Problems with nvidia driver on Fedora 14

[Florist]

Hello!
I bought new Lenovo Z560 a few month ago. I installed Fedora 14 on my laptop and was trying to setup nvidia drivers. But probles is that Lenovo Z560 has Nvidia Optimus technology. In other words my laptop has two video cards that worked in turn.Because of that I can't install drivers from repository.
There someone said that situation can be fixed by compiling binary drivers from official site. I tried to do that, but I don't know how to
1) stop X server in Fedora 14
2) kill / disable nouveau in Fedora 14.

Do anybody know how to do it?

[SiliconSlick]

It isn't clear to me why you think the nvidia drivers from rpmfusion won't work while the nvidia drivers from nvidia will.

Did you even try the rpmfusion ones? If not, you should. See the yellow link at the top of the page for how to set up nvidia drivers. It is currently version 260.19.36. Checking nvidia's site says the current 310M drivers is 260.19.44 (released yesterday). Nothing in its release notes indicate anything about Optimus. The rpmfusion ones are directly packaged from nvidia's so I don't know why you think nvidia's would be better unless you think Optimus support was added somewhere between .36 and .44.

If you use the akmod-nvidia package

a) Nvidia's horrible installer won't destroy your mesa-libGL libraries

b) your module will be automatically built each time your kernel is rebuilt.

If you choose the path you are going you will have to repeat the process you are about to do everytime a kernel is updated... and you'll have to do it from the command line (no GUI).

So unless you have a definite need to use nvidia's (I saw nothing in the thread you linked that said you did), I would try the preferred way of using rpmfusion's solution.

SS

[leigh123linux]

Originally Posted by SiliconSlick

It isn't clear to me why you think the nvidia drivers from rpmfusion won't work while the nvidia drivers from nvidia will.

Did you even try the rpmfusion ones? If not, you should. See the yellow link at the top of the page for how to set up nvidia drivers. It is currently version 260.19.36. Checking nvidia's site says the current 310M drivers is 260.19.44 (released yesterday). Nothing in its release notes indicate anything about Optimus. The rpmfusion ones are directly packaged from nvidia's so I don't know why you think nvidia's would be better unless you think Optimus support was added somewhere between .36 and .44.

If you use the akmod-nvidia package

a) Nvidia's horrible installer won't destroy your mesa-libGL libraries

b) your module will be automatically built each time your kernel is rebuilt.

If you choose the path you are going you will have to repeat the process you are about to do everytime a kernel is updated... and you'll have to do it from the command line (no GUI).

So unless you have a definite need to use nvidia's (I saw nothing in the thread you linked that said you did), I would try the preferred way of using rpmfusion's solution.

SS

There is

Linux x64 (AMD64/EM64T) Display Driver





Version:

260.19.44 Certified

Release Date:

2011.03.07

Operating System:

Linux 64-bit

Language:

English (U.S.)

File Size:

45.8 MB





Release Highlights

• Updated the NVIDIA X driver to not update mode timings for G-Sync compatibility when NVIDIA 3D Vision or NVIDIA 3D VisionPro is enabled along with a G-Sync device.

Supported products


Additional information

Note that many Linux distributions provide their own packages of the NVIDIA Linux Graphics Driver in the distribution's native package management format. This may interact better with the rest of your distribution's framework, and you may want to use this rather than NVIDIA's official package.

Also note that SuSE users should read the SuSE NVIDIA Installer HOWTO before downloading the driver.

Installation instructions: Once you have downloaded the driver, change to the directory containing the driver package and install the driver by running, as root, sh ./NVIDIA-Linux-x86_64-260.19.44.run

One of the last installation steps will offer to update your X configuration file. Either accept that offer, edit your X configuration file manually so that the NVIDIA X driver will be used, or run nvidia-xconfig

Note that the list of supported GPU products is provided to indicate which GPUs are supported by a particular driver version. Some designs incorporating supported GPUs may not be compatible with the NVIDIA Linux driver: in particular, notebook and all-in-one desktop designs with switchable (hybrid) or Optimus graphics will not work if means to disable the integrated graphics in hardware are not available. Hardware designs will vary from manufacturer to manufacturer, so please consult with a system's manufacturer to determine whether that particular system is compatible.
See the README for more detailed instructions.

[Florist]

SiliconSlick - I think leigh123linux answered you question. My questions are still actual, because I want try to compil binary drivers.

[SiliconSlick]

So he needs .44 (which isn't in rpmfusion yet) but it may not work (due to manufacturer wiring)??? (I didn't connect 3D Gsync issues with Optimus)

OK...to answer the OP

1) either:
a) pull up a virtual terminal with <Ctrl><Alt><F2> and run (as root) 'telinit 3'... then do a 'ps axfu | grep gdm' and kill all the tasks with bugs and haven't quit, or,
b) (easier/cleaner) when you boot, pause grub and edit the kernel boot line to append a '3' to the end, then boot using that line

2) Do what rpmfusion does and edit/create the file /etc/modprobe.d/blacklist-nouveau.conf with the line 'blacklist nouveau'. You will also need to edit /boot/grub/grub.conf and add the bits 'nouveau.modeset=0 rdblacklist=nouveau nomodeset' to your kernel line (though the last nomodeset might not be necessary). You will also need to edit your /etc/X11/xorg.conf and change nouveau to nvidia... assuming you have such a file. If not...

Also make sure you have your kernel-devel packages installed. Then run the nvidia installer (and have it trash Mesa-libGL and need to be rebuilt on each kernel update).

SS

[DBelton]

only time I had to compile the nvidia drivers myself was awhile back when I had a laptop with a geforce4 440 go adapter in it, but Dell in their infinite wisdom changed the device ID string where the nvidia drivers wouldn't detect it. (not in windows or in linux) I had to download the drivers, add the device string my adapter was returning, then compile. (had to mod the ini file in windows as well to get it to work on there)

I haven't run across anymore that were that bad since, though. Most will at least be recognized by the nvidia drivers now.

[Florist]

SiliconSlick, thx you very and very much.

i think thread is over.