VGA Console Warning

Hi, I’m setting up CUDA 5.0 on an Ubuntu 12.04 headless server… Initially, GPU 0 was crashing almost instantly, which was probably caused by conflicts with the EFI frame buffer device… Adding:

video=vesa:off video=efifb:off vga=normal

to the linux commandline seems to solve the problem. However, I still see these message in dmesg:

[   12.230635] nvidia: module license 'NVIDIA' taints kernel.
[   12.230638] Disabling lock debugging due to kernel taint
[   12.236784] vgaarb: device changed decodes: PCI:0000:01:00.0,olddecodes=io+mem,decodes=none:owns=io+mem
[   12.236786] vgaarb: transferring owner from PCI:0000:01:00.0 to PCI:0000:02:00.0
[   12.236843] nvidia 0000:02:00.0: enabling device (0000 -> 0003)
[   12.236848] vgaarb: device changed decodes: PCI:0000:02:00.0,olddecodes=io+mem,decodes=none:owns=none
[   12.236893] NVRM: loading NVIDIA UNIX x86_64 Kernel Module  310.14  Tue Oct  9 11:52:41 PDT 2012
[   14.701738] NVRM: GPU at 0000:01:00: GPU-bfccfb8f-e0c9-c73b-664a-50c5a3dde5f4
[   14.701775] NVRM: Your system is not currently configured to drive a VGA console
[   14.701776] NVRM: on the primary VGA device. The NVIDIA Linux graphics driver
[   14.701777] NVRM: requires the use of a text-mode VGA console. Use of other console
[   14.701778] NVRM: drivers including, but not limited to, vesafb, may result in
[   14.701787] NVRM: corruption and stability problems, and is not supported.

Is this warning about the system not being configured to drive a VGA console something I should worry about?

I don’t see it appearing on our older servers running the same configuration.

Looking at just the log, you’re probably in the clear. Try running nvidia-smi, and building and running the deviceQuery sample.