• 版本、环境信息：
     1）PaddlePaddle版本：2.1
     2）CPU：8700K
     3）GPU：3060 CUDA 11.0 CUDNN 81077 tensorRt 7234
     4）系统环境：win10，Python 3.8.8 cmake 3.20.1 vs2019
  -预测信息
     1）C++预测：请您提供预测库安装包的版本信息，及其中的version.txt文件
  GIT COMMIT ID: 1e62c23
  WITH_MKL: ON
  WITH_MKLDNN: ON
  WITH_GPU: ON
  WITH_ROCM: OFF
  CUDA version: 11.0
  CUDNN version: v8.0
  CXX compiler version: 19.16.27045.0
  WITH_TENSORRT: ON
  TensorRT version: v7

 4）预测库来源：官网下载/

使用PPOCR开启编译C++ GPU环境后使用报错

出错如下:

自编译Paddle2.1预测库版本错误日志:
Found Paddle host system: win32, version:
Found Paddle host system's CPU: 12 cores
Selecting Windows SDK version 10.0.19041.0 to target Windows 10.0.17763.
The CXX compiler identification is MSVC 19.29.30136.0
The C compiler identification is MSVC 19.29.30136.0
Detecting CXX compiler ABI info
Detecting CXX compiler ABI info - done
Check for working CXX compiler: D:/VS2019/VC/Tools/MSVC/14.29.30133/bin/Hostx64/x64/cl.exe - skipped
Detecting CXX compile features
Detecting CXX compile features - done
Detecting C compiler ABI info
Detecting C compiler ABI info - done
Check for working C compiler: D:/VS2019/VC/Tools/MSVC/14.29.30133/bin/Hostx64/x64/cl.exe - skipped
Detecting C compile features
Detecting C compile features - done
Looking for pthread.h
Looking for pthread.h - not found
Found Threads: TRUE
The CUDA compiler identification is NVIDIA 11.0.194
Detecting CUDA compiler ABI info
Detecting CUDA compiler ABI info - done
Check for working CUDA compiler: C:/PPOCR/CUDA/V110/bin/nvcc.exe - skipped
Detecting CUDA compile features
Detecting CUDA compile features - done
CUDA compiler: C:/PPOCR/CUDA/V110/bin/nvcc.exe, version: NVIDIA 11.0.194
CXX compiler: D:/VS2019/VC/Tools/MSVC/14.29.30133/bin/Hostx64/x64/cl.exe, version: MSVC 19.29.30136.0
C compiler: D:/VS2019/VC/Tools/MSVC/14.29.30133/bin/Hostx64/x64/cl.exe, version: MSVC 19.29.30136.0
AR tools: D:/VS2019/VC/Tools/MSVC/14.29.30133/bin/Hostx64/x64/lib.exe
Use static C runtime time, refer to https://docs.microsoft.com/en-us/cpp/c-runtime-library/crt-library-features?view=vs-2019
Found Git: D:/MyWork/Git/cmd/git.exe (found version "2.20.1.windows.1")
Performing Test MMX_FOUND
Performing Test MMX_FOUND - Failed
Performing Test SSE2_FOUND
Performing Test SSE2_FOUND - Success
Performing Test SSE3_FOUND
Performing Test SSE3_FOUND - Success
Performing Test AVX_FOUND
Performing Test AVX_FOUND - Success
Performing Test AVX2_FOUND
Performing Test AVX2_FOUND - Success
Performing Test AVX512F_FOUND
Performing Test AVX512F_FOUND - Failed
CMake Warning at CMakeLists.txt:237 (MESSAGE):
Disable NCCL when compiling for Windows. Force WITH_NCCL=OFF.

CMake Warning at CMakeLists.txt:263 (MESSAGE):
If the environment is multi-card, the WITH_NCCL option needs to be turned
on, otherwise only a single card can be used.

CUDA detected: 11.0.194
WARNING: This is just a warning for publishing release.
You are building GPU version without supporting different architectures.
So the wheel package may fail on other GPU architectures.
You can add -DCUDA_ARCH_NAME=All in cmake command
to get a full wheel package to resolve this warning.
While, this version will still work on local GPU architecture.
Automatic GPU detection failed. Building for all known architectures.
NVCC_FLAGS_EXTRA: -gencode arch=compute_52,code=sm_52 -gencode arch=compute_60,code=sm_60 -gencode arch=compute_61,code=sm_61 -gencode arch=compute_70,code=sm_70 -gencode arch=compute_75,code=sm_75 -gencode arch=compute_80,code=sm_80
Current cuDNN header is C:/PPOCR/CUDA/V110/include/cudnn_version.h Current cuDNN version is v8.1.0.
CMake Warning at CMakeLists.txt:292 (MESSAGE):
Disable RCCL when compiling without ROCM. Force WITH_RCCL=OFF.

Disable XBYAK in Windows and MacOS
Looking for C++ include shlwapi.h
Looking for C++ include shlwapi.h - found
BOOST_VERSION: 1.41.0, BOOST_URL: http://paddlepaddledeps.bj.bcebos.com/boost_1_41_0.tar.gz
warp-ctc library: E:/Paddle/build/third_party/install/warpctc/bin/warpctc.dll
MKLML_VER: mklml_win_2019.0.5.20190502, MKLML_URL: https://paddlepaddledeps.bj.bcebos.com/mklml_win_2019.0.5.20190502.zip
Found cblas and lapack in MKLML (include: E:/Paddle/build/third_party/install/mklml/include, library: E:/Paddle/build/third_party/install/mklml/lib/mklml.lib)
Set E:/Paddle/build/third_party/install/mkldnn/lib to runtime path
MKLDNN library: E:/Paddle/build/third_party/install/mkldnn/bin/mkldnn.lib
Protobuf protoc executable: E:/Paddle/build/third_party/install/protobuf/bin/protoc.exe
Protobuf-lite library: E:/Paddle/build/third_party/install/protobuf/lib/libprotobuf-lite.lib
Protobuf library: E:/Paddle/build/third_party/install/protobuf/lib/libprotobuf.lib
Protoc library: E:/Paddle/build/third_party/install/protobuf/lib/libprotoc.lib
Protobuf version: 3.1.0
Found PythonInterp: C:/Anaconda3/python.exe (found suitable version "3.8.8", minimum required is "3.6")
Found PythonLibs: C:/Anaconda3/libs/python38.lib (found suitable version "3.8.8", minimum required is "3.6")
Found PY_pip: C:\Anaconda3\lib\site-packages\pip_init.py
Found PY_numpy: C:\Anaconda3\lib\site-packages\numpyinit.py
Found PY_wheel: C:\Anaconda3\lib\site-packages\wheelinit.py
Found PY_google.protobuf: C:\Anaconda3\lib\site-packages\google\protobufinit_.py
CMake Deprecation Warning at cmake/FindNumPