CUDA driver version is insufficient for CUDA runtime version

hello everyone, I have some problem with my cuda,it is in the same situation with https://devtalk.nvidia.com/default/topic/617414/-solved-cuda-driver-version-is-insufficient-for-cuda-runtime-version-fedora-18-rpmfusion-driver/

But when i follow the solution from this topic ,it doesnot work.

thats what i got from his solution

[suice07@suice07-pc ~]$ cd /usr/lib64/nvidia
[suice07@suice07-pc nvidia]$ sudo ln -s libcuda.so.1 libcuda.so
[sudo] password for suice07: 
ln: failed to create symbolic link 'libcuda.so': File exists
[suice07@suice07-pc nvidia]$

thats what i got

th> require 'cunn'
THCudaCheck FAIL file=/tmp/luarocks_cutorch-scm-1-3884/cutorch/lib/THC/THCGeneral.c line=70 error=35 : CUDA driver version is insufficient for CUDA runtime version
/home/suice07/torch/install/share/lua/5.1/trepl/init.lua:389: /home/suice07/torch/install/share/lua/5.1/trepl/init.lua:389: cuda runtime error (35) : CUDA driver version is insufficient for CUDA runtime version at /tmp/luarocks_cutorch-scm-1-3884/cutorch/lib/THC/THCGeneral.c:70
stack traceback:
        [C]: in function 'error'
        /home/suice07/torch/install/share/lua/5.1/trepl/init.lua:389: in function 'require'
        [string "_RESULT={require 'cunn'}"]:1: in main chunk
        [C]: in function 'xpcall'
        /home/suice07/torch/install/share/lua/5.1/trepl/init.lua:661: in function 'repl'
        ...ce07/torch/install/lib/luarocks/rocks/trepl/scm-1/bin/th:204: in main chunk
        [C]: at 0x00405c90
[suice07@suice07-pc nvidia]$ nvidia-smi
Fri Jan  5 19:02:51 2018       
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 384.98                 Driver Version: 384.98                    |
|-------------------------------+----------------------+----------------------+
| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|===============================+======================+======================|
|   0  GeForce GT 740M     Off  | 00000000:01:00.0 N/A |                  N/A |
| N/A   41C    P0    N/A /  N/A |      0MiB /  2004MiB |     N/A      Default |
+-------------------------------+----------------------+----------------------+
                                                                               
+-----------------------------------------------------------------------------+
| Processes:                                                       GPU Memory |
|  GPU       PID   Type   Process name                             Usage      |
|=============================================================================|
|    0                    Not Supported                                       |
+-----------------------------------------------------------------------------+
[suice07@suice07-pc nvidia]$
[suice07@suice07-pc nvidia]$ lspci -v | grep -i nvidia
01:00.0 3D controller: NVIDIA Corporation GK208M [GeForce GT 740M] (rev a1)
        Kernel driver in use: nvidia
        Kernel modules: nouveau, nvidia_drm, nvidia

i am using manjaro,and the cudatoolkit 9.1

cuda toolkit 9.1 does not work with driver 384.98. The error message you are getting is expected.

If you want to install CUDA properly, start with a clean OS load, get your installers from here:

http://www.nvidia.com/getcuda

and follow the instructions in the linux install guide carefully:

http://docs.nvidia.com/cuda/cuda-installation-guide-linux/index.html

I also try the driver 384.111, does not work too. Should I get cuda 9.0? Cause i remember it works with 384.98.

CUDA 9.0 will work with 384.xx drivers.

thank you for the solution.
seems I have new problem

cp: error writing '/tmp/yaourt-tmp-suice07/aur-cuda-8.0/pkg/cuda-8.0/opt/cuda/bin/cuda-install-samples-8.0.sh': No space left on device
Logging to /tmp/cuda-installer-2208

========================================

Configuring samples Makefile...
cp: error writing 'uninstall_cuda_samples_8.0.pl': No space left on device

========================================

Please make sure that
 -   PATH includes /tmp/yaourt-tmp-suice07/aur-cuda-8.0/pkg/cuda-8.0/opt/cuda/bin
 -   LD_LIBRARY_PATH includes /tmp/yaourt-tmp-suice07/aur-cuda-8.0/pkg/cuda-8.0/opt/cuda/lib64, or, add /tmp/yaourt-tmp-suice07/aur-cuda-8.0/pkg/cuda-8.0/opt/cuda/lib64 to /etc/ld.so.conf and run ldconfig as root

To uninstall the NVIDIA CUDA Samples, run the uninstall script in /tmp/yaourt-tmp-suice07/aur-cuda-8.0/pkg/cuda-8.0/opt/cuda/samples
Installation Complete

install: error writing '/tmp/yaourt-tmp-suice07/aur-cuda-8.0/pkg/cuda-8.0/etc/profile.d/cuda.sh': No space left on device
==> ERROR: A failure occurred in package().
    Aborting...
==> ERROR: Makepkg was unable to build cuda-8.0.
==> Restart building cuda-8.0 ? [y/N]

this is what i got when i installing cuda 8.0…seems out of memory.

