How to determine which card I am using

Accelerated Computing CUDA CUDA Programming and Performance
1

My computer has two nvidia cards, quadro nvs290 and tesla 1060. I am already able to run some simple CUDA examples.

But I recently found the results of this command “lspci | grep -i nVidia” is

01:00.0 VGA compatible controller: nVidia Corporation Quadro NVS 290 (rev a1)
02:00.0 3D controller: nVidia Corporation Unknown device 05e7 (rev a1)

I did installed the tesla’s driver.

Does it mean that I am actually running my program on that quadro card? Is there any method I can determine which card I am using

Thanks in advance

2

I also checked NVIDIA X Server Settings in System->Preference

The two cards are both listed

3

There is nothing unusual about the lspci output. All it means is that the Tesla device ID hasn’t made its way into the device file that the linux pciutils uses.

The NVS290 is also CUDA capable, so I guess there is a possibility you are running your code on it rather than the Tesla. The SDK example deviceQuery will show the features of all the detected CUDA devices on your system, and you can use cudaChooseDevice() to explicitly ensure that your code get the Telsa rather than your display GPU.

4

Thank you for your help. I will try that