为什么我的 Jetson Nano 不能运行 deepstream-test1

该问题发给 NVIDIA 论坛，下面是与管理员的交流内容：

https://forums.developer.nvidia.com/t/can-deepstream-test1-run-on-jetson-nano-2g/179912

xuyeping

Hi，

After I compile deepstream-test1 directly, the program can’t run on my Jetson Nano 2G. However, after removing the deep learning model nvinferfrom the codes, the program can run. Is it because my nano 2G memory is too small, so the deep learning model can’t run? Or other reasons?

Fiona.Chen

What kind of error have you met?

xuyeping

When I enter the command line ./deepstream-test1-app sample_720p.h264, the window deepstream-test1-app was showed. However, the window only draws the title and the border, the image is not displayed in it. The canvas of the window we saw is the desktop of Jetson Nano. The output of this program is as follows:

jetson@jetson-desktop:/opt/nvidia/deepstream/deepstream/sources/apps/sample_apps/deepstream-test1$ ./deepstream-test1-app sample_720p.h264 
Now playing: sample_720p.h264Using winsys: x11 
Opening in BLOCKING MODE 
0:00:11.198570435  8436   0x55934fcad0 INFO                 nvinfer gstnvinfer.cpp:619:gst_nvinfer_logger:<primary-nvinference-engine> NvDsInferContext[UID 1]: Info from NvDsInferContextImpl::deserializeEngineAndBackend() <nvdsinfer_context_impl.cpp:1702> [UID = 1]: deserialized trt engine from :/opt/nvidia/deepstream/deepstream-5.1/samples/models/Primary_Detector/resnet10.caffemodel_b1_gpu0_fp16.engine
INFO: [Implicit Engine Info]: layers num: 3
0   INPUT  kFLOAT input_1         3x368x640       
1   OUTPUT kFLOAT conv2d_bbox     16x23x40        
2   OUTPUT kFLOAT conv2d_cov/Sigmoid 4x23x40         0:00:11.218911904  8436   0x55934fcad0 INFO                 nvinfer gstnvinfer.cpp:619:gst_nvinfer_logger:<primary-nvinference-engine> NvDsInferContext[UID 1]: Info from NvDsInferContextImpl::generateBackendContext() <nvdsinfer_context_impl.cpp:1806> [UID = 1]: Use deserialized engine model: /opt/nvidia/deepstream/deepstream-5.1/samples/models/Primary_Detector/resnet10.caffemodel_b1_gpu0_fp16.engine
0:00:11.316350994  8436   0x55934fcad0 INFO                 nvinfer gstnvinfer_impl.cpp:313:notifyLoadModelStatus:<primary-nvinference-engine> [UID 1]: Load new model:dstest1_pgie_config.txt sucessfully
Running...
NvMMLiteOpen : Block : BlockType = 261 
NVMEDIA: Reading vendor.tegra.display-size : status: 6 
NvMMLiteBlockCreate : Block : BlockType = 261 
ERROR: [TRT]: ../rtSafe/cuda/reformat.cu (925) - Cuda Error in NCHWToNCHHW2: 719 (unspecified launch failure)
ERROR: [TRT]: FAILED_EXECUTION: std::exception
ERROR: Failed to enqueue inference batch
ERROR: Infer context enqueue buffer failed, nvinfer error:NVDSINFER_TENSORRT_ERROR
0:00:12.668284961  8436   0x5593091de0 WARN                 nvinfer gstnvinfer.cpp:1225:gst_nvinfer_input_queue_loop:<primary-nvinference-engine> error: Failed to queue input batch for inferencing
ERROR: Failed to make stream wait on event, cuda err_no:4, err_str:cudaErrorLaunchFailure
ERROR: Preprocessor transform input data failed., nvinfer error:NVDSINFER_CUDA_ERROR
ERROR from element primary-nvinference-engine: Failed to queue input batch for inferencing
0:00:12.668905003  8436   0x5593091de0 WARN                 nvinfer gstnvinfer.cpp:1225:gst_nvinfer_input_queue_loop:<primary-nvinference-engine> error: Failed to queue input batch for inferencing
ERROR: Failed to make stream wait on event, cuda err_no:4, err_str:cudaErrorLaunchFailure
ERROR: Preprocessor transform input data failed., nvinfer error:NVDSINFER_CUDA_ERROR
0:00:12.669423588  8436   0x5593091de0 WARN                 nvinfer gstnvinfer.cpp:1225:gst_nvinfer_input_queue_loop:<primary-nvinference-engine> error: Failed to queue input batch for inferencing
ERROR: Failed to make stream wait on event, cuda err_no:4, err_str:cudaErrorLaunchFailure
ERROR: Preprocessor transform input data failed., nvinfer error:NVDSINFER_CUDA_ERROR
0:00:12.669702021  8436   0x5593091de0 WARN                 nvinfer gstnvinfer.cpp:1225:gst_nvinfer_input_queue_loop:<primary-nvinference-engine> error: Failed to queue input batch for inferencing
Error details: /dvs/git/dirty/git-master_linux/deepstream/sdk/src/gst-plugins/gst-nvinfer/gstnvinfer.cpp(1225): gst_nvinfer_input_queue_loop (): /GstPipeline:dstest1-pipeline/GstNvInfer:primary-nvinference-engine
Returned, stopping playback
Frame Number = 0 Number of objects = 0 Vehicle Count = 0 Person Count = 0** (deepstream-test1-app:8436): WARNING **: 14:51:19.402: Use gst_egl_image_allocator_alloc() to allocate from this allocator
0:00:13.099598980  8436   0x5593091c00 WARN                 nvinfer gstnvinfer.cpp:1984:gst_nvinfer_output_loop:<primary-nvinference-engine> error: Internal data stream error.
0:00:13.099640229  8436   0x5593091c00 WARN                 nvinfer gstnvinfer.cpp:1984:gst_nvinfer_output_loop:<primary-nvinference-engine> error: streaming stopped, reason error (-5)
Frame Number = 1 Number of objects = 0 Vehicle Count = 0 Person Count = 0** (deepstream-test1-app:8436): WARNING **: 14:51:19.417: Use gst_egl_image_allocator_alloc() to allocate from this allocator
Frame Number = 2 Number of objects = 0 Vehicle Count = 0 Person Count = 0** (deepstream-test1-app:8436): WARNING **: 14:51:19.429: Use gst_egl_image_allocator_alloc() to allocate from this allocator
Frame Number = 3 Number of objects = 0 Vehicle Count = 0 Person Count = 0** (deepstream-test1-app:8436): WARNING **: 14:51:19.442: Use gst_egl_image_allocator_alloc() to allocate from this allocator

Fiona.Chen

I just tried with Nano board with deepstream-test1-app, it can run and I can see the video displayed.

Can you provide platform information?

xuyeping

Thank you so much! My running environment is as follows:

The information displayed in jtop is shown in the figure below：

Fiona.Chen

Deepstream5.1 is based on JetPack 4.5.1 GA(corresponding to L4T 32.5.1 release) .

Please install Deepstream5.1 and the dependencies correctly.

https://docs.nvidia.com/metropolis/deepstream/dev-guide/text/DS_Quickstart.html#jetson-setup

按照 Fiona.Chen 的答复，问题原因应该是我的 Jetson Nano 相关软件版本不匹配所致。我会按照所提的 URL 网址下载系统镜像，重新制作系统。稍后我会把实验结果详细写成文章与大家共享！