Tag Archives: paddle

[Solved] paddle:FatalError: `Segmentation fault` is detected by the operating system.

Question

When using the paddle framework for relevant model training, an error is reported when using the GPU:

C++ Traceback (most recent call last):
--------------------------------------
0 std::thread::_Impl<std::_Bind_simple<ThreadPool::ThreadPool(unsigned long)::{lambda()#1} ()> >::_M_run()
1 std::__future_base::_State_baseV2::_M_do_set(std::function<std::unique_ptr<std::__future_base::_Result_base, std::__future_base::_Result_base::_Deleter> ()>*, bool*)
2 paddle::framework::SignalHandle(char const*, int)
3 paddle::platform::GetCurrentTraceBackString[abi:cxx11]()

----------------------
Error Message Summary:
----------------------
FatalError: `Segmentation fault` is detected by the operating system.
[TimeInfo: *** Aborted at 1639543591 (unix time) try "date -d @1639543591" if you are using GNU date ***]
[SignalInfo: *** SIGSEGV (@0x0) received by PID 168238 (TID 0x7fc05d8c5700) from PID 0 ***]

edition:

python=3.8 paddlepaddle-gpu=2.2.1 cuda=10.1 cudnn=7.6

Check the issue and find that many of these situations occur. Make a record here

Solution:

Reinstall using CONDA.

Execute command:

conda install paddlepaddle-gpu==2.2.1 cudatoolkit=10.1 --channel https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/Paddle/

How to Solve QGIS installation paddy GPU error

QGIS installation paddy GPU error

Error content

Use the following code to check whether the paddle GPU is installed successfully

import paddle
paddle.utils.run_check()

Get the following error reports

RuntimeError: (PreconditionNotMet) The third-party dynamic library (cublas64_102.dll;cublas64_10.dll) that Paddle depends on is not configured correctly. (error code is 126)
  Suggestions:
  1. Check if the third-party dynamic library (e.g. CUDA, CUDNN) is installed correctly and its version is matched with paddlepaddle you installed.
  2. Configure third-party dynamic library environment variables as follows:
  - Linux: set LD_LIBRARY_PATH by `export LD_LIBRARY_PATH=...`
  - Windows: set PATH by `set PATH=XXX; (at C:\home\workspace\Paddle_release\paddle\fluid\platform\dynload\dynamic_loader.cc:265)

It is found that other CONDA environments can be used normally, CUDA installation and environment variable configuration are correct, and the environment variable is checked in QGIS. It is found that there is no C: \program files\NVIDIA GPU computing toolkit\CUDA\v10.2\bin in path.

Solution:

For the desktop, you can add C:\program files\NVIDIA GPU computing toolkit\CUDA\v10.2\bin
after the path in QGIS\bin\qgis-ltr-bin.env. For plug-ins developed in vscode, you can add o4w in the same folder, Change set path to set path =% osgeo4w in env.bat_ROOT%\bin;% WINDIR%\system32;% WINDIR%;% WINDIR%\system32\WBem;% CUDA_BIN_PATH%