I’m trying to troubleshoot a CUDA problem on a Ubuntu 16.04 LTS. I have a GTX 960 that I’d like to use with CUDA support (it’s for genome assembly), but I’ve hit a few snags.
When I installed the latest nvidia drivers, I got stuck in a login loop in Ubuntu, so I ended up removing the driver, and installing “nvidia-352” driver to get the OS working again. I then installed the latest CUDA toolkit, but I’m getting an error message now that there is no CUDA support for the card (CUDA_ERROR_NO_BINARY_FOR_GPU). Updating nvidia and CUDA drivers again hasn’t helped.
Long story short, I’m trying to find compatible versions of the nvidia driver and CUDA driver, to work with a GTX 960.
Any help would be much appreciated!
Here’s my nvidia-smi output:
Wed Feb 20 12:02:27 2019
±----------------------------------------------------------------------------+
| NVIDIA-SMI 410.79 Driver Version: 410.79 CUDA Version: 10.0 |
|-------------------------------±---------------------±---------------------+
| 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 960 Off | 00000000:09:00.0 On | N/A |
| 0% 35C P8 8W / 160W | 166MiB / 1999MiB | 0% Default |
±------------------------------±---------------------±---------------------+
±----------------------------------------------------------------------------+
| Processes: GPU Memory |
| GPU PID Type Process name Usage |
|=============================================================================|
| 0 1121 G /usr/lib/xorg/Xorg 90MiB |
| 0 2331 G compiz 72MiB |
±----------------------------------------------------------------------------+
And the error that I’m getting:
Loading fatbin file shared.fatbin failed with: CUDA error at /builds/ofan/ont_core_cpp/ont_core/common/CUDAHelper.cpp:40: CUDA_ERROR_NO_BINARY_FOR_GPU