Fedora 14 nVidia Drivers Install Guide (disable nouveau driver) - Comment Page: 2
Looking Fedora 22/21 nVidia Drivers Install Guide?
[inttf_post_ad1]
This is guide, howto install nVidia proprietary drivers on Fedora 14 and disable Nouveau driver. I write this guide about two weeks ago, but I Delayed the publication of this guide, because of pyxf86config bug, which cause following livna-config-display errors on boot:
Checking for module nvidia.ko: [ OK ]
Enabling the nvidia driver: Traceback (most recent call last):
File "/usr/sbin/nvidia-config-display", line 28, in
import livnaConfigDisplay.ConfigDisplay
File "/usr/lib/python2.7/site-packages/livnaConfigDisplay/ConfigDisplay.py", line 29, in
import xf86config
File...
I have an i7 2.66 with a gtx 285.
The NVIDIA driver, and the KMOD driver have been freezing and not working for 3 kernels and 3 drivers versions now. I’ve spent about 10 hours troubleshooting and trying every single trick listed on net. Its quite a joke really… Not quite sure what to think of Fedora when i can run better 3d graphics on centos.
Problem: Keep getting Black screen when it tries to load X server: (Can’t seem to get rid of Nouveau driver) Could only recover and get back into init 3 mode by adding to the command line at bootup.
Installed as per directions. Tried both kmod and akmod versions. rpm ran fine. However, when I checked /etc/X11/xorg.conf file the Section “files” was not in the file therefore I typed it in and saved file. (I left the rest of the file as is) I also checked /usr/lib64/xorg/modules/extensions folder and found the nvidia file.
The /boot/grub/grub.conf file had the proper rdblacklist=nouveau and nouveau.modeset=0 appended to the kernel line.
I am running very latest F14 Kernel and have a GForce 9800 GT.
Would like to make my video card run its 3D acceleration mode.
Any suggestions?
I’m pretty sure the problem is its just not getting rid of the nouveau driver.
Note: (I tried the sh .run nvidia installation method as well (in init 3 mode) and it told me it could not run without getting rid of nouveau driver as well. It made the same changes to the grub.conf file with no result. (I made sure I performed a reboot too).
Would appreciate any help.
Hi Paul,
Could you post output of following commands after booting:
rpmfusion nvidia driver fails to install on FX 5200 Fedora 14 with message “non-fatal POSTTRANS scriptlet failure in rpm package xorg-x11-drv-nvidia-173xx-173.14………….”
Pls help
Hi PR Karpe,
Have you tried use nVidia drivers despite of error message?
Could you post full output when installing drivers?
I tried first with easyLife, but that never works for me.
Now works fine in my Dell Vostro 1510 with a Gforce 8400M GS. The boot screen turns “a border loading”, but its just about to set up the vga=resolution in kernel line on grub.
Thanks for your help and time.
Just a quick note to thank you for putting this guide together. After many hours of mucking around, several tips in this guide nailed the issues completely. Well done!
I did as you describe in this guide but:
1)The screen started to have dots
2)The performance of rendering started to degrade
3)Finally the whole machine freezed.
My graphics card is NVIDIA 8400M GT, and my computer VGN-AR41L
Thank You.
Works great [8600 gt /lcd ]
All the best
worked perfect for my geforce 6100 integrated card on emachine with sempron 3100+..just used the first choice”kmod-nvidia” rebooted and it works awesome and love the interface as well thanks for the awesome support here…
Thank you for this guide.
However, it is not working well for me. I have a Quadro FX 570 and Fedora 14 x86_64. After all the steps I reboot and the X server gets stuck (black screen).
The kernel works fine so I can manage the machine via SSH from another computer (the keyboard doesn’t work either, so Ctrl+Alt+Fx does nothing).
The nouveau driver is not loaded (“lsmod |grep nouveau” outputs nothing) and the nvidia driver is loaded correctly (“lsmod |grep nvidia” output is ok). However, top shows that Xorg is stuck in an endless loop, running on a single core at 100% while the screen is black.
This is the Xorg.0.log file after booting with the nvidia driver enabled:
[ 34.471]
X.Org X Server 1.9.3
Release Date: 2010-12-13
[ 34.471] X Protocol Version 11, Revision 0
[ 34.471] Build Operating System: x86-02 2.6.32-71.el6.x86_64
[ 34.471] Current Operating System: Linux cygnus.pixinsight.com 2.6.35.10-74.fc14.x86_64 #1 SMP Thu Dec 23 16:04:50 UTC 2010 x86_64
[ 34.473] Kernel command line: ro root=UUID=c1b0be8c-ebcd-44cd-bb7d-0228bcb84fb1 rd_MD_UUID=2364a3e0:e8197806:d34f9bec:0fdb0c0d rd_MD_UUID=b626b78b:1af598d7:709431f3:f3e44837 rd_NO_LUKS rd_NO_LVM rd_NO_DM LANG=en_US.UTF-8 SYSFONT=latarcyrheb-sun16 KEYTABLE=es rhgb quiet nouveau.modeset=0 rdblacklist=nouveau
[ 34.473] Build Date: 20 December 2010 03:37:11PM
[ 34.473] Build ID: xorg-x11-server 1.9.3-3.fc14
[ 34.474] Current version of pixman: 0.18.4
[ 34.474] Before reporting problems, check http://wiki.x.org
to make sure that you have the latest version.
[ 34.474] Markers: (--) probed, (**) from config file, (==) default setting,
(++) from command line, (!!) notice, (II) informational,
(WW) warning, (EE) error, (NI) not implemented, (??) unknown.
[ 34.474] (==) Log file: "/var/log/Xorg.0.log", Time: Tue Jan 11 21:34:03 2011
[ 34.523] (==) Using config file: "/etc/X11/xorg.conf"
[ 34.523] (==) Using config directory: "/etc/X11/xorg.conf.d"
[ 34.523] (==) Using system config directory "/usr/share/X11/xorg.conf.d"
[ 34.591] (==) No Layout section. Using the first Screen section.
[ 34.591] (==) No screen section available. Using defaults.
[ 34.591] (**) |-->Screen "Default Screen Section" (0)
[ 34.591] (**) | |-->Monitor "
[ 34.591] (==) No device specified for screen "Default Screen Section".
Using the first device section listed.
[ 34.591] (**) | |-->Device "Videocard0"
[ 34.591] (==) No monitor specified for screen "Default Screen Section".
Using a default monitor configuration.
[ 34.591] (==) Automatically adding devices
[ 34.591] (==) Automatically enabling devices
[ 34.591] (==) FontPath set to:
catalogue:/etc/X11/fontpath.d,
built-ins
[ 34.591] (**) ModulePath set to "/usr/lib64/xorg/modules/extensions/nvidia,/usr/lib64/xorg/modules"
[ 34.591] (II) The server relies on udev to provide the list of input devices.
If no devices become available, reconfigure udev or disable AutoAddDevices.