Nothing related to memory. Whatever volume your /tmp directory is on, maybe just your root (/) filesystem is full… or in other words, you are out of disk space on your storage device that has the OS install. Possibly the CUDA 9.1 / 9.0 installers dumped stuff on your /tmp directory that has remained on disk.

Hello everyone! I have nvidia GeForce 330 built in my notebook.This GPU is supported with drivers up to 340 version, but the latest cuda - cuda-9.1 requires nvidia-387 package (it appears during installation via deb.). So could you please help me to fix this or advise another version of cuda (which works with 340 drivers). My OS is Ubuntu 16.04 LTS

Hello everyone,

I have the same problem in win 10. I have Geforce 710M graphics card, I installed the cuda 9.2 for windows 10 and with cudaGetDeviceCount I got error number 35. Please help me what to do!

GeForce 710m is a Fermi device (compute capability 2.x)

Fermi devices are not supported by CUDA 9.x. The last CUDA version supporting those devices was CUDA 8.0.

Hello everyone! I have the same problem, I saw this comment:

However, i still got this problem.

+-----------------------------------------------------------------------------+
| NVIDIA-SMI 384.130                Driver Version: 384.130                   |
|-------------------------------+----------------------+----------------------+
| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|===============================+======================+======================|
|   0  GeForce GT 740M     Off  | 00000000:01:00.0 N/A |                  N/A |
| N/A   41C    P8    N/A /  N/A |    151MiB /  2001MiB |     N/A      Default |
+-------------------------------+----------------------+----------------------+
                                                                               
+-----------------------------------------------------------------------------+
| Processes:                                                       GPU Memory |
|  GPU       PID   Type   Process name                             Usage      |
|=============================================================================|
|    0                    Not Supported                                       |
+-----------------------------------------------------------------------------+
./deviceQuery Starting...

 CUDA Device Query (Runtime API) version (CUDART static linking)

Detected 1 CUDA Capable device(s)

Device 0: "GeForce GT 740M"
  CUDA Driver Version / Runtime Version          9.0 / 9.0
  CUDA Capability Major/Minor version number:    3.0

The error in python:

tensorflow.python.framework.errors_impl.InternalError: cudaGetDevice() failed. Status: CUDA driver version is insufficient for CUDA runtime version

Hello, i am also facing the same problem.
I have Nvidia driver version 384 with Cuda 9 and CuDnn 7. If anyone solved this error, please share the results. Thanks in advance.

rohit@rohit-MS-7A71:/usr/local/cuda-9.0/samples/1_Utilities/deviceQuery$ ./deviceQuery
./deviceQuery Starting…

CUDA Device Query (Runtime API) version (CUDART static linking)

cudaGetDeviceCount returned 35
-> CUDA driver version is insufficient for CUDA runtime version
Result = FAIL
nvidia-smi
Thu Apr 2 20:01:36 2020
±----------------------------------------------------------------------------+
| NVIDIA-SMI 384.130 Driver Version: 384.130 |
|-------------------------------±---------------------±---------------------+
| GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |
|===============================+======================+======================|
| 0 GeForce GTX 1080 Off | 00000000:01:00.0 On | N/A |
| 37% 36C P8 12W / 180W | 380MiB / 8110MiB | 0% Default |
±------------------------------±---------------------±---------------------+

±----------------------------------------------------------------------------+
| Processes: GPU Memory |
| GPU PID Type Process name Usage |
|=============================================================================|
| 0 1048 G /usr/lib/xorg/Xorg 308MiB |
| 0 2013 G compiz 69MiB |
±----------------------------------------------------------------------------+
rohit@rohit-MS-7A71:~ nvcc --version nvcc: NVIDIA (R) Cuda compiler driver Copyright (c) 2005-2017 NVIDIA Corporation Built on Fri_Sep__1_21:08:03_CDT_2017 Cuda compilation tools, release 9.0, V9.0.176 rohit@rohit-MS-7A71:~
rohit@rohit-MS-7A71:~$ cat /usr/local/cuda/include/cudnn.h | grep CUDNN_MAJOR -A 2
#define CUDNN_MAJOR 7
#define CUDNN_MINOR 3
#define CUDNN_PATCHLEVEL 1

#define CUDNN_VERSION (CUDNN_MAJOR * 1000 + CUDNN_MINOR * 100 + CUDNN_PATCHLEVEL)

#include “driver_types.h”