libtorch crashing core dump

[natxopedreira]

🐛 Describe the bug

Just trying to integrate libtorch in another c++ platform, is building but crashing on launch

Im compiling with gcc -L./libtorch/lib/ -ltorch_cpu -lc10

terminate called after throwing an instance of 'c10::Error'
  what():  _ivalue_INTERNAL ASSERT FAILED at "../torch/csrc/jit/api/object.h":35, please report a bug to PyTorch. 
Exception raised from _ivalue at ../torch/csrc/jit/api/object.h:35 (most recent call first):
frame #0: c10::Error::Error(c10::SourceLocation, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >) + 0x6b (0x7ffa965bb0eb in /home/natxo/of_v0.11.2_linux64gcc6_release/apps/myApps/pytorch_example/libtorch/lib/libc10.so)
frame #1: c10::detail::torchCheckFail(char const*, char const*, unsigned int, char const*) + 0xd1 (0x7ffa965b6c41 in /home/natxo/of_v0.11.2_linux64gcc6_release/apps/myApps/pytorch_example/libtorch/lib/libc10.so)
frame #2: torch::jit::Object::find_method(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&) const + 0x33f (0x7ffa999aeb5f in /home/natxo/of_v0.11.2_linux64gcc6_release/apps/myApps/pytorch_example/libtorch/lib/libtorch_cpu.so)
frame #3: <unknown function> + 0x93753 (0x55775d422753 in ./pytorch_example)
frame #4: <unknown function> + 0xad337 (0x55775d43c337 in ./pytorch_example)
frame #5: <unknown function> + 0x1d5c14 (0x55775d564c14 in ./pytorch_example)
frame #6: <unknown function> + 0xa1c74 (0x55775d430c74 in ./pytorch_example)
frame #7: <unknown function> + 0xa945c (0x55775d43845c in ./pytorch_example)
frame #8: <unknown function> + 0xa965c (0x55775d43865c in ./pytorch_example)
frame #9: <unknown function> + 0x9e76d (0x55775d42d76d in ./pytorch_example)
frame #10: main + 0xd5 (0x55775d4039b5 in ./pytorch_example)
frame #11: __libc_start_main + 0xf3 (0x7ffa947670b3 in /lib/x86_64-linux-gnu/libc.so.6)
frame #12: <unknown function> + 0x9039e (0x55775d41f39e in ./pytorch_example)

Aborted (core dumped)

Versions

PyTorch version: 1.9.0+cu102
Is debug build: False
CUDA used to build PyTorch: 10.2
ROCM used to build PyTorch: N/A

OS: Ubuntu 20.04.4 LTS (x86_64)
GCC version: (Ubuntu 9.4.0-1ubuntu1~20.04.1) 9.4.0
Clang version: Could not collect
CMake version: version 3.21.4
Libc version: glibc-2.31

Python version: 3.8.10 (default, Mar 15 2022, 12:22:08) [GCC 9.4.0] (64-bit runtime)
Python platform: Linux-5.13.0-40-generic-x86_64-with-glibc2.29
Is CUDA available: True
CUDA runtime version: 11.2.67
GPU models and configuration: GPU 0: NVIDIA GeForce GTX 1070
Nvidia driver version: 470.103.01
cuDNN version: Probably one of the following:
/usr/lib/x86_64-linux-gnu/libcudnn.so.7.6.5
/usr/local/cuda-11.2/targets/x86_64-linux/lib/libcudnn.so.8.1.0
/usr/local/cuda-11.2/targets/x86_64-linux/lib/libcudnn_adv_infer.so.8.1.0
/usr/local/cuda-11.2/targets/x86_64-linux/lib/libcudnn_adv_train.so.8.1.0
/usr/local/cuda-11.2/targets/x86_64-linux/lib/libcudnn_cnn_infer.so.8.1.0
/usr/local/cuda-11.2/targets/x86_64-linux/lib/libcudnn_cnn_train.so.8.1.0
/usr/local/cuda-11.2/targets/x86_64-linux/lib/libcudnn_ops_infer.so.8.1.0
/usr/local/cuda-11.2/targets/x86_64-linux/lib/libcudnn_ops_train.so.8.1.0
HIP runtime version: N/A
MIOpen runtime version: N/A
Is XNNPACK available: True

Versions of relevant libraries:
[pip3] mypy-extensions==0.4.3
[pip3] numpy==1.22.3
[pip3] numpydoc==0.7.0
[pip3] torch==1.9.0
[pip3] torchsummary==1.5.1
[pip3] torchvision==0.10.0
[conda] blas 1.0 mkl
[conda] mkl 2021.4.0 h06a4308_640
[conda] mkl-service 2.4.0 py38h7f8727e_0
[conda] mkl_fft 1.3.1 py38hd3c417c_0
[conda] mkl_random 1.2.2 py38h51133e4_0
[conda] mypy_extensions 0.4.3 py38h06a4308_1
[conda] numpy 1.19.5 pypi_0 pypi
[conda] numpydoc 1.2 pyhd3eb1b0_0

cc @jbschlosser

[natxopedreira]

I compiled from source, installed and works now.

[D-jojo]

@natxopedreira Hi,
May I ask how you solved this problem? I have recently encountered a similar problem.
terminate called after throwing an instance of 'c10::Error'
what(): _ivalue_INTERNAL ASSERT FAILED at "../torch/csrc/jit/api/object.h":35, please report a bug to PyTorch.
Exception raised from ivalue at ../torch/csrc/jit/api/object.h:35 (most recent call first):
frame #0: c10::Error::Error(c10::SourceLocation, std::cxx11::basic_string<char, std::char_traits, std::allocator >) + 0xa0 (0xffff9ea71e30 in /home/xyb/sd/djj/env_djj/lib/python3.8/site-packages/torch/lib/libc10.so)
frame #1: c10::detail::torchCheckFail(char const*, char const*, unsigned int, char const*) + 0xe8 (0xffff9ea53ad4 in /home/xyb/sd/djj/env_djj/lib/python3.8/site-packages/torch/lib/libc10.so)
frame #2: torch::jit::Object::find_method(std::cxx11::basic_string<char, std::char_traits, std::allocator > const&) const + 0x554 (0xffffb080e6c4 in /home/xyb/sd/djj/env_djj/lib/python3.8/site-packages/torch/lib/libtorch_cpu.so)
frame #3: Model::evaluate(std::vector<double, std::allocator > const&, bool) + 0x278 (0xffffb262d9a8 in /home/xyb/sd/djj/cms_ws/catkin_cms/devel/lib/libcpp_policy.so)
frame #4: agile_locomotion::AgileLocomotion::startWalking() + 0xb50 (0xaaaacbd32190 in /home/xyb/sd/djj/cms_ws/catkin_cms/devel/lib/agile_locomotion/agile_locomotion)
frame #5: agile_locomotion::AgileLocomotion::startExecutionCallback(boost::shared_ptr<std_msgs::Empty<std::allocator > const> const&) + 0x130 (0xaaaacbd33318 in /home/xyb/sd/djj/cms_ws/catkin_cms/devel/lib/agile_locomotion/agile_locomotion)
frame #6: boost::detail::function::void_function_obj_invoker1<boost::function<void (boost::shared_ptr<std_msgs::Empty<std::allocator > const> const&)>, void, boost::shared_ptr<std_msgs::Empty<std::allocator > const> >::invoke(boost::detail::function::function_buffer&, boost::shared_ptr<std_msgs::Empty<std::allocator > const>) + 0x3c (0xaaaacbd4347c in /home/xyb/sd/djj/cms_ws/catkin_cms/devel/lib/agile_locomotion/agile_locomotion)
frame #7: ros::SubscriptionCallbackHelperT<boost::shared_ptr<std_msgs::Empty<std::allocator > const> const&, void>::call(ros::SubscriptionCallbackHelperCallParams&) + 0x1e8 (0xaaaacbd43f48 in /home/xyb/sd/djj/cms_ws/catkin_cms/devel/lib/agile_locomotion/agile_locomotion)
frame #8: ros::SubscriptionQueue::call() + 0x9a4 (0xffffb28c8e6c in /opt/ros/noetic/lib/libroscpp.so)
frame #9: ros::CallbackQueue::callOneCB(ros::CallbackQueue::TLS*) + 0x274 (0xffffb2874a6c in /opt/ros/noetic/lib/libroscpp.so)
frame #10: ros::CallbackQueue::callOne(ros::WallDuration) + 0x1d8 (0xffffb2875780 in /opt/ros/noetic/lib/libroscpp.so)
frame #11: ros::AsyncSpinnerImpl::threadFunc() + 0x64 (0xffffb28cb8bc in /opt/ros/noetic/lib/libroscpp.so)
frame #12: + 0x10624 (0xffff9e3c2624 in /lib/aarch64-linux-gnu/libboost_thread.so.1.71.0)
frame #13: + 0x7624 (0xffffb2714624 in /lib/aarch64-linux-gnu/libpthread.so.0)
frame #14: + 0xd149c (0xffff9e78249c in /lib/aarch64-linux-gnu/libc.so.6)

I am looking forward to your reply very much, thank you!