You are not logged in.
After an upgrade, a downgrade, and then another upgrade, I've run into some issues with my nvidia drivers.
I upgraded to 270, didn't work, downgraded to 260 (and Xorg and all the rest, too, with it), and then I wanted to get 270.x working, so I want to try again. Anyway, I figured out most of my issues, but somewhere on the down/upgrading processes, I must have somehow missed some traces of the old driver (260.x), and now I'm getting an API Mismatch. I've tried manually unloading the module, with modprobe -r nvidia, then reinstalling the driver, rebooting, to no avail. I have either missed a trace of it, or have done something in the wrong order, perhaps.
A few things: pacman -Q | grep nvidia
lib32-nvidia-cg-toolkit 3.0-1
lib32-nvidia-utils 270.41.03-1
nvidia 270.41.03-1
nvidia-cg-toolkit 3.0-3
nvidia-utils 270.41.03-1
Additionally, I added nvidia to MODULES in mkinitcpio. I'm not sure if that's needed or not, or if I may have left something else in or out.
Any other details needed?
Thanks!
Last edited by Neikron (2011-04-20 05:48:03)
Offline
If you don't have any packages in IgnorePkg than a 'pacman -Syu' update should work.
Offline
If you don't have any packages in IgnorePkg than a 'pacman -Syu' update should work.
I don't believe I've ignored any packages. There were even some minor updates to nvidia and xorg-server since then, and it's still the same error. The old package is still lying around somewhere. Also, when I try to update anyway, it says there's nothing to update.
Offline
i suggest to remove the module from mkinitcpio.conf and regenerate the initrd with mkinitcpio -p kernel26 and reboot
Give what you have. To someone, it may be better than you dare to think.
Offline
i suggest to remove the module from mkinitcpio.conf and regenerate the initrd with mkinitcpio -p kernel26 and reboot
I removed the entry from mkinitcpio then regenerated initrd with the following command followed by a reboot. It's displaying the same error as before.
Offline
I was just thinking that it could have something to do with me installing the nvidia driver from the official nvidia site, and it mentioned some kernel issue back then, and I didn't think about it until now. There may still be some traces of the driver on the system, and that's what's happening with the API error.
Edit: Additionally, after downgrading to the versions that work, I can only run it without restarting. So, for example, I downgrade kernel, xorg, xf86*, etc, and startx to boot fine. But, I can't mount anything without rebooting, and then the initial error comes up when I reboot. Some, damned if I do, damned if I don't.
Mounting example: mount /dev/sda1 /mnt
FATAL: Could not load /lib/modules/2.6.38-ARCH/modules.dep: No such file or directory
fuse: device not found, try modprobe fuse (which gives the same, Could not load thing...)
Anyway, I am really having lots of issues with this. Not sure if I should back everything up, or if there's a fix.
Last edited by Neikron (2011-04-20 03:23:00)
Offline
If you installed the nvidia driver using their installer you're basically on your own, there's a reason you're not supposed to do that. Fortunately, I believe their installer has an uninstall capability you'd want to look at. You'd then need to reinstall the packages whose files are touched by the nvidia installer (not exactly sure which, kernel26 and xorg-related I believe).
Allan-Volunteer on the (topic being discussed) mailn lists. You never get the people who matters attention on the forums.
jasonwryan-Installing Arch is a measure of your literacy. Maintaining Arch is a measure of your diligence. Contributing to Arch is a measure of your competence.
Griemak-Bleeding edge, not bleeding flat. Edge denotes falls will occur from time to time. Bring your own parachute.
Offline
Thanks a lot for the reply. Turns out all I had to do was
nvidia-*.run --uninstall
Update to nvidia 270 and newest xorg-server, etc, and now it's working flawlessly again.
Note to self: Always run native Arch packages!
Thanks to everyone else who helped.
Offline