Have error for run yolov4 darknet on manifold2 jetson tx2 without sudo

Hi!
I have a problem with launch yolov4 darknet on manifold 2G (jetson tx2).
I’am use darknet from this github(AlexeyAB): GitHub - AlexeyAB/darknet: YOLOv4 / Scaled-YOLOv4 / YOLO - Neural Networks for Object Detection (Windows and Linux version of Darknet )
command for launch:
./darknet detector test obj.data yolov4.cfg yolov4.weights image.jpg -out -dont_show

CUDA status Error: file: ./src/dark_cuda.c : () : line: 373 : build time: Oct 15 2020 - 14:57:53

CUDA Error: unknown error
CUDA Error: unknown error: Permission denied
darknet: ./src/utils.c:325: error: Assertion `0’ failed.
Aborted (core dumped)

if run this command with sudo privilege, will be ok.
how can I use darknet without sudo privilege?

1 Like

Hi,

Suppose you don’t need the root privilege for the darkenet app.

May I know the location you clone the darknet source?
You can try to clone the source to /home/nvidia/ to see if helps.

Thanks.

Hi!
Yes, I need the darknet app to work without su privileges
Sources darknet location in home dir.
image
The darknet compiled with parameters (Makefile):

GPU=1
CUDNN=1
CUDNN_HALF=0
OPENCV=1
AVX=0
OPENMP=0
LIBSO=1
ZED_CAMERA=0
ZED_CAMERA_v2_8=0
I tried compiling without CUDNN, but have same result

I use a built-in computer from DJI, this device contains a jetson TX2. But the OS (ubuntu) has some modifications: the default user is dji, not nvidia

Hi,

Please check if you have enabled the privilege of the customized user name.

Thanks.

Hi,
I check, yes, user is member of group “video”.
image
Thanks.

Hi,

Sorry for the late update.

We are going to reproduce this issue in our TX2 board.
Will share more information with you once we got a progress.

Thanks.

Hi,

We cannot reproduce this in our environment.
So this issue should come from the custom BSP of manifold2.

Have you checked this with the vender first?

Thanks.

Hi,
I also wrote a letter about this problem to the support service of DJI. but I haven’t received a response yet

There is no update from you for a period, assuming this is not an issue any more.
Hence we are closing this topic. If need further support, please open a new one.
Thanks

Hi,

Do you get any feedback from the DJI?

Thanks.

Hi!
I’m working on running yolo on Tx2 Dev kit. I have installed all of requirements for yolo, OpenCV, CUDA, Cudnn.
I’ve tested if the onboard camera’s working via gstreamer : gst-launch-1.0 nvarguscamerasrc ! nvvidconv ! xvimagesink
It works properly.
Then, I tried to run yolo by your code ./darknet detector demo cfg/coco.data cfg/yolo.cfg yolo.weights “nvcamerasrc ! video/x-raw(memory:NVMM), width=(int)1280, height=(int)720,format=(string)I420, framerate=(fraction)30/1 ! nvvidconv flip-method=0 ! video/x-raw, format=(string)BGRx ! videoconvert ! video/x-raw, format=(string)BGR ! appsink”
The result is “Video-stream stopped!”
How could i track the issue?
Thanks!