[ 34.591] (II)
***************************************************************
***************************************************************
** Fedora switched to udev-based device detection with **
** xorg-x11-server-1.7.99.901-1. Custom HAL configuration **
** stored in /etc/hal/fdi/policy will not be seen by this **
** server version. If you have such configuration you will **
** need to update it to the new format. **
** For more information, see the Fedora wiki page **
** https://fedoraproject.org/wiki/Input_device_configuration **
***************************************************************
***************************************************************
[ 34.592] (II) Loader magic: 0x7d29c0
[ 34.592] (II) Module ABI versions:
[ 34.592] X.Org ANSI C Emulation: 0.4
[ 34.592] X.Org Video Driver: 8.0
[ 34.592] X.Org XInput driver : 11.0
[ 34.592] X.Org Server Extension : 4.0
[ 34.602] (--) PCI:*(0:10:0:0) 10de:040e:10de:0474 rev 161, Mem @ 0xfd000000/16777216, 0xd0000000/268435456, 0xfa000000/33554432, I/O @ 0x0000ec00/128, BIOS @ 0x????????/131072
[ 34.603] (II) LoadModule: "extmod"
[ 34.663] (II) Loading /usr/lib64/xorg/modules/extensions/libextmod.so
[ 34.679] (II) Module extmod: vendor="X.Org Foundation"
[ 34.679] compiled for 1.9.3, module version = 1.0.0
[ 34.679] Module class: X.Org Server Extension
[ 34.679] ABI class: X.Org Server Extension, version 4.0
[ 34.679] (II) Loading extension SELinux
[ 34.679] (II) Loading extension MIT-SCREEN-SAVER
[ 34.679] (II) Loading extension XFree86-VidModeExtension
[ 34.679] (II) Loading extension XFree86-DGA
[ 34.679] (II) Loading extension DPMS
[ 34.679] (II) Loading extension XVideo
[ 34.679] (II) Loading extension XVideo-MotionCompensation
[ 34.679] (II) Loading extension X-Resource
[ 34.679] (II) LoadModule: "dbe"
[ 34.681] (II) Loading /usr/lib64/xorg/modules/extensions/libdbe.so
[ 34.684] (II) Module dbe: vendor="X.Org Foundation"
[ 34.684] compiled for 1.9.3, module version = 1.0.0
[ 34.684] Module class: X.Org Server Extension
[ 34.684] ABI class: X.Org Server Extension, version 4.0
[ 34.684] (II) Loading extension DOUBLE-BUFFER
[ 34.684] (II) LoadModule: "glx"
[ 34.684] (II) Loading /usr/lib64/xorg/modules/extensions/nvidia/libglx.so
[ 35.016] (II) Module glx: vendor="NVIDIA Corporation"
[ 35.025] compiled for 4.0.2, module version = 1.0.0
[ 35.025] Module class: X.Org Server Extension
[ 35.026] (II) NVIDIA GLX Module 260.19.29 Wed Dec 8 12:24:30 PST 2010
[ 35.026] (II) Loading extension GLX
[ 35.026] (II) LoadModule: "record"
[ 35.027] (II) Loading /usr/lib64/xorg/modules/extensions/librecord.so
[ 35.070] (II) Module record: vendor="X.Org Foundation"
[ 35.075] compiled for 1.9.3, module version = 1.13.0
[ 35.075] Module class: X.Org Server Extension
[ 35.075] ABI class: X.Org Server Extension, version 4.0
[ 35.075] (II) Loading extension RECORD
[ 35.075] (II) LoadModule: "dri"
[ 35.076] (II) Loading /usr/lib64/xorg/modules/extensions/libdri.so
[ 35.126] (II) Module dri: vendor="X.Org Foundation"
[ 35.126] compiled for 1.9.3, module version = 1.0.0
[ 35.127] ABI class: X.Org Server Extension, version 4.0
[ 35.127] (II) Loading extension XFree86-DRI
[ 35.127] (II) LoadModule: "dri2"
[ 35.128] (II) Loading /usr/lib64/xorg/modules/extensions/libdri2.so
[ 35.129] (II) Module dri2: vendor="X.Org Foundation"
[ 35.129] compiled for 1.9.3, module version = 1.2.0
[ 35.129] ABI class: X.Org Server Extension, version 4.0
[ 35.129] (II) Loading extension DRI2
[ 35.129] (II) LoadModule: "nvidia"
[ 35.214] (II) Loading /usr/lib64/xorg/modules/drivers/nvidia_drv.so
