Welcome, Guest. Please login or register.
Did you miss your activation email?

Author Topic: [EN] [SOLVED] opencl / error loading nvidia_uvm  (Read 5416 times)

Miho

  • Guest
[EN] [SOLVED] opencl / error loading nvidia_uvm
« on: 2014/07/12, 14:36:14 »
Hi all,

I'm trying to get opencl running.
AFAIK the nvidia_uvm module ist required which can't be loaded allthough it was build.

Code: [Select]
modprobe: ERROR: could not insert 'nvidia_uvm': Invalid argument

journalctl:
Code: [Select]
Jul 12 13:28:52 rebel kernel: nvidia_uvm: no symbol version for nvUvmInterfaceChannelDestroy
Jul 12 13:28:52 rebel kernel: nvidia_uvm: Unknown symbol nvUvmInterfaceChannelDestroy (err -22)
Jul 12 13:28:52 rebel kernel: nvidia_uvm: no symbol version for nvUvmInterfaceQueryCaps
Jul 12 13:28:52 rebel kernel: nvidia_uvm: Unknown symbol nvUvmInterfaceQueryCaps (err -22)
Jul 12 13:28:52 rebel kernel: nvidia_uvm: no symbol version for nvUvmInterfaceMemoryAllocSys
Jul 12 13:28:52 rebel kernel: nvidia_uvm: Unknown symbol nvUvmInterfaceMemoryAllocSys (err -22)
Jul 12 13:28:52 rebel kernel: nvidia_uvm: no symbol version for nvUvmInterfaceMemoryCpuMap
Jul 12 13:28:52 rebel kernel: nvidia_uvm: Unknown symbol nvUvmInterfaceMemoryCpuMap (err -22)
Jul 12 13:28:52 rebel kernel: nvidia_uvm: no symbol version for nvUvmInterfaceKillChannel
Jul 12 13:28:52 rebel kernel: nvidia_uvm: Unknown symbol nvUvmInterfaceKillChannel (err -22)
Jul 12 13:28:52 rebel kernel: nvidia_uvm: no symbol version for nvUvmInterfaceAddressSpaceCreateMirrored
Jul 12 13:28:52 rebel kernel: nvidia_uvm: Unknown symbol nvUvmInterfaceAddressSpaceCreateMirrored (err -22)
Jul 12 13:28:52 rebel kernel: nvidia_uvm: no symbol version for nvUvmInterfaceServiceDeviceInterruptsRM
Jul 12 13:28:52 rebel kernel: nvidia_uvm: Unknown symbol nvUvmInterfaceServiceDeviceInterruptsRM (err -22)
Jul 12 13:28:52 rebel kernel: nvidia_uvm: no symbol version for nvUvmInterfaceDeRegisterUvmOps
Jul 12 13:28:52 rebel kernel: nvidia_uvm: Unknown symbol nvUvmInterfaceDeRegisterUvmOps (err -22)
Jul 12 13:28:52 rebel kernel: nvidia_uvm: no symbol version for nvUvmInterfaceGetUvmPrivRegion
Jul 12 13:28:52 rebel kernel: nvidia_uvm: Unknown symbol nvUvmInterfaceGetUvmPrivRegion (err -22)
Jul 12 13:28:52 rebel kernel: nvidia_uvm: no symbol version for nvUvmInterfaceGetAttachedUuids
Jul 12 13:28:52 rebel kernel: nvidia_uvm: Unknown symbol nvUvmInterfaceGetAttachedUuids (err -22)
Jul 12 13:28:52 rebel kernel: nvidia_uvm: no symbol version for nvUvmInterfaceSessionDestroy
Jul 12 13:28:52 rebel kernel: nvidia_uvm: Unknown symbol nvUvmInterfaceSessionDestroy (err -22)
Jul 12 13:28:52 rebel kernel: nvidia_uvm: no symbol version for nvUvmInterfaceCheckEccErrorSlowpath
Jul 12 13:28:52 rebel kernel: nvidia_uvm: Unknown symbol nvUvmInterfaceCheckEccErrorSlowpath (err -22)
Jul 12 13:28:52 rebel kernel: nvidia_uvm: no symbol version for nvUvmInterfaceCopyEngineAllocate
Jul 12 13:28:52 rebel kernel: nvidia_uvm: Unknown symbol nvUvmInterfaceCopyEngineAllocate (err -22)
Jul 12 13:28:52 rebel kernel: nvidia_uvm: no symbol version for nvUvmInterfaceAddressSpaceDestroy
Jul 12 13:28:52 rebel kernel: nvidia_uvm: Unknown symbol nvUvmInterfaceAddressSpaceDestroy (err -22)
Jul 12 13:28:52 rebel kernel: nvidia_uvm: no symbol version for nvUvmInterfaceRegisterUvmCallbacks
Jul 12 13:28:52 rebel kernel: nvidia_uvm: Unknown symbol nvUvmInterfaceRegisterUvmCallbacks (err -22)
Jul 12 13:28:52 rebel kernel: nvidia_uvm: no symbol version for nvUvmInterfaceGetGpuArch
Jul 12 13:28:52 rebel kernel: nvidia_uvm: Unknown symbol nvUvmInterfaceGetGpuArch (err -22)
Jul 12 13:28:52 rebel kernel: nvidia_uvm: no symbol version for nvUvmInterfaceChannelAllocate
Jul 12 13:28:52 rebel kernel: nvidia_uvm: Unknown symbol nvUvmInterfaceChannelAllocate (err -22)
Jul 12 13:28:52 rebel kernel: nvidia_uvm: no symbol version for nvUvmInterfaceSessionCreate
Jul 12 13:28:52 rebel kernel: nvidia_uvm: Unknown symbol nvUvmInterfaceSessionCreate (err -22)

