Chris Kloiber wrote:
> Is that using rpmfusion packages, or nVidia's blob installer? The
> company I work for offers CUDA servers now, and they seem to be running
> the same nvidia driver version I use on my desktop (without CUDA,
> obviously). It seems the CUDA libraries aren't packaged by rpmfusion,
> but may get installed by the blob installer. I would try the nVidia
> installer from their website, and if that works compare the libraries
> installed to the rpmfusion packages. Find the differences and bug
> rpmfusion packagers to package the extra libraries if possible, perhaps
> as a separate "nvidia-cuda" package?
This is easy to figure out if you have a NVIDIA card or not.
# yum whatprovides */libcuda.so.1
[snip long output]
1:xorg-x11-drv-nvidia-libs-260.19.29-2.fc14.x86_64 : Libraries for
: xorg-x11-drv-nvidia
Repo : rpmfusion-nonfree-updates
Matched from:
Filename : /usr/lib64/nvidia/libcuda.so.1
[snip more long output]
Any NVIDIA or CUDA related questions are best asked in areas[1] where
NVIDIA will see them. No one on this list can help you fix the errors.
[1] http://www.nvnews.net/vbulletin/forumdisplay.php?f=14
--
users mailing list
users@xxxxxxxxxxxxxxxxxxxxxxx
To unsubscribe or change subscription options:
https://admin.fedoraproject.org/mailman/listinfo/users
Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
