I use docker container to run all this.
My Host:
- Hardware Platform : x86_64 + RTX2080Ti
- NVIDIA GPU Driver Version: 530.41.03
Docker Container:
base image: PyTorch 22.12 On NGC
The following components are included:
- Ubuntu 20.04 including Python 3.8
- NVIDIA CUDA® 11.8.0
- NVIDIA cuBLAS 11.11.3.6
- NVIDIA cuDNN 8.7.0.84
- NVIDIA TensorRT™ 8.5.1
Build my Dockerfile via base image:
I wrote the following Dockerfile through the official guidelines to install deepstream 6.2.
Dockerfile (8.7 KB)
Docker run :
xhost +
docker run -itd --gpus all --name aiaio-22.12 --ipc=host -e DISPLAY=$DISPLAY --device /dev/snd -v /tmp/.X11-unix/:/tmp/.X11-unix aiaio:22.12
My x11 forwarding is fine because I can run xclock to display a clock on my host.
Issue:
When I run the official sample program, only a black window pops up without any content.(I think it’s a problem with the libEGL library, but I don’t know how to fix it)
root@ee30e84098b6:/opt/nvidia/deepstream/deepstream-6.2/samples/configs/deepstream-app# deepstream-app -c source4_1080p_dec_infer-resnet_tracker_sgie_tiled_display_int8.txt
libEGL warning: DRI3: Screen seems not DRI3 capable
libEGL warning: DRI2: failed to authenticate
0:00:03.568952466 367 0x5638a1a23300 INFO nvinfer gstnvinfer.cpp:680:gst_nvinfer_logger:<secondary_gie_2> NvDsInferContext[UID 6]: Info from NvDsInferContextImpl::deserializeEngineAndBackend() <nvdsinfer_context_impl.cpp:1909> [UID = 6]: deserialized trt engine from :/opt/nvidia/deepstream/deepstream-6.2/samples/configs/deepstream-app/../../models/Secondary_CarMake/resnet18.caffemodel_b16_gpu0_int8.engine
INFO: ../nvdsinfer/nvdsinfer_model_builder.cpp:610 [Implicit Engine Info]: layers num: 2
0 INPUT kFLOAT input_1 3x224x224
1 OUTPUT kFLOAT predictions/Softmax 20x1x1
0:00:03.597210407 367 0x5638a1a23300 INFO nvinfer gstnvinfer.cpp:680:gst_nvinfer_logger:<secondary_gie_2> NvDsInferContext[UID 6]: Info from NvDsInferContextImpl::generateBackendContext() <nvdsinfer_context_impl.cpp:2012> [UID = 6]: Use deserialized engine model: /opt/nvidia/deepstream/deepstream-6.2/samples/configs/deepstream-app/../../models/Secondary_CarMake/resnet18.caffemodel_b16_gpu0_int8.engine
0:00:03.621300772 367 0x5638a1a23300 INFO nvinfer gstnvinfer_impl.cpp:328:notifyLoadModelStatus:<secondary_gie_2> [UID 6]: Load new model:/opt/nvidia/deepstream/deepstream-6.2/samples/configs/deepstream-app/config_infer_secondary_carmake.txt sucessfully
0:00:05.253186016 367 0x5638a1a23300 INFO nvinfer gstnvinfer.cpp:680:gst_nvinfer_logger:<secondary_gie_1> NvDsInferContext[UID 5]: Info from NvDsInferContextImpl::deserializeEngineAndBackend() <nvdsinfer_context_impl.cpp:1909> [UID = 5]: deserialized trt engine from :/opt/nvidia/deepstream/deepstream-6.2/samples/configs/deepstream-app/../../models/Secondary_CarColor/resnet18.caffemodel_b16_gpu0_int8.engine
INFO: ../nvdsinfer/nvdsinfer_model_builder.cpp:610 [Implicit Engine Info]: layers num: 2
0 INPUT kFLOAT input_1 3x224x224
1 OUTPUT kFLOAT predictions/Softmax 12x1x1
0:00:05.281197712 367 0x5638a1a23300 INFO nvinfer gstnvinfer.cpp:680:gst_nvinfer_logger:<secondary_gie_1> NvDsInferContext[UID 5]: Info from NvDsInferContextImpl::generateBackendContext() <nvdsinfer_context_impl.cpp:2012> [UID = 5]: Use deserialized engine model: /opt/nvidia/deepstream/deepstream-6.2/samples/configs/deepstream-app/../../models/Secondary_CarColor/resnet18.caffemodel_b16_gpu0_int8.engine
0:00:05.282464598 367 0x5638a1a23300 INFO nvinfer gstnvinfer_impl.cpp:328:notifyLoadModelStatus:<secondary_gie_1> [UID 5]: Load new model:/opt/nvidia/deepstream/deepstream-6.2/samples/configs/deepstream-app/config_infer_secondary_carcolor.txt sucessfully
0:00:06.898347201 367 0x5638a1a23300 INFO nvinfer gstnvinfer.cpp:680:gst_nvinfer_logger:<secondary_gie_0> NvDsInferContext[UID 4]: Info from NvDsInferContextImpl::deserializeEngineAndBackend() <nvdsinfer_context_impl.cpp:1909> [UID = 4]: deserialized trt engine from :/opt/nvidia/deepstream/deepstream-6.2/samples/configs/deepstream-app/../../models/Secondary_VehicleTypes/resnet18.caffemodel_b16_gpu0_int8.engine
INFO: ../nvdsinfer/nvdsinfer_model_builder.cpp:610 [Implicit Engine Info]: layers num: 2
0 INPUT kFLOAT input_1 3x224x224
1 OUTPUT kFLOAT predictions/Softmax 6x1x1
0:00:06.927166384 367 0x5638a1a23300 INFO nvinfer gstnvinfer.cpp:680:gst_nvinfer_logger:<secondary_gie_0> NvDsInferContext[UID 4]: Info from NvDsInferContextImpl::generateBackendContext() <nvdsinfer_context_impl.cpp:2012> [UID = 4]: Use deserialized engine model: /opt/nvidia/deepstream/deepstream-6.2/samples/configs/deepstream-app/../../models/Secondary_VehicleTypes/resnet18.caffemodel_b16_gpu0_int8.engine
0:00:06.928422698 367 0x5638a1a23300 INFO nvinfer gstnvinfer_impl.cpp:328:notifyLoadModelStatus:<secondary_gie_0> [UID 4]: Load new model:/opt/nvidia/deepstream/deepstream-6.2/samples/configs/deepstream-app/config_infer_secondary_vehicletypes.txt sucessfully
gstnvtracker: Loading low-level lib at /opt/nvidia/deepstream/deepstream/lib/libnvds_nvmultiobjecttracker.so
gstnvtracker: Batch processing is ON
gstnvtracker: Past frame output is ON
[NvMultiObjectTracker] Initialized
0:00:08.608975900 367 0x5638a1a23300 INFO nvinfer gstnvinfer.cpp:680:gst_nvinfer_logger:<primary_gie> NvDsInferContext[UID 1]: Info from NvDsInferContextImpl::deserializeEngineAndBackend() <nvdsinfer_context_impl.cpp:1909> [UID = 1]: deserialized trt engine from :/opt/nvidia/deepstream/deepstream-6.2/samples/configs/deepstream-app/../../models/Primary_Detector/resnet10.caffemodel_b4_gpu0_int8.engine
INFO: ../nvdsinfer/nvdsinfer_model_builder.cpp:610 [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
0:00:08.637573388 367 0x5638a1a23300 INFO nvinfer gstnvinfer.cpp:680:gst_nvinfer_logger:<primary_gie> NvDsInferContext[UID 1]: Info from NvDsInferContextImpl::generateBackendContext() <nvdsinfer_context_impl.cpp:2012> [UID = 1]: Use deserialized engine model: /opt/nvidia/deepstream/deepstream-6.2/samples/configs/deepstream-app/../../models/Primary_Detector/resnet10.caffemodel_b4_gpu0_int8.engine
0:00:08.638605999 367 0x5638a1a23300 INFO nvinfer gstnvinfer_impl.cpp:328:notifyLoadModelStatus:<primary_gie> [UID 1]: Load new model:/opt/nvidia/deepstream/deepstream-6.2/samples/configs/deepstream-app/config_infer_primary.txt sucessfully
Runtime commands:
h: Print this help
q: Quit
p: Pause
r: Resume
NOTE: To expand a source in the 2D tiled display and view object details, left-click on the source.
To go back to the tiled display, right-click anywhere on the window.
**PERF: FPS 0 (Avg) FPS 1 (Avg) FPS 2 (Avg) FPS 3 (Avg)
**PERF: 0.00 (0.00) 0.00 (0.00) 0.00 (0.00) 0.00 (0.00)
** INFO: <bus_callback:239>: Pipeline ready
cuGraphicsGLRegisterBuffer failed with error(304) gst_eglglessink_cuda_init texture = 1
** INFO: <bus_callback:225>: Pipeline running
**PERF: 758.05 (1.46) 758.05 (1.46) 758.05 (1.46) 631.71 (1.21)
nvstreammux: Successfully handled EOS for source_id=2
nvstreammux: Successfully handled EOS for source_id=1
nvstreammux: Successfully handled EOS for source_id=0
nvstreammux: Successfully handled EOS for source_id=3
ERROR from secondary_gie_bin_queue: Internal data stream error.
Debug info: gstqueue.c(988): gst_queue_handle_sink_event (): /GstPipeline:pipeline/GstBin:secondary_gie_bin/GstQueue:secondary_gie_bin_queue:
streaming stopped, reason not-negotiated (-4)
Quitting
[NvMultiObjectTracker] De-initialized
App run failed
btw, I run deepstream6.1 on nvcr.io/nvidia/deepstream:6.1.1-samples image everything works fine.