SlackwareThis Forum is for the discussion of Slackware Linux.
Notices
Welcome to LinuxQuestions.org, a friendly and active Linux Community.
You are currently viewing LQ as a guest. By joining our community you will have the ability to post topics, receive our newsletter, use the advanced search, subscribe to threads and access many other special features. Registration is quick, simple and absolutely free. Join our community today!
Note that registered members see fewer ads, and ContentLink is completely disabled once you log in.
If you have any problems with the registration process or your account login, please contact us. If you need to reset your password, click here.
Having a problem logging in? Please visit this page to clear all LQ-related cookies.
Get a virtual cloud desktop with the Linux distro that you want in less than five minutes with Shells! With over 10 pre-installed distros to choose from, the worry-free installation life is here! Whether you are a digital nomad or just looking for flexibility, Shells can put your Linux machine on the device that you want to use.
Exclusive for LQ members, get up to 45% off per month. Click here for more info.
I've seen sooo many posts about this in the forum and I decided to write a little how-to for everyone that has trouble installing these drivers. Personally I have a NVIDIA GeForce2 MX400 64mb card and I have never had any trouble installing the proper drivers...
As many of you have seen the install file that NVIDIA has made for you to install the drivers doesn't seem to work on the slackware-based systems, well here's a way around this.
You'll need to go to the NVIDIA site and grab the latest drivers, but here's the trick, get them from here... ftp://download.nvidia.com/XFree86/Linux-x86/ This is the driver source for the latest NVIDIA drivers.
You will need to download: NVIDIA_GLX-1.0-4349.tar.gz
and NVIDIA_kernel-1.0-4349.tar.gz
These are the current drivers at the time of this post ... keep in mind you may be looking for newer ones later on.
Now on to installing them...
You'll need to gunzip and untar them to a directory of your choice and then su to root. Once you are root, cd to the NVIDIA_kernel-1.0-4349 directory and type make then make install
Once this is done do a cd ../NVIDIA_GLX-1.0-4349 and run make and make install
Once this is completed the drivers should be in place though you still need to change a few things. After this is completed run xf86config and select the default nvidia driver that XF86 gives you.
Now you'll have to locate your XF86Config file which is usually located in /etc/X11/XF86Config
Open this file with your favorite editor and search for the text that looks something like Section "Device" or something that says Driver "nv" this can be acomplished with the ctrl+w search feature of pico, possibly close to line 336... Once you find this you'll need to change the Driver "nv" like to resemble Driver "nvidia"
Once you do that, save the file and give a shot at starting X.... that should be all that you need to do other than changing screen resolutions if you havn't done so already. Any questions or concerns about this.... or problems with the method I described let me know please and I'll fix it immediatly.
I hope this helps some of you 's out there in installing these drivers, and anyone else that has some problems...
I compiled everything just fine, but when I change "nv" to "nvidia" in the XF86Config file, then X simply hangs at a black screen when I try to start it. If I change "nvidia" back to the "nv" it originally was, then everything's great. Maybe I missed something?
Try adding another user to the system, then logging in as that user and starting X... I've had this problem before wityh my computer but I'm not sure how I fixed it other than removing my username and home dir and readding it.
Yeah, I just used the regular nvidia driver .run pack, built the kernel module myself (part of the install), and then followed the Nvidia readme, and it worked great. No problems.
This may help people:
If you are having trouble installing the drivers, and it complains about kernel headers. Make sure you install the kernel-headers package from the slackware CD, and the slackware source from the /extra directory on the Slackware FTP.
I used the executable that NVIDIA are now providing on their drivers page. It came up with the common 'couldn't find kernel headers' error the first time I ran it, but there's a command-line switch you can add to point the setup program at the kernel headers (to get the advanced options when running the NVIDIA installer, type 'sh NVIDIA-Linux-x86-1.0-4363.run -A)
I believe the exact command line I used to install the kernel headers was 'sh NVIDIA-Linux-x86-1.0-4363.run --kernel-include-path=/usr/src/linux-2.4.20'
The above worked for me, I said no to trying to find kernel headers on the nvidia website, and it then went ahead and built what it needed from my kernel headers installed locally.
Please bear in mind with the nvidia driver that if you recompile your kernel for any reason, you will subsequently have to re-run the nvidia setup program so that the correct modules are re-inserted into your new kernel.
Hope this is of some help,
Mark
P.S. cyoul8r, if you're still having problems, try looking into /var/log/XFree86.0.log and posting up any lines that begin with (EE). The (EE) lines are errors being reported by XFree86.
Astro,
I realized that about your thread. And it's a good thing. For my part, I was just putting in a little counter to the part that said:
Quote:
...the install file that NVIDIA has made for you to install the drivers doesn't seem to work on the slackware-based systems
And, it's mostly just for the benefit of those searching on the subject later.
I guess, and this is from a new user of slack, that your 'howto' is probably a better thing to know than the "stupid .RUN file" thing. And, I'll probably go with that on my next slack install. Which won't be long, given my maniacal penchant for reformatting and reinstalling.
After you compile the kernel and everything, where does it stick it? Shouldn't I have another entry in Lilo or something?
I also tried to grab the header files and use the .run, but I couldn't figure out which stuff I actually needed to download. I don't have kernel source in /usr/source. Just a /rpm directory.
Oh, and I tried to look through the log. No lines with EE. It's completely hanging, and I have to hard reboot the system once it hangs. It also seems to be writing over the XF86Config file with html garbage when I have to hard restart as well.
Dunno if this has actually got anything to do with this, but i'm starting to run out of ideas.
Got an AMD AthlonXP 2400+, 512DDR (dual channel configured), on a Soltek 74FRN2-L mobo (nvidia2 chipset), and a Abit Siluro GeForce 4 Ti4200 OTES
Everything runs ok, X starts, UT2K3 runs, BUT the framerate is way down, going as low as 12 FPM. This is a new problem, SuSE didn't have the issue, despite using the same kernel package (in other words the downloaded self compiled one, rather than the distro supplied tarball).
Whilst i acknowledge, life could be worse, with X not running at all, i know that this should be working.
I've tracked the primary effect to be:
bash-2.05b# cat /proc/driver/nvidia/agp/status
Status: Disabled
as a result i have searched to find something to change said status to Enabled, with all the associated bells and whistles (read better performance). Instead I found that agpgart is part of the problem. Recompiling the kernel, making agpgart a module, or not yields nothing. Additionally i've found that dmesg's out put ends with:
Linux agpgart interface v0.99 (c) Jeff Hartmann
agpgart: Maximum main memory to use for agp memory: 439M
agpgart: unsupported bridge
agpgart: no supported devices found.
0: nvidia: loading NVIDIA Linux x86 nvidia.o Kernel Module 1.0-4363 Sat Apr 19 17:46:46 PDT 2003
I am going to do a source code install of the NVidia drivers, as described here, since this has previously worked cure some SuSE issues (read failure of X to start). But frankly it seems that there is something else amiss.
Hi Chakkerz.
One thing you have to do is using Nvidia AGP driver (it is already part of the driver, just must be activated), as kernel's AGP driver doesn't support nForce2 chipset. To activate it you have to insert "option "nvAGP" "1""(or something similar, check in readme)in XF86Config, it forces X to use Nvidia AGP driver. I have near the same computer configuration and have about 3000 fps with Nvidia driver (500 fps full screen 1024x768).
Yeah i tried that ("NvAGP" "1" which forces the use of NVidia's AGP driver ... or rather "tries" to use NVidia's) but the XFree86 log still states:
(WW) NVIDIA(0): Failed to verify AGP usage
and agp ain't enabled according to /proc/... either (framerates in UT2K3 support the off notion)
Here is something else i just noticed that's odd:
(WW) NVIDIA(0): OpenGL is only supported in depths 16 and 24
This is odd, because i'm running at 24bit, according to /etc/X11/XF86Config
Am i missing something about Slackware?? in SuSE and debian the XF86Config file had a -4 postfix, in Slackware it doesn't. And in either of those distro's a simple install yielded working results, in the GFX department ... ok debian was a struggle, but no problem like this. SuSE first up though.
LinuxQuestions.org is looking for people interested in writing
Editorials, Articles, Reviews, and more. If you'd like to contribute
content, let us know.