On 08/24/2015 02:56 PM, Gustav Degreef wrote:
Hello,
I recently upgraded my "testing" (that I use as the production system once it is stable) system from 13.1 to 13.2 online (not via DVD). I removed all repositories except oss and non-oss and changed to the newer versions. The upgrade seemed to go flawlessly, with no error messages. However, the system now only has the vesa driver working (I can get a low resolution desktop). I installed the nvidia drivers from the suse repos, but somehow the wrong drivers were installed (the g03 and g04 series). I have previously used the g02 series with multiple suse KDE installations on this laptop without any issues.
I removed the incorrect drivers and reinstalled the correct ones (three rpm's) a number of times and tried a number of fixes suggested in the SDB without success.
I am attaching parts of the xor.log and messages logs which seem to show that there was a mismatch in the expected driver 304.125 (which is what I have installed) and what the "client" is looking for (340.65).
I I am visually impaired and have difficulty looking through logs, so I don't know hot to be more precise.
It looks to me like the incorrect installation of the g03 or g04 series left something incorrect in place that prevents the loading of the now correctly installed kernel module, is that correct? Any Ideas on how to proceed other than re-installing 13.2? I need the nvidia driver, nouveau does not work well enough for a dual monitor setup I use. Thanks in advance, Gustav.
/var/log/xorg.olg extracts
[ 31.559] (II) Loading /usr/lib64/xorg/modules/extensions/libglx.so [ 32.337] (II) Module glx: vendor="NVIDIA Corporation" [ 32.337] compiled for 4.0.2, module version = 1.0.0 [ 32.337] Module class: X.Org Server Extension [ 32.337] (II) NVIDIA GLX Module 304.125 Mon Dec 1 20:22:48 PST 2014
32.339] (II) NVIDIA dlloader X Driver 340.65 Tue Dec 2 08:47:36 PST 2014 [ 32.339] (II) NVIDIA Unified Driver for all Supported NVIDIA GPUs
[ 32.341] (EE) NVIDIA: Failed to initialize the NVIDIA kernel module. Please see the [ 32.341] (EE) NVIDIA: system's kernel log for additional error messages and [ 32.341] (EE) NVIDIA: consult the NVIDIA README for details.
/var/log/messages extract
2015-06-27T02:06:39.135680+02:00 linux-aigg kernel: [ 20.393421] [drm] Initialized nvidia-drm 0.0.0 20130102 for 0000:01:00.0 on minor 0 2015-06-27T02:06:39.135681+02:00 linux-aigg kernel: [ 20.393432] NVRM: loading NVIDIA UNIX x86_64 Kernel Module 340.65 Tue Dec 2 09:50:34 PST 2014
2015-08-24T15:38:29.194968+02:00 linux-aigg kernel: [ 32.342614] NVRM: API mismatch: the client has the version 340.65, but 2015-08-24T15:38:29.194988+02:00 linux-aigg kernel: [ 32.342614] NVRM: this kernel module has the version 304.125. Please 2015-08-24T15:38:29.194990+02:00 linux-aigg kernel: [ 32.342614] NVRM: make sure that this kernel module and all NVIDIA driver 2015-08-24T15:38:29.194992+02:00 linux-aigg kernel: [ 32.342614] NVRM: components have the same version. 2015-08-24T15:38:29.195000+02:00 linux-aigg kernel: [ 32.343036] NVRM: API mismatch: the client has the version 340.65, but 2015-08-24T15:38:29.195002+02:00 linux-aigg kernel: [ 32.343036] NVRM: this kernel module has the version 304.125. Please 2015-08-24T15:38:29.195004+02:00 linux-aigg kernel: [ 32.343036] NVRM: make sure that this kernel module and all NVIDIA driver 2015-08-24T15:38:29.195006+02:00 linux-aigg kernel: [ 32.343036] NVRM: components have the same version.
I don't know if you've got this fixed yet, you've certainly got lots of help here, but, I also had some problems recently with the SuSE rpms for NVIDIA cards. Current kernel-desktop is revision 3.16.7-21.1 but the G02 NVIDIA kernel driver I needed is at 3.16.6_2-38.1 and will NOT work with the 3.16.7-21 kernel. I had to download the installer from NVIDIA to get it all working. Regards Mark -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org