[ 35.278] (II) Module nvidia: vendor="NVIDIA Corporation"
[ 35.282] compiled for 4.0.2, module version = 1.0.0
[ 35.282] Module class: X.Org Video Driver
[ 35.298] (II) NVIDIA dlloader X Driver 260.19.29 Wed Dec 8 12:10:14 PST 2010
[ 35.298] (II) NVIDIA Unified Driver for all Supported NVIDIA GPUs
[ 35.298] (++) using VT number 1
[ 35.305] (II) Loading sub module "fb"
[ 35.305] (II) LoadModule: "fb"
[ 35.307] (II) Loading /usr/lib64/xorg/modules/libfb.so
[ 35.320] (II) Module fb: vendor="X.Org Foundation"
[ 35.320] compiled for 1.9.3, module version = 1.0.0
[ 35.320] ABI class: X.Org ANSI C Emulation, version 0.4
[ 35.320] (II) Loading sub module "wfb"
[ 35.320] (II) LoadModule: "wfb"
[ 35.322] (II) Loading /usr/lib64/xorg/modules/libwfb.so
[ 35.324] (II) Module wfb: vendor="X.Org Foundation"
[ 35.324] compiled for 1.9.3, module version = 1.0.0
[ 35.324] ABI class: X.Org ANSI C Emulation, version 0.4
[ 35.324] (II) Loading sub module "ramdac"
[ 35.324] (II) LoadModule: "ramdac"
[ 35.324] (II) Module "ramdac" already built-in
[ 35.332] (II) NVIDIA(0): Creating default Display subsection in Screen section
"Default Screen Section" for depth/fbbpp 24/32
[ 35.332] (==) NVIDIA(0): Depth 24, (==) framebuffer bpp 32
[ 35.332] (==) NVIDIA(0): RGB weight 888
[ 35.332] (==) NVIDIA(0): Default visual is TrueColor
[ 35.332] (==) NVIDIA(0): Using gamma correction (1.0, 1.0, 1.0)
[ 35.336] (**) NVIDIA(0): Enabling RENDER acceleration
[ 35.336] (II) NVIDIA(0): Support for GLX with the Damage and Composite X extensions is
[ 35.336] (II) NVIDIA(0): enabled.
It seems X hungs after enabling GLX ?
This is my xorg.conf:
# RPM Fusion - nvidia-xorg.conf
#
Section "Device"
Identifier "Videocard0"
Driver "nvidia"
EndSection
Section "Files"
ModulePath "/usr/lib64/xorg/modules/extensions/nvidia"
ModulePath "/usr/lib64/xorg/modules"
EndSection
(BTW, I had to add the Section “Files” block manually after installing the driver).
I’d really appreciate if you could help me. Nouveau is a real nightmare. It crashes on my machine and I really need hardware OpenGL acceleration working…
Thanks in advance!
Hi Juan,
Your problem looks like very tricky, because everything seems to be ok.
You could try following:
What happens?
Another thing what you could try is add following to your /etc/X11/xorg.conf file:
And then reboot and see what happens then?
Hi JR,
Thanks for your help.
killall Xorg
I can’t kill Xorg for some reason that eludes me; the above command does nothing (?).
Anyway, I’ve changed the default run level to 3 and rebooted. Then I’ve entered the rmmod and modprobe commands you suggest, and after startx everything is just as before: black screen, X stuck and Xorg.0.log exactly as I posted before.
I’ve tried also your modifications to the xorg.conf file to no avail.
FWIW, the machine is a dual Intel Xeon E5420 and the motherboard is an ASUS DSEB-DG/SAS. The graphics card is an nVidia Quadro FX 570. This hardware worked like a charm with Fedora 11 and the proprietary nVidia driver . Then since Fedora 13 we have lost hardware acceleration and X crashes several times each day ‘thanks’ to nouveau… these are the kind of things that make Linux acceptance difficult, IMO, especially in professional environments.
Help! :(
Hi again Juan,
Sounds really strange, because you do not get any error messages. One thing what you could try is create initramfs (initial ramdisk) images for preloading modules again:
1. Double check that you have rdblacklist=nouveau and nouveau.modeset=0 options on kernel line on /boot/grub/grub.conf file.
