ERROR from dsexample0: surface-gpu-id=1,dsexample0-gpu-id=0

Accelerated Computing Intelligent Video Analytics DeepStream SDK
1

Hi,
I set gpu-id=1 in configfile.And I use GPU No.1 to run my deepstream-app program smoothly.
But when I add [ds-example] in my configfile.
An error message appears：
ERROR from dsexample0: Input surface gpu-id doesnt match with configured gpu-id for element, please allocate input using unified memory, or use same gpu-ids
Debug info: gstdsexample.cpp(904): gst_dsexample_transform_ip (): /GstPipeline:pipeline/GstBin:dsexample_bin/GstDsExample:dsexample0:
surface-gpu-id=1,dsexample0-gpu-id=0

but I set gpu-id=1 in my configfile [ds-example] like this:
enable=1
full-frame=0
unique-id=0
gpu-id=1

3

In which platform? dGPU or Jetson?

4

dGPU tesla T4

5

Can you set “nvbuf-memory-type=3” in [ds-example] group?

6

I’ve tried with T4, the following config works fine with gpu-id=1 settings.

[ds-example0]
ensble=1
full-frame=0
unique-id=1
gpu-id=1

8

give [ds-example] a number,like this [ds-example0].it works!
Thanks a lot!