No X with nvidia-current 270.41.06-0ubuntu1.1
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
nvidia-graphics-drivers (Ubuntu) |
Invalid
|
Undecided
|
Unassigned |
Bug Description
X refuses to start on 32 bit PAE Ubuntu after todays kernel and nvidia updates.
When starting the PAE kernel, I get the following message in Xorg.0.log:
[ 42.536] (II) LoadModule: "nvidia"
[ 42.536] (II) Loading /usr/lib/
[ 42.536] (II) Module nvidia: vendor="NVIDIA Corporation"
[ 42.536] compiled for 4.0.2, module version = 1.0.0
[ 42.536] Module class: X.Org Video Driver
[ 42.539] (EE) NVIDIA: Failed to load the NVIDIA kernel module. Please check your
[ 42.539] (EE) NVIDIA: system's kernel log for additional error messages.
[ 42.540] (II) UnloadModule: "nvidia"
[ 42.540] (II) Unloading nvidia
[ 42.540] (EE) Failed to load module "nvidia" (module-specific error, 0)
[ 42.540] (EE) No drivers available.
[ 42.540]
Fatal server error:
[ 42.540] no screens found
I had the same symptoms when I started using the PAE kernel with nvidia drivers installed using the non-PAE kernel (long time ago). Starting the non-PAE kernel gives me a working X, but unused nvidia driver since the driver was installed while running the PAE kernel.
lsb_release -rd:
Description: Ubuntu 11.04
Release: 11.04
apt-cache policy nvidia-current:
270.41.
What I expected to happen:
X and gdm starting up and allowing me to graphically log in as usual.
What happened instead:
No GUI since X refused to start.
What I have not tried so far:
Reinstall nvidia-current while running the non-PAE kernel.
Erm... fixed by removing nvidia-current and reinstalling while running PAE kernel.
What I did in more detail:
Remove nvidia-current while running non-PAE kernel.
Restart with PAE kernel.
Log in without X.
Delete xorg.conf.
Start gdm and login.
Reinstall nvidia-current.
Restart with PAE kernel.
Works.
I have no idea why it did not work from the beginning like that.