Parameter check failed at: runtime/api/executionContext.cpp::enqueueInternal::322, condition: bindings[x] != nullptr

[mmksir]

When i run "./build/segment ./config/config.yaml", i get an Error "[E] [TRT] 3: [executionContext.cpp::enqueueInternal::322] Error Code 3: Internal Error (Parameter check failed at: runtime/api/executionContext.cpp::enqueueInternal::322, condition: bindings[x] != nullptr)", and what maybe the reason? If i get the right ONNX and tensorrt_model_bin? This is my output information of running "build_model" and "demo":

./build/build_model ./config/config.yaml

~/Solov2-TensorRT-CPP/cmake-build-debug/build_model ./config/config.yaml config_file:./config/config.yaml createInferBuilder [05/25/2022-22:57:19] [I] [TRT] [MemUsageChange] Init CUDA: CPU +299, GPU +0, now: CPU 301, GPU 309 (MiB) createNetwork createBuilderConfig createParser parseFromFile:~/Solov2-TensorRT-CPP/ONNX/SOLOv2_light_R34.onnx [05/25/2022-22:57:19] [I] [TRT] ---------------------------------------------------------------- [05/25/2022-22:57:19] [I] [TRT] Input filename: ~/Solov2-TensorRT-CPP/ONNX/SOLOv2_light_R34.onnx [05/25/2022-22:57:19] [I] [TRT] ONNX IR version: 0.0.4 [05/25/2022-22:57:19] [I] [TRT] Opset version: 11 [05/25/2022-22:57:19] [I] [TRT] Producer name: pytorch [05/25/2022-22:57:19] [I] [TRT] Producer version: 1.3 [05/25/2022-22:57:19] [I] [TRT] Domain:
[05/25/2022-22:57:19] [I] [TRT] Model version: 0 [05/25/2022-22:57:19] [I] [TRT] Doc string:
[05/25/2022-22:57:19] [I] [TRT] ---------------------------------------------------------------- [05/25/2022-22:57:19] [W] [TRT] onnx2trt_utils.cpp:364: Your ONNX model has been generated with INT64 weights, while TensorRT does not natively support INT64. Attempting to cast down to INT32. input shape:input (1, 3, 384, 1152) output shape:cate_pred (3872, 80) enableDLA buildEngineWithConfig [05/25/2022-22:57:20] [I] [TRT] [MemUsageSnapshot] Builder begin: CPU 664 MiB, GPU 671 MiB [05/25/2022-22:57:21] [I] [TRT] [MemUsageChange] Init cuBLAS/cuBLASLt: CPU +70, GPU +68, now: CPU 822, GPU 1012 (MiB) [05/25/2022-22:57:21] [I] [TRT] [MemUsageChange] Init cuDNN: CPU +0, GPU +10, now: CPU 822, GPU 1022 (MiB) [05/25/2022-22:57:21] [W] [TRT] Detected invalid timing cache, setup a local cache instead [05/25/2022-22:57:24] [I] [TRT] Some tactics do not have sufficient workspace memory to run. Increasing workspace size may increase performance, please check verbose output. [05/25/2022-22:58:39] [I] [TRT] Detected 1 inputs and 13 output network tensors. [05/25/2022-22:58:39] [I] [TRT] Total Host Persistent Memory: 274640 [05/25/2022-22:58:39] [I] [TRT] Total Device Persistent Memory: 83921920 [05/25/2022-22:58:39] [I] [TRT] Total Scratch Memory: 0 [05/25/2022-22:58:39] [I] [TRT] [MemUsageStats] Peak memory usage of TRT CPU/GPU memory allocators: CPU 158 MiB, GPU 675 MiB [05/25/2022-22:58:39] [I] [TRT] [MemUsageChange] Init cuBLAS/cuBLASLt: CPU +0, GPU +8, now: CPU 1298, GPU 1635 (MiB) [05/25/2022-22:58:39] [I] [TRT] [MemUsageChange] Init cuDNN: CPU +0, GPU +8, now: CPU 1298, GPU 1643 (MiB) [05/25/2022-22:58:39] [I] [TRT] [MemUsageChange] Init cuBLAS/cuBLASLt: CPU +0, GPU +0, now: CPU 1298, GPU 1627 (MiB) [05/25/2022-22:58:39] [I] [TRT] [MemUsageChange] Init cuBLAS/cuBLASLt: CPU +0, GPU +0, now: CPU 1297, GPU 1611 (MiB) [05/25/2022-22:58:39] [I] [TRT] [MemUsageSnapshot] Builder end: CPU 1210 MiB, GPU 1381 MiB serializeModel done

