How to update the nvidia driver on PX2

Hi,

  1. How can i get the information about the nvidia driver on PX2. We can get the information by using “nvidia-smi” on x86 . The nvidia driver information on x86 is as follows:

±----------------------------------------------------------------------------+
| NVIDIA-SMI 418.56 Driver Version: 418.56 CUDA Version: 10.1 |
|-------------------------------±---------------------±---------------------+
| 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 105… Off | 00000000:01:00.0 On | N/A |
| 25% 34C P8 N/A / 75W | 611MiB / 4038MiB | 0% Default |
±------------------------------±---------------------±---------------------+

±----------------------------------------------------------------------------+
| Processes: GPU Memory |
| GPU PID Type Process name Usage |
|=============================================================================|
| 0 1310 G /usr/lib/xorg/Xorg 374MiB |
| 0 2703 G compiz 20MiB |
| 0 7553 G /usr/lib/firefox/firefox 1MiB |
| 0 21939 G …-token=BA591A318972537F9F871D5934279122 37MiB |
| 0 30475 G /usr/lib/xorg/Xorg 47MiB |
±----------------------------------------------------------------------------+
We can get the version is “Driver Version: 418.56”.
Is there any similar way to get the information that on PX2.

  1. And How to update the nvidia driver of PX2 except by SDKmanager.

Could someone give me some guidance.

Thanks

Dear lyzeng,

May I know if you have any nvidia driver issue on DrivePX2?
Because there is no way to get nvidia driver info on DrivePX2 like nvidia-smi on x86.
You can get some dGPU/iGPU info with deviceQuery included in CUDA samples. Thanks.

Hi SteveNV,

Thanks for your fast reply. There is no issue on my DrivePX2. But i want to use tensorrt5 on PX2. I know that PX2 only support is tensorrt4. But we still want to use tensorrt5 on PX2. By now we have update all the tensorrt libs and cuda libs, and other necessary environment.
Then we run the cuda sample “./deviceQuery” in folder:
mywork/local/cuda-10.0/samples/1_Utilities/deviceQuery/
and we got the message:
./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

So how can we solve this problem?

Thanks

Dear lyzeng,

Unfortunately, up to TRT4.0 is supported for DPX2. Please refer to the following links. Thanks.
https://docs.nvidia.com/deeplearning/sdk/tensorrt-archived/index.html#trt_4
https://devtalk.nvidia.com/default/topic/1044167/general/a-few-questions-about-tensorrt-versioning/