10.127438] NVRM: This PCI I/O region assigned to your NVIDIA device is invalid:
NVRM: BAR1 is 16384M @ 0x0 (PCI:0000:02:00.0)
[   10.127442] NVRM: The system BIOS may have misconfigured your GPU.
[   10.127447] nvidia: probe of 0000:02:00.0 failed with error -1
[   10.127499] vgaarb: device changed decodes: PCI:0000:03:00.0,olddecodes=io+mem,decodes=none:owns=none
[   10.127942] [drm] Initialized nvidia-drm 0.0.0 20150116 for 0000:03:00.0 on minor 0
[   10.128114] NVRM: The NVIDIA probe routine failed for 1 device(s).
[   10.128117] NVRM: loading NVIDIA UNIX x86_64 Kernel Module  352.39  Fri Aug 14 18:09:10 PDT 2015

So should I be battling with BIOS configurations or what? Any chance there is just some kernel parameter or /dev node creation I am missing?

2

Just to answer my own question, the problem really was that I had to set up the BIOS to allow for the large memory from the K40m.

3

and how did you do that?

4

@xor0110, Could you please share the BIOS changes that you made?