2. Backup old initramfs image and create new initramfs for preloading modules with dracut command.
Hi,
I’d install nVidia 5200 FX graphic card driver on Fedora Core 14 completely, but when I click on desktop effects icon, it warn me : “Accelerated 3D graphics is not available”
please help me! thank you!
[[email protected] /]# grep “EE” /var/log/Xorg.0.log
(WW) warning, (EE) error, (NI) not implemented, (??) unknown.
[ 4260.626] (II) Loading extension MIT-SCREEN-SAVER
[ 4260.669] (EE) NVIDIA(0): Failed to initialize the GLX module; please check in your X
[ 4260.669] (EE) NVIDIA(0): log file that the GLX module has been loaded in your X
[ 4260.669] (EE) NVIDIA(0): server, and that the module is the NVIDIA GLX module. If
[ 4260.669] (EE) NVIDIA(0): you continue to encounter problems, Please try
[ 4260.669] (EE) NVIDIA(0): reinstalling the NVIDIA driver.
[[email protected] /]# lsmod |grep nvidia
nvidia 7081681 18
i2c_core 21328 2 nvidia,i2c_viapro
[[email protected] /]# lsmod |grep nouveau
[[email protected] /]#
Easier way is:
Yum update kernel*
Yum install kernel*-devel
Reboot
Download ur driver from nvidia
On reboot, hold shift to get to grub loader screen. Press e to edit kernel init command line. Append 3 to the command press enter and then b to boot with that command. Once in terminal, log in type su then password to assume root.
Navigate to folder where your driver file is downloaded. Type sh nvidia*****.run
Reboot and done.
THANKYOU!!!! been trying to install the Nvidia drivers for aaages and even had to reinstall F14 twice but now it works!!
Thanks for the instructions. I am about to try them. I ran into the same problem of Nouveau vs NV on my other, OpenSuSE box.
I have to say that the way Nouveau is installed/integrated/adopted by some of the distro makers is irritating. A FOSS video driver is not better by default. The Nouveau NV driver does not seem to support CUDA and I’m not clear on which driver does OpenGL better.
I always have felt FOSS was about choice… the couple of apps I’ve released under GPL, I did so people could use them easily or easily not use them.
This Nouveau driver stuff however – it used to be the case sometimes that getting a driver for a really proprietary piece of HW to work was difficult, now getting rid of an FOSS driver is difficult.
I hesitate to say this because I have a lot of gratitude for Fedora/Redhat/Suse/Slackware/Debian/etc, but the way Nouveau was included in Fedora 14 and OpenSuSE 11.3 is rather fascistic… it seems the Fedora and OpenSuSE devs want to tell me which driver to use.
It’s my box, my instance of the kernel, I can taint that kernel it if I wish. But getting rid of Nouveau really requires me to jump through a lot of hoops.
Well that’s my $0.02 FWIW.
