BIDData/BIDMach

0 CUDA devices found

Opened this issue · 4 comments

JoshuaC3 commented

I had the message 0 CUDA devices found within the tutorial notebooks. I think this then results in other issues like java.lang.RuntimeException: Couldnt find any files.

~$ nvcc -V
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2015 NVIDIA Corporation
Built on Tue_Aug_11_14:27:32_CDT_2015
Cuda compilation tools, release 7.5, V7.5.17

$ echo $LD_LIBRARY_PATH
/usr/local/cuda-7.5/lib64:

echo $PATH/usr/local/cuda-7.5/bin:

Any more information please ask!

GearFear commented

What's the output of nvidia-smi ?

JoshuaC3 commented

This was the return from nvidia-smi:

Thu Nov 17 12:02:21 2016       
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 367.48                 Driver Version: 367.48                    |
|-------------------------------+----------------------+----------------------+
| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|===============================+======================+======================|
|   0  Quadro K3100M       Off  | 0000:01:00.0      On |                  N/A |
| N/A   32C    P8     9W /  N/A |     46MiB /  4034MiB |     54%      Default |
+-------------------------------+----------------------+----------------------+

+-----------------------------------------------------------------------------+
| Processes:                                                       GPU Memory |
|  GPU       PID  Type  Process name                               Usage      |
|=============================================================================|
|    0      1019    G   /usr/lib/xorg/Xorg                              45MiB |
+-----------------------------------------------------------------------------+

I have made a few changes to nvidia drivers/installs and installed CUDA 8 for Ubuntu 16.04. I will have another crack at installing BIDMACH tonight as I have not tried again since the CUDA 8 update.

svanschalkwyk commented

[SOLVED]
Any success? Mine only detects one of my two devices.
I have export CUDA_VISIBLE_DEVICES=1,2 set.
Have to use the device number as reported by nvidia-smi. In my case
"export CUDA_VISIBLE_DEVICES=0,1"

svanschalkwyk commented

Try "export CUDa_VISIBLE_DEVICES=0"