I'm trying to build a 3 monitor setup using 2 nvidia DVI outputs and 1 intel VGA onboard output. I have this weird issue that when I enable my onboard intel vga out I can't load the nvidia x.org module anymore:

[   363.899] (WW) Warning, couldn't open module nvidia
[   363.899] (II) UnloadModule: "nvidia"
[   363.899] (II) Unloading nvidia
[   363.899] (EE) Failed to load module "nvidia" (module does not exist, 0)

When I disable the onboard intel VGA ouput my Nvidia module loads fine. I have all nvidia packages installed, and the kernel modules are correctly compiled and the nvidia kernel module is loaded.

I've tried many config settings, but I always have the same problem, I just can't use my Nvidia card somehow when this onboard intel gpu is enabled. I think this x.org layout config is very similar to what I try to accomplish: Triple-head on a Lenovo T520

Using this config (only changing the busid) the result is that I only have a working desktop on my intel output when the intel gpu is enabled, and only desktop on my nvidia screens when intel gpu is disabled.

In both cases the nvidia card is listen when lspci is used. When I use jockey-text -l while Intel GPU is enabled I get nothing, while disabled I get xorg:nvidia_current and xorg:nvidia_current_updates.

any suggestion?


Do they both show up in lspci output? Since the same config works when one of the gpus is disabled, I don't trust the error message.

You could try fixing the interrupt and memory apertures in BIOS.

  • Yes they both show up with lspci when I enable the onboard intel gpu. – Gijs Molenaar Jun 22 '12 at 9:08

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.