Thank you for a very clear and concise guide, but I am still having the same problems after using your guide as all the others and am hoping you might be willing to help me figure out what is going wrong…
Usually the screen just sits there black with a cursor in the top left. Sometimes I get the white bar moving across the bottom of the screen, but once it gets all the way across it just sits there…
Based upon your requests in others above I am attaching the following information:
# yum –enablerepo=rpmfusion-nonfree-updates-testing repolist
Loaded plugins: fastestmirror, kmdl, langpacks, presto, refresh-packagekit
Adding en_US to language list
Loading mirror speeds from cached hostfile
* fedora: mirror.cc.columbia.edu
* rpmfusion-free: mirror.hiwaay.net
* rpmfusion-free-updates: mirror.hiwaay.net
* rpmfusion-nonfree: mirror.hiwaay.net
* rpmfusion-nonfree-updates: mirror.hiwaay.net
* rpmfusion-nonfree-updates-testing: mirror.hiwaay.net
* updates: mirror.cc.columbia.edu
rpmfusion-nonfree-updates-testing | 3.3 kB 00:00
rpmfusion-nonfree-updates-testing/primary_db | 11 kB 00:00
repo id repo name status
fedora Fedora 14 – x86_64 22,161
rpmfusion-free RPM Fusion for Fedora 14 – Free 411
rpmfusion-free-updates RPM Fusion for Fedora 14 – Free – Updates 309
rpmfusion-nonfree RPM Fusion for Fedora 14 – Nonfree 181
rpmfusion-nonfree-updates RPM Fusion for Fedora 14 – Nonfree – Updates 127
rpmfusion-nonfree-updates-testing RPM Fusion for Fedora 14 – Nonfree – Test Updates 10
updates Fedora 14 – x86_64 – Updates 6,202
repolist: 29,401
# ls -la /etc/yum.repos.d/
total 60
drwxr-xr-x. 2 root root 4096 Feb 14 18:33 .
drwxr-xr-x. 124 root root 12288 Feb 14 19:49 ..
-rw-r–r– 1 root root 1155 Feb 6 21:10 fedora.repo
-rw-r–r– 1 root root 1116 Feb 6 21:11 fedora-updates.repo
-rw-r–r–. 1 root root 1163 Oct 14 17:35 fedora-updates-testing.repo
-rw-r–r– 1 root root 1301 Oct 10 07:48 rpmfusion-free-rawhide.repo
-rw-r–r– 1 root root 1202 Oct 10 07:48 rpmfusion-free.repo
-rw-r–r– 1 root root 1200 Oct 10 07:48 rpmfusion-free-updates.repo
-rw-r–r– 1 root root 1260 May 17 2009 rpmfusion-free-updates-testing.repo
-rw-r–r– 1 root root 1355 Oct 10 07:48 rpmfusion-nonfree-rawhide.repo
-rw-r–r– 1 root root 1247 Oct 10 07:48 rpmfusion-nonfree.repo
-rw-r–r– 1 root root 1245 Oct 10 07:48 rpmfusion-nonfree-updates.repo
-rw-r–r– 1 root root 1305 May 17 2009 rpmfusion-nonfree-updates-testing.repo
/var/log/Xorg.0.log
[ 50.046] (II) Loading extension DRI2
[ 50.046] (II) LoadModule: “nvidia”
[ 50.053] (II) Loading /usr/lib64/xorg/modules/drivers/nvidia_drv.so
[ 50.096] (II) Module nvidia: vendor=”NVIDIA Corporation”
[ 50.097] compiled for 4.0.2, module version = 1.0.0
[ 50.097] Module class: X.Org Video Driver
[ 50.125] (EE) NVIDIA: Failed to load the NVIDIA kernel module. Please check your
[ 50.125] (EE) NVIDIA: system’s kernel log for additional error messages.
[ 50.125] (II) UnloadModule: “nvidia”
[ 50.125] (II) Unloading /usr/lib64/xorg/modules/drivers/nvidia_drv.so
[ 50.125] (EE) Failed to load module “nvidia” (module-specific error, 0)
[ 50.125] (EE) No drivers available.
[ 50.126]
Fatal server error:
[ 50.126] no screens found
My kernel line in GRUB:
kernel /vmlinuz-2.6.35.11-83.fc14.x86_64 ro root=UUID=ec1eeb0a-baaf-4ede-8d72-44cbb75e21ba rd_NO_LUKS rd_NO_LVM rd_NO_MD rd_NO_DM LANG=en_US.UTF-8 SYSFONT=latarcyrheb-sun16 KEYTABLE=us rhgb quiet nouveau.modeset=0 rdblacklist=nouveau
# lsmod |grep nvidia
# lsmod |grep nouveau
Both returned blank
Hi ViperSBT,
Try following command on command line:
What happens?
Thanks a million, your post helped me recover my system!