nvidia binary driver?

[krausest]

Did anyone manage to install the nvidia binary driver (e.g. 190.25)? If so details on how you did would be really appreciated...

[soxs060389]

Does Nvidia binary blob support kernel 2.6.31? Has anyone tried yet?
Thx for any tips and reply.

[AdamW]

Last I tried it's broken at present.

[AdamW]

It's not currently possible.

[Hlingler]

Threads merged. Other thread(s) pertaining to the same topic have been merged into this one. Please note that some posts may appear out of sequence as a result. Members are reminded to check for recent threads that address the same topic before starting a new thread.

FedoraForum.org Guidelines (http://www.fedoraforum.org/?view=guide)

V

[vallimar]

Works fine for me, don't know why you think it doesn't.

[lmcogs]

So vallimar how did you do it?

[soxs060389]

kmod, akamod, binary blob? Which one did you use to make it work?

[lmcogs]

This is what I'm getting when trying to compile either nvidia 185 or the 190 drivers. Could it be same as audio hard ware not being detected?

Kernel module compilation complete.
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.

Please see the log entries 'Kernel module load error' and 'Kernel
messages' at the end of the file '/var/log/nvidia-installer.log' for
more information.
-> Kernel module load error: insmod: error inserting './usr/src/nv/nvidia.ko':
-1 No such device
-> Kernel messages:
[<ffffffff81461622>] __qdisc_run+0x10a/0x231
RPC: Registered udp transport module.
RPC: Registered tcp transport module.
SELinux: initialized (dev rpc_pipefs, type rpc_pipefs), uses genfs_contexts
Installing knfsd (copyright (C) 1996 okir@monad.swb.de).
SELinux: initialized (dev nfsd, type nfsd), uses genfs_contexts
eth0: no IPv6 routers present
nvidia: module license 'NVIDIA' taints kernel.
Disabling lock debugging due to kernel taint
NVRM: The NVIDIA probe routine was not called for 1 device(s).
NVRM: This can occur when a driver such as rivafb, nvidiafb or
NVRM: rivatv was loaded and obtained ownership of the NVIDIA
NVRM: device(s).
NVRM: Try unloading the rivafb, nvidiafb or rivatv kernel module
NVRM: (and/or reconfigure your kernel without rivafb/nvidiafb
NVRM: support), then try loading the NVIDIA kernel module again.
NVRM: No NVIDIA graphics adapter probed!
NVRM: The NVIDIA probe routine was not called for 1 device(s).
NVRM: This can occur when a driver such as rivafb, nvidiafb or
NVRM: rivatv was loaded and obtained ownership of the NVIDIA
NVRM: device(s).
NVRM: Try unloading the rivafb, nvidiafb or rivatv kernel module
NVRM: (and/or reconfigure your kernel without rivafb/nvidiafb
NVRM: support), then try loading the NVIDIA kernel module again.
NVRM: No NVIDIA graphics adapter probed!

[vallimar]

Just do a standard manual install, I don't use the various kmod thingies.
After that, just run Xorg using the ignoreABI option. Nothing to it.

# uname -r
2.6.31-0.185.rc7.git6.xps.x86_64

# rpm -q xorg-x11-server-Xorg
xorg-x11-server-Xorg-1.6.99-44.20090901.fc12.x86_64

# grep NVIDIA dmesg
nvidia: module license 'NVIDIA' taints kernel.
NVRM: loading NVIDIA UNIX x86_64 Kernel Module 190.25 Thu Aug 20 16:36:08 PDT 2009

[vallimar]

Just did my usual custom handbuild of the latest kernel, kernel-2.6.31-0.190.rc8.fc12.src.rpm.
Built and installed fine, was able to build the 190.25 nvidia drivers against it fine, rebooted fine.
Latest Xorg, kernel and nvidia all running happily along. Now if only gnome wasn't halfways broke.

Make sure you have the correct kernel-headers and kernel-devel for the kernel you are trying
to compile the drivers for. Also, make sure that a nvidia module isn't already loaded when you
try to do your rebuild or however those auto-mods work.

I suppose it's possible your problem is caused by something in the kernel that I remove for
my custom hand-builds. That seems somewhat unlikely though, as your module is compiling,
just not loading it sounds like.

[vallimar]

Hmm, looking at that error message, it is suggesting a framebuffer driver might be loaded.
Make sure that the nouveau module is not loaded / loading. It will grab the nvidia device
and conflict with the proprietary nvidia driver. In other words, blacklist it. I think one of the
howto's tells you to do that as well. It might be loading early as part of KMS or something now.
In which case, you would have to disable rhgb altogether possibly. I disabled all that stuff
long ago so I can't really say for sure.

[lmcogs]

I wonder if final release version intends to have 3D ready in nouveau driver?

[soxs060389]

I wonder if final release version intends to have 3D ready in nouveau driver?

How do you think of that without nvidia releasing ANY data sheets at all? This won't happen within the next 5 years, reverse enginering with complex chips is somewhat time expensive. Nuveau is restriced to basic 2D and fur thee forseeable future will be.

Greets

[soxs060389]

Just did my usual custom handbuild of the latest kernel, kernel-2.6.31-0.190.rc8.fc12.src.rpm.
Built and installed fine, was able to build the 190.25 nvidia drivers against it fine, rebooted fine.
Latest Xorg, kernel and nvidia all running happily along. Now if only gnome wasn't halfways broke.

Make sure you have the correct kernel-headers and kernel-devel for the kernel you are trying
to compile the drivers for. Also, make sure that a nvidia module isn't already loaded when you
try to do your rebuild or however those auto-mods work.

I suppose it's possible your problem is caused by something in the kernel that I remove for
my custom hand-builds. That seems somewhat unlikely though, as your module is compiling,
just not loading it sounds like.

Thx thats great :-) you saved me a day of pure experimental work :-)

[phoenixpb]

i have a geforce 4 mx 400
with kernel superior at 128 monitor power off at boot
any idea ?
i use binary driver from nvidia 96.43.13

[AdamW]

soxs: you're being slightly too pessimistic. nouveau is explicitly aiming at usable 3D acceleration. Some experimental support for a few chipsets is already available in nouveau upstream, though it's disabled in the Fedora build as it's not really ready for general consumption yet. But I suspect we'll have something usable for some chips in well under five years.

[soxs060389]

soxs: you're being slightly too pessimistic. nouveau is explicitly aiming at usable 3D acceleration. Some experimental support for a few chipsets is already available in nouveau upstream, though it's disabled in the Fedora build as it's not really ready for general consumption yet. But I suspect we'll have something usable for some chips in well under five years.

Well you say yourself for some chipsets experimental support.. that's the fact of taking a hell of time to reverse engineer complex structured chips and as the manpower isn't that great, so it actually will take years to support a wide spectrum of availiable chips, as nvidia doesn't seem to even tend to release anything more than basci 2D withepapers and diffs....

Though we should quit that discussion, kind of offtopic ;-)

[krausest]

Hmm - why does my installer (190.32) fail with kernel 2.6.31-0.204.rc9.fc12.x86_64 with the following message?

FATAL: modpost: GPL-incompatible module nvidia.ko uses GPL-only symbol 'lock
dep_init_map'

How did you get around that?

[NoFate]

Hmm - why does my installer (190.32) fail with kernel 2.6.31-0.204.rc9.fc12.x86_64 with the following message?

FATAL: modpost: GPL-incompatible module nvidia.ko uses GPL-only symbol 'lock
dep_init_map'

How did you get around that?

http://forums.fedoraforum.org/showthread.php?t=229602

read this.