dpkg -l | grep nvidia:
Code: [Select]
ii  glx-alternative-nvidia                                      0.4.1                             amd64        allows the selection of NVIDIA as GLX provider
ii  libgl1-nvidia-glx:amd64                                     340.24-0.siduction.1              amd64        NVIDIA binary OpenGL libraries
ii  libnvidia-compiler:amd64                                    340.24-0.siduction.1              amd64        NVIDIA runtime compiler library
ii  libnvidia-ml1:amd64                                         340.24-0.siduction.1              amd64        NVIDIA Management Library (NVML) runtime library
ii  nvidia-alternative                                          340.24-0.siduction.1              amd64        allows the selection of NVIDIA as GLX provider
ii  nvidia-driver                                               340.24-0.siduction.1              amd64        NVIDIA metapackage
ii  nvidia-installer-cleanup                                    20131102+1                        amd64        cleanup after driver installation with the nvidia-installer
ii  nvidia-kernel-common                                        20131102+1                        amd64        NVIDIA binary kernel module support files
ii  nvidia-kernel-dkms                                          340.24-0.siduction.1              amd64        NVIDIA binary kernel module DKMS source
ii  nvidia-libopencl1:amd64                                     340.24-0.siduction.1              amd64        NVIDIA OpenCL ICD Loader library
ii  nvidia-modprobe                                             334.16-1                          amd64        utility to load NVIDIA kernel modules and create device nodes
ii  nvidia-opencl-common                                        340.24-0.siduction.1              amd64        NVIDIA OpenCL driver
ii  nvidia-opencl-icd:amd64                                     340.24-0.siduction.1              amd64        NVIDIA OpenCL ICD
ii  nvidia-support                                              20131102+1                        amd64        NVIDIA binary graphics driver support files
ii  nvidia-vdpau-driver:amd64                                   340.24-0.siduction.1              amd64        NVIDIA vdpau driver
ii  xserver-xorg-video-nvidia                                   340.24-0.siduction.1              amd64        NVIDIA binary Xorg driver

uname -r:
Code: [Select]
3.15-5.towo-siduction-amd64
As far as I found, there was a bug, e.g. [1], but should have been resolved since version 331.79-1.
I tested the versions 331.79-1, 337.25-0... and also the driver directly from nvidia.com – all to no avail.

Code: [Select]
LANG=C apt-cache policy nvidia-driver:
nvidia-driver:
  Installed: 340.24-0.siduction.1
  Candidate: 340.24-0.siduction.1
  Version table:
 *** 340.24-0.siduction.1 0
        500 http://packages.siduction.org/fixes/ unstable/non-free amd64 Packages
        100 /var/lib/dpkg/status
     340.17-0.siduction.2 0
        500 http://packages.siduction.org/user/ unstable/non-free amd64 Packages
     337.25-0.siduction.3 0
        100 http://packages.siduction.org/experimental/ unstable/non-free amd64 Packages
     337.19-1 0
          1 http://http.debian.net/debian/ experimental/non-free amd64 Packages
     331.79-1 0
        500 http://ftp.at.debian.org/debian/ unstable/non-free amd64 Packages

I'm stuck and therefore I would appreciate any suggestions. Please ask if you need any further information.
thanks in advance and best regards

[1] https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=747366
« Last Edit: 2014/07/13, 11:36:24 by Miho »

Online towo

  • Administrator
  • User
  • *****
  • Posts: 2.948
Re: opencl / error loading nvidia_uvm
« Reply #1 on: 2014/07/12, 15:17:08 »
Pleas update your system, i have added the patch from your linked bugreport to the drivers in our fixes-repo.
Ich gehe nicht zum Karneval, ich verleihe nur manchmal mein Gesicht.

Miho

  • Guest
Re: opencl / error loading nvidia_uvm
« Reply #2 on: 2014/07/13, 11:34:46 »
Thanks a lot towo for the quick fix – appreciate it!

I can load the module by hand now and it seems to work but I still can't get opencl working. I'll have to dig into this for a sulotion to the problem. Until then I consider this topic solved.

best regards