[Solved, kinda]Nvidia driver problem

[William Case]

Hi gilpel

On Thu, 2009-08-27 at 00:07 +0500, gilpel at altern.org wrote:
> Having already removed akmod-nvidia, I booted with kernel
> 2.6.29.6-217.2.3.fc11.x86_64, which was the only one still working and
> just did:
> 
> yum remove kmod-nvidia*
> 
> yum install kmod-nvidia
> 
> which took care of dependencies. So, not only was:
> 
> kmod-nvidia-2.6.29.6-217.2.3.fc11.x86_64.x86_64
>                            185.18.31-1.fc11
> 
> removed, but also:
> 
> xorg-x11-drv-nvidia.x86_64        185.18.31-1.fc11
> xorg-x11-drv-nvidia-libs.x86_64   185.18.31-1.fc11
> 
> How those packages ever got to my system, I have no idea. I never enabled
> rpmfusion-nonfree-updates-testing. (Maybe somebody at rpmfusion is keeping
> a low profile for supplying these as dependencies with akmod?)
> 
See the description below

akmod is just a script(s).  It does not supply any dependencies.

akmod-nvidia-185-18.14 supplies or directs akmod to the files that are
needed for akmod to build an rpm that can be used by kmod for various
kernels as modules for the nvidia driver.

If you have akmod and akmod-185.18.14 installed I am sure that
2.6.29.6-217.2.7 and 2.6.29.6-217.2.3 will work.  The correct nvidia
modules will be automatically built when you first boot into those
kernels.  Whatever you think you may have done with akmod previously, it
is very unlikely that your problems will re-occur.

http://rpmfusion.org/Packaging/KernelModules/Akmods

Quick Overview

RPM Fusion/Livna distributes kernel-modules as kmod packages that
contain modules precompiled for the latest kernels released by Fedora.
That works fine for most people, but it doesn't work on systems with use
different kernel -- like a self-compiled kernel, an older Fedora kernel
or the quickly changing kernels from updates-testing/rawhide. 

[This includes new stable kernels for which RPMFusion has not yet built
and distributed a kmod-nvidia-kernel_number-driver_version.]

The kmods-srpms can easily be rebuilt for those kernels using rpmbuild
with a kmod-specific parameter that defines what kernel to build the
kmod for. 

But that requires some knowledge of how to build rpms; this is what the
script akmods tries to make easier for the end user, as it does all the
steps required to build a kmod.rpm for the running kernel from a
kmod-srpm.

But the user still needs to do something manually when he needs a kmod
for a newly installed kernel. This is what the akmodsd daemon is trying
to fix: it's a script normally started from init on bootup that checks
if all kmods are present. 

If a kmod is not found then akmods tries to rebuild kmod.srpms found in
a certain place in the filesystem; if that works it will install the
rebuilt kmod into the running kernel automatically.

What is akmod-nvidia?

akmod-nvidia provides the files required by the akmod system to build a
new kmod-nvidia package when the kernel module is found to be missing.
Akmods are completely compatible with "regular" kmods, and you can mix
the two if you'd like.

> Now, kernel 2.6.29.6-217.2.8 works, kernel 2.6.29.6-217.2.7 which never
> worked and I considered destroyed by my experiments with akmod, still
> doesn't work, and kernel 2.6.29.6-217.2.3, which always worked flawlessly,
> now doesn't work. Maybe, I should have booted to kernel 2.6.29.6-217.2.8
> and worked from a terminal for the remove/install operations, but I still
> find it strange that the .3 kernel, which always worked fine with the
> 185.18.14-3.fc11 Nvidia modules, doesn't work anymore with the same
> modules.
> 

Modules for each kernel have to be built.

> Though the Nvidia drivers now seem to work very well with the .8 kernel, I
> have those error messages:
> 

Use 'locate' to ensure that all 185.18.31 files are gone.  They have all
been removed on my system, then install akmod and
akmod-nvidia-185.18.14. and reboot for each kernel.

> boot message:
> 
> Enabling the nvidia driver: /etc/rc.d/init.d/functions: line 513 :1298
> Segmentation fault "$@"
> 
> 
> dmesg | grep nvidia
> 
> nvidia: module license 'NVIDIA' taints kernel.
> nvidia 0000:01:00.0: PCI INT A -> GSI 18 (level, low) -> IRQ 18
> nvidia 0000:01:00.0: setting latency timer to 64
> nvidia-config-d[1300]: segfault at 7f5ac4000000 ip 0000003e7dc799a4 sp
> 00007ffffb6f0448 error 4 in libc-2.10.1.so[3e7dc00000+164000]
> 
> Note that I got the boot message, that otherwise only flashes by, with the
> icon at the bottom of the login screen. It would have been impossible to
> read this message if the boot process had not terminated at the login
> screen. 

> The "I" for interactive set-up has absolutely no effect and Shift
> pg-up doesn't get you to the previous boot messages screen. Those are not
> always present in dmesg or /var/log/messages and might prove helpful for
> troubleshooting.
> 

I have the same problem.  I would hope someone on the list can give us a
fix or an alternate way to step through the boot process.

-- 
Regards Bill
Fedora 11, Gnome 2.26.3
Evo.2.26.3, Emacs 23.1.1