link to nvstreammux cause an internal data flow error

Accelerated Computing Intelligent Video Analytics DeepStream SDK
1

hi.
I add nvstreammux and tiler in deepstream-test1. After processing one image, pad throw an internal data flow error.
pipeline: source–h264parser–nvdec–queue1–nvstreammux–nvinfer–tiler–queue3–filter–nvvidconv–filter2–nvosd–fakesink

error occur in queue1.src.
do you have any idea with it?
thanks.

2

Hi,

Could you check this topic first:
[url]https://devtalk.nvidia.com/default/topic/1037232/how-to-use-nvstreammux/[/url]

Thanks.

3

I had tried that, but it didn’t work, ending with ‘one element cannot be created’. gst-inspect-1.0 says no such element or plugin, ‘nvcuvid_h264dec’.

4

Hi

Have you applied this command in the README?

sudo ln -s /usr/lib/nvidia-<driver-version>/libnvcuvid.so \
    /usr/lib/x86_64-linux-gnu/libnvcuvid.so

Thanks.

5

Yes, I had applied that, and one more time after your answer. But that didn’t help.
Still confused, what’s the difference between nvcuvid_h264dec and nvdec_h264.
I found that error won’t occur if I link more than one src to mux using nvdec_h264. nvstreammux doesn’t support single channel? That’s weird.

6

Hi,

YES. Please use nvdec_h264 currently.
Thanks.

7

What is nvidia-?

$ sudo ln -s /usr/lib/<b>nvidia-<driver-version></b>/libnvcuvid.so \
>     /usr/lib/x86_64-linux-gnu/libnvcuvid.so
bash: driver-version: No such file or directory

$ nvidia-smi

Wed Aug 21 02:45:06 2019       
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 430.40       Driver Version: 430.40       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 1080    Off  | 00000000:01:00.0  On |                  N/A |
| 33%   44C    P5    12W / 180W |   1409MiB /  8118MiB |      0%      Default |
+-------------------------------+----------------------+----------------------+
                                                                               
+-----------------------------------------------------------------------------+
| Processes:                                                       GPU Memory |
|  GPU       PID   Type   Process name                             Usage      |
|=============================================================================|
|    0      1409      G   /usr/lib/xorg/Xorg                           496MiB |
|    0      1644      G   /usr/bin/gnome-shell                         338MiB |
|    0      9914      G   ...quest-channel-token=6768318621729276611   572MiB |
+-----------------------------------------------------------------------------+
suryadi@xtal:/usr/lib$ sudo ln -s /usr/lib/nvidia-430.40/libnvcuvid.so     /usr/lib/x86_64-linux-gnu/libnvcuvid.so
[sudo] password for suryadi: 
ln: failed to create symbolic link '/usr/lib/x86_64-linux-gnu/libnvcuvid.so': File exists

/usr/lib/x86_64-linux-gnu$ ls -la libnvcuvid.so
lrwxrwxrwx 1 root root 15 Jul 31 13:15 libnvcuvid.so -> libnvcuvid.so.1

How to rectify this problem?
Thanks.

Suryadi