maxkukartsev

[Log in to get rid of this advertisement]

Hi everyone,

I am using Debian Lenny (2.6.26-19).

I compiled a new kernel with nVidia driver support using the downloaded script NVIDIA-Linux-x86-185.18.36-pkg1.run. Booting up that kernel launched the nVidia driver successfully.

Then I installed the Intel wireless driver by following the instructions in http://wiki.debian.org/iwlagn, which compiled a new kernel for me.

Now I have 2 entries in my GRUB boot menu,
1) Debian GNU/Linux, kernel 2.6.30-bpo.1-686 ## after wireless driver
2) Debian GNU/Linux, kernel 2.6.26-2-686 ## after nVidia driver

If I boot (1), I get wireless support but the nVidia driver does not get loaded. So I need to have a generic xorg.conf (without nVidia module) for this to work.

If I boot (2), I get nVidia, but no wireless.

How can I have both modules loaded?

Thanks,
Max Kukartsev

AwesomeMachine

Install the Intel wireless module first. Then, into the same kernel, install the Nvidia module. I can't remember exactly where it is, but there is a fairly recent guide to installing Nvidia modules in Debian, on the Debian site: debian.org

maxkukartsev

I booted up into my wireless-enabled kernel, and from there ran the NVIDIA-Linux-x86-185.18.36-pkg1.run script to add the nVidia module to the kernel as well. Now if I boot up into my (1) Debian GNU/Linux, kernel 2.6.30-bpo.1-686, I get both the nVidia and wireless modules.

HOWEVER:
When the config script for the iwlagn package rebuilt the kernel, it appeared to IGNORE the nVidia module which was loaded into the kernel at the time. I did not reboot after installing the nVidia driver (just started the GDM init script), so perhaps this is why the iwlagn script did not include the nVidia driver in its kernel recompilation?

Thanks,
Max Kukartsev

freet15

Max, I have the same problem as you got. but I ran the NVIDIA-Linux-x86-185.18.36-pkg1.run in my kernel environment(2.6.30-bpo.1-686) still failed.

the message show as follow:

ERROR: Unable to load the kernel module 'nvidia.ko'. This happens most
frequently when this kernel module was built against the wrong or
improperly configured kernel sources, with a version of gcc that differs
from the one used to build the target kernel, or if a driver such as
rivafb/nvidiafb is present and prevents the NVIDIA kernel module from
obtaining ownership of the NVIDIA graphics device(s), or NVIDIA GPU
installed in this system is not supported by this NVIDIA Linux graphics
driver release.

maxkukartsev

Hi freet15,

I would assume the message is correct, and there is a mismatch in compiler version used to build the current kernel and the nVidia module.

Here's what I would do:

1. Check the compiler used to build the current kernel:
   Code:

   cat /proc/version

   Look for gcc version x.y.z and compare that with the version of gcc in your PATH:
   Code:

   gcc -v

2. Download that version of the compiler. Older versions of gcc appear as gcc-x.y, so to install:
   Code:

   sudo apt-get install gcc-x.y

   Binaries from such packages will also be named gcc-x.y
3. Before running NVIDIA-Linux-x86-185.18.36-pkg1.run, set the $CC environment variable so that it'll use that particular compiler:
   Code:

   export CC=gcc-x.y