On 2018-02-05 13:25, Per Jessen wrote:
Carlos E. R. wrote:
On 2018-02-05 12:52, Peter Suetterlin wrote:
Per Jessen wrote:
Carlos E. R. wrote:
Current Operating System: Linux Telcontar 4.4.104-39-default #1 SMP
[snip]
Apparently everything is updated. I'm using the nvidia rpms, and I have reinstalled them twice:
Telcontar:~ # rpm -qa | grep -i nvidia nvidia-gfxG03-kmp-default-340.104_k4.4.76_1-11.1.x86_64 nvidia-uvm-gfxG03-kmp-default-340.104_k4.4.76_1-11.1.x86_64
That looks like the wrong kernel version - 4.4.76_1 ?
zypper lr --details
24 | download.nvidia.com-leap | nVidia Graphics Drivers | Yes | (r ) Yes | Yes | 99 | rpm-md | http://download.nvidia.com/opensuse/leap/42.3
Funny, in https://en.opensuse.org/SDB:NVIDIA_drivers it is a slightly different URL:
http://http.download.nvidia.com/opensuse/leap/42.3
The contents seem to be identical.
I'll try changing that.
What happens if you 'modprobe nvidia'?
Telcontar:/lib/modules # modprobe nvidia modprobe: ERROR: could not find module by name='nvidia' modprobe: ERROR: could not insert 'nvidia': Function not implemented
That should be read as "cannot find nvidia module that matches this kernel version".
Right.
Telcontar:/lib/modules #
I copy the tree from /lib/modules/4.4.104-18.44-default, and now I have:
According to your OP, you're running 4.4.104-39
Yes. But the nvidia module is for 4.4.76_1-11.1.x86_64, and the symlinks point to it: Telcontar:/lib/modules # l 4.4.104-39-default/weak-updates/updates/ total 8 drwxr-xr-x 2 root root 4096 Feb 4 20:16 ./ drwxr-xr-x 3 root root 4096 Jan 11 01:27 ../ lrwxrwxrwx 1 root root 47 Feb 5 13:03 nvidia.ko -> /lib/modules/4.4.76-1-default/updates/nvidia.ko lrwxrwxrwx 1 root root 51 Feb 5 13:03 nvidia-uvm.ko -> /lib/modules/4.4.76-1-default/updates/nvidia-uvm.ko Telcontar:/lib/modules # Maybe the links were not in place because the last thing I did yesterday was reinstall the entire OSS repo with YaST (all in this list, update unconditionally). I did not reinstall the nvidia drivers after that. I'm reinstalling them now, again, after changing the repo to point to the URL you wrote. There are no symptoms in YaST that it compiles anything. I look at the /var/log/YaST2/y2log and what I see is: 2018-02-05 13:45:53 <1> Telcontar(5069) [Ruby] clients/sw_single.rb:794 Summary dialog needed 2018-02-05 13:45:53 <2> Telcontar(5069) [Ruby] yast/convert.rb:73 Conversion from integer to float lead to loose precision. 2018-02-05 13:45:53 <2> Telcontar(5069) [Ruby] yast/convert.rb:73 Conversion from integer to float lead to loose precision. 2018-02-05 13:45:53 <1> Telcontar(5069) [Ruby] modules/PackagesUI.rb:635 Installation summary: <p><h3>Installation Successfully Finished</h3></p><p><h3>Packages</h3><ul><li>Updated Packages: 5<BR>nvidia-computeG03, nvidia-gfxG03-kmp-default, nvidia-glG03, nvidia-uvm-gfxG03-kmp-default, x11-video-nvidiaG03</li></ul></p><p><h3>Statistics</h3><ul><li>Elapsed Time: 03:01</li><li>Total Installed Size: 343.78 MiB</li><li>Total Downloaded Size: 70.42 MiB</li></ul></p><p><h3>Details</h3><ul><li><a href="install_log">Installation log</a></li></ul></p> 2018-02-05 13:45:53 <1> Telcontar(5069) [Ruby] modules/PackagesUI.rb:664 Displaying installation report: {"installed"=>0, "updated"=>5, "removed"=>0, "installed_list"=>[], "updated_list"=>["nvidia-gfxG03-kmp-default", "nvidia-uvm-gfxG03-kmp-default", "nvidia-glG03", "nvidia-computeG03", "x11-video-nvidiaG03"], "removed_list"=>[], "downloaded_bytes"=>73836279, "installed_bytes"=>360477836, "time_seconds"=>181.09334326199996, "success"=>true, "remaining"=>[], "install_log"=>"Downloading nvidia-gfxG03-kmp-default (download size 3.56 MiB)\nDownloading nvidia-uvm-gfxG03-kmp-default (download size 3.57 MiB)\nDownloading nvidia-glG03 (download size 27.56 MiB)\nDownloading nvidia-computeG03 (download size 17.41 MiB)\nDownloading x11-video-nvidiaG03 (download size 18.31 MiB)\nInstalling nvidia-gfxG03-kmp-default-340.104_k4.4.76_1-11.1.x86_64.rpm (installed size 28.03 MiB)\nInstalling nvidia-uvm-gfxG03-kmp-default-340.104_k4.4.76_1-11.1.x86_64.rpm (installed size 14.65 MiB)\nInstalling nvidia-glG03-340.104-11.1.x86_64.rpm (installed size 173.38 MiB)\nInstalling nvidia-computeG03-340.104-11.1.x86_64.rpm (installed size 55.74 MiB)\nInstalling x11-video-nvidiaG03-340.104-11.1.x86_64.rpm (installed size 71.99 MiB)\n"} 2018-02-05 13:46:26 <1> Telcontar(5069) [Ruby] modules/PackagesUI.rb:706 input: install_log 2018-02-05 13:46:27 <1> Telcontar(5069) [ui] YPushButton.cc(setFunctionKey):202 Guessing button role YOKButton for YPushButton "OK" at 0x7fe9797675a0 from function key F10 2018-02-05 13:47:03 <1> Telcontar(5069) [Ruby] modules/PackagesUI.rb:706 input: `next 2018-02-05 13:47:03 <1> Telcontar(5069) [Ruby] modules/PackagesUI.rb:746 Installation Summary result: `next I try "startx". Success! So the trick was that: reinstall everything from OSS (because I did not know what was bad), then reinstall again nvidia drivers, in that order. And the culprit was that the upgrade did not upgrade everything and in order. I'll explain why the upgrade did wrong, I have to retrieve the photos from the camera when system is fully running. Basically, the upgrade stopped and finished off when there were still *hundreds* of rpms that were not installed. Lets hop the logs have not rotated -I fear they have :-/ Yes, they have. The upgrade was finished 2018-02-04 18:52:55+01:00, and the earliest yast log is dated 2018-02-04 22:22:36 :-// It is absurd, the yast log rotation is way too aggressive, it only saves 4 MB of compressed logs :-/ -- Cheers / Saludos, Carlos E. R. (from 42.3 x86_64 "Malachite" (Minas Tirith))