GPU not supported in Nvidia-smi

Accelerated Computing CUDA CUDA Setup and Installation
1

Sat Feb 24 16:38:21 2018
±----------------------------------------------------------------------------+
| NVIDIA-SMI 387.34 Driver Version: 387.34 |
|-------------------------------±---------------------±---------------------+
| 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 880M Off | 00000000:01:00.0 N/A | N/A |
| N/A 43C P8 N/A / N/A | 329MiB / 8109MiB | N/A Default |
±------------------------------±---------------------±---------------------+
| 1 GeForce GTX 880M Off | 00000000:02:00.0 N/A | N/A |
| N/A 36C P8 N/A / N/A | 0MiB / 8111MiB | N/A Default |
±------------------------------±---------------------±---------------------+

±----------------------------------------------------------------------------+
| Processes: GPU Memory |
| GPU PID Type Process name Usage |
|=============================================================================|
| 0 Not Supported |
| 1 Not Supported |
±----------------------------------------------------------------------------+

How to get my GPUs supported? Thanks!

Screenshot from 2018-02-24 16-47-06.png
Screenshot from 2018-02-24 16-47-06.png714×353 35.2 KB

2

The GTX 880M is a device with compute capability 3.0, and is supported by currently shipping CUDA components, tools, and drivers.

The output shown above does not show that the GTX 880M is not supported. It merely shows that the display of processes that use the GPU is not supported. Which I believe is standard nvidia-smi behavior for consumer-grade GPUs, and has been for many years. In other words, you would have to use a professional-grade GPU (Quadro or Tesla) if you need this information in the output of nvidia-smi.

3

Thanks very much.

4

Hi,
I also have the problem.GPU not supported in Nvidia -smi. And I use the GPU 820M to run tensorflow-gpu. it seems use the CPU to work, because the CPU has a very high usage rate when the code is running.So, I wonder thhat is the GPU working ?
I install tensorflow-gpu, cuda and cunn, and I can run tensoflow. But it seems use the CPU. I’m very confused.