how to set different gpu id?

Accelerated Computing Intelligent Video Analytics DeepStream SDK
1

Hi,
I tried to set the gpu-id 1:

[property]
gpu-id=1
net-scale-factor=0.0039215697906911373
model-file=../../../../samples/models/Primary_Detector/resnet10.caffemodel
proto-file=../../../../samples/models/Primary_Detector/resnet10.prototxt
labelfile-path=../../../../samples/models/Primary_Detector/labels.txt
int8-calib-file=../../../../samples/models/Primary_Detector/cal_trt4.bin
batch-size=1
network-mode=1
num-detected-classes=4
interval=0
gie-unique-id=1
parse-func=4
output-blob-names=conv2d_bbox;conv2d_cov/Sigmoid

[class-attrs-all]
threshold=0.2
eps=0.2
group-threshold=1

When I run deepstream-test1-app, I get an error message:

./deepstream-test1-app ../../../../samples/streams/sample_720p.h264
Now playing: ../../../../samples/streams/sample_720p.h264
>>> Generating new TRT model engine
Using INT8 data type.

 ***** Storing serialized engine file as /opt/DeepStream_Release/sources/apps/sample_apps/deepstream-test1/../../../../samples/models/Primary_Detector/resnet10.caffemodel_b1_int8.engine batchsize = 1 *****

Running...
ERROR from element primary-nvinference-engine: Input surface gpu-id doesnt match with configured gpu-id for element, please allocate input using unified memory, or use same gpu-ids
Error details: gstnvinfer.c(1929): gst_nvinfer_prepare (): /GstPipeline:dstest1-pipeline/GstNvInfer:primary-nvinference-engine:
surface-gpu-id=0,primary-nvinference-engine-gpu-id=1
Returned, stopping playback
Deleting pipeline

where do I set the surface-gpu-id?

2

Did you change deepstream-test1-app source code ?

3

yes, I have changed the code as follow:

/* Finally render the osd output */
//sink = gst_element_factory_make ("nveglglessink", "nvvideo-renderer");
sink = gst_element_factory_make ("fakesink", "nvvideo-renderer");

ennn, I have soluted the issue:

g_object_set (G_OBJECT (decoder), "gpu-id", 1, NULL);

Now, it’s all fine.