Forums - Access Adreno GPU inside Docker

3 posts / 0 new
Last post
Access Adreno GPU inside Docker
marcelo.benencase
Join Date: 25 Sep 23
Posts: 4
Posted: Mon, 2023-09-25 13:17

I have an EB5 device with an Adreno 650 GPU.

 

I'm able to run normally my object detection model outside a Docker environment using SNPE SDK.

However, when I try to run the same software I found a problem. After inspecting it deeper, I tried to run the following:

./snpe-platform-validator --runtime gpu --debug

This script is present in the SNPE SDK.

Outside the Docker, it doesn't throw any error, inside Docker I receive:

PF_VALIDATOR: DEBUG: Calling PlatformValidator->setRuntime
PF_VALIDATOR: DEBUG: Calling PlatformValidator->RuntimeCheck
PF_VALIDATOR: ERROR: Failed to query OpenCL platform.
Unit Test on the runtime GPU: Failed.
SNPE is NOT supported for runtime GPU on the device.
PF_VALIDATOR: DEBUG: Calling PlatformValidator->IsRuntimeAvailable
PF_VALIDATOR: ERROR: Could not find libOpenCL.so or libOpenCL_system.so.
Runtime GPU Prerequisites: Absent.
Error in saving the results 


I tried to install libOpenCL.so as follows: sudo apt install ocl-icd-opencl-dev

The message changed to:
 
PF_VALIDATOR: DEBUG: Calling PlatformValidator->setRuntime
PF_VALIDATOR: DEBUG: Calling PlatformValidator->RuntimeCheck
PF_VALIDATOR: ERROR: Failed to query OpenCL platform.
Unit Test on the runtime GPU: Failed.
SNPE is NOT supported for runtime GPU on the device.
PF_VALIDATOR: DEBUG: Calling PlatformValidator->IsRuntimeAvailable
Runtime GPU Prerequisites: Present.
Error in saving the results 

 

Did anyone already face and was able to overcome this?

  • Up0
  • Down0
abnave.m
Join Date: 12 Sep 23
Posts: 1
Posted: Thu, 2023-10-05 03:22

Did you have any luck with this?!! I encountered exactly same issue. Not able to access GPU from inside docker container. 

  • Up0
  • Down0
marcelo.benencase
Join Date: 25 Sep 23
Posts: 4
Posted: Fri, 2023-10-06 05:25

No, I'm still on the same page.

  • Up0
  • Down0
or Register

Opinions expressed in the content posted here are the personal opinions of the original authors, and do not necessarily reflect those of Qualcomm Incorporated or its subsidiaries (“Qualcomm”). The content is provided for informational purposes only and is not meant to be an endorsement or representation by Qualcomm or any other party. This site may also provide links or references to non-Qualcomm sites and resources. Qualcomm makes no representations, warranties, or other commitments whatsoever about any non-Qualcomm sites or third-party resources that may be referenced, accessible from, or linked to this site.