Process finished with exit code 0

---

./build/demo ./config/config.yaml

~/Solov2-TensorRT-CPP/cmake-build-debug/segment ./config/config.yaml config_file:./config/config.yaml [05/25/2022-23:35:10] [I] [TRT] [MemUsageChange] Init CUDA: CPU +298, GPU +0, now: CPU 411, GPU 309 (MiB) [05/25/2022-23:35:11] [I] [TRT] Loaded engine size: 81 MB [05/25/2022-23:35:11] [I] [TRT] [MemUsageSnapshot] deserializeCudaEngine begin: CPU 411 MiB, GPU 309 MiB [05/25/2022-23:35:22] [I] [TRT] [MemUsageChange] Init cuBLAS/cuBLASLt: CPU +2140, GPU +980, now: CPU 2804, GPU 1731 (MiB) [05/25/2022-23:35:22] [I] [TRT] [MemUsageChange] Init cuDNN: CPU +0, GPU +10, now: CPU 2804, GPU 1741 (MiB) [05/25/2022-23:35:22] [I] [TRT] [MemUsageChange] Init cuBLAS/cuBLASLt: CPU +0, GPU +0, now: CPU 2804, GPU 1725 (MiB) [05/25/2022-23:35:22] [I] [TRT] [MemUsageSnapshot] deserializeCudaEngine end: CPU 2804 MiB, GPU 1725 MiB [05/25/2022-23:35:22] [I] [TRT] [MemUsageSnapshot] ExecutionContext creation begin: CPU 2804 MiB, GPU 1725 MiB [05/25/2022-23:35:22] [I] [TRT] [MemUsageChange] Init cuBLAS/cuBLASLt: CPU +0, GPU +8, now: CPU 2804, GPU 1733 (MiB) [05/25/2022-23:35:22] [I] [TRT] [MemUsageChange] Init cuDNN: CPU +0, GPU +8, now: CPU 2804, GPU 1741 (MiB) [05/25/2022-23:35:22] [I] [TRT] [MemUsageSnapshot] ExecutionContext creation end: CPU 2811 MiB, GPU 2166 MiB [05/25/2022-23:35:23] [E] [TRT] 3: [executionContext.cpp::enqueueInternal::322] Error Code 3: Internal Error (Parameter check failed at: runtime/api/executionContext.cpp::enqueueInternal::322, condition: bindings[x] != nullptr ) terminate called after throwing an instance of 'c10::Error' what(): CUDA error: invalid argument Exception raised from getDeviceFromPtr at ../aten/src/ATen/cuda/CUDADevice.h:13 (most recent call first): frame #0: c10::Error::Error(c10::SourceLocation, std::__cxx11::basic_string<char, std::char_traits, std::allocator >) + 0x69 (0x7f30d25c1b29 in ~/NVIDIA/libtorch/lib/libc10.so) frame #1: c10::detail::torchCheckFail(char const*, char const*, unsigned int, std::__cxx11::basic_string<char, std::char_traits, std::allocator > const&) + 0xd2 (0x7f30d25beab2 in /home/cqyd/NVIDIA/libtorch/lib/libc10.so) frame #2: + 0x36d1ea7 (0x7f306f824ea7 in ~/NVIDIA/libtorch/lib/libtorch_cuda.so) frame #3: + 0x7c87c (0x559bd949387c in ~/Solov2-TensorRT-CPP/cmake-build-debug/segment) frame #4: + 0x7cdf1 (0x559bd9493df1 in ~/Solov2-TensorRT-CPP/cmake-build-debug/segment) frame #5: + 0x7d2a7 (0x559bd94942a7 in ~/Solov2-TensorRT-CPP/cmake-build-debug/segment) frame #6: + 0x7dff8 (0x559bd9494ff8 in ~/Solov2-TensorRT-CPP/cmake-build-debug/segment) frame #7: + 0x7e0b2 (0x559bd94950b2 in ~/Solov2-TensorRT-CPP/cmake-build-debug/segment) frame #8: + 0x79af5 (0x559bd9490af5 in ~/SOLOV2model/Solov2-TensorRT-CPP/cmake-build-debug/segment) frame #9: + 0x84b4d (0x559bd949bb4d in ~/SOLOV2model/Solov2-TensorRT-CPP/cmake-build-debug/segment) frame #10: + 0x84827 (0x559bd949b827 in ~/Solov2-TensorRT-CPP/cmake-build-debug/segment) frame #11: + 0x16f77 (0x559bd942df77 in ~/Solov2-TensorRT-CPP/cmake-build-debug/segment) frame #12: __libc_start_main + 0xf3 (0x7f306b82c083 in /lib/x86_64-linux-gnu/libc.so.6) frame #13: + 0x1606e (0x559bd942d06e in ~/Solov2-TensorRT-CPP/cmake-build-debug/segment)

Process finished with exit code 134 (interrupted by signal 6: SIGABRT)

[chenjianqu]

According to "Error Code 3: Internal Error (Parameter check failed at: runtime/api/executionContext.cpp::enqueueInternal::322, condition: bindings[x] != nullptr", I guess there some problems with the ONNX model .

[chenjianqu]

Try a predefined model:baidudisk, fetch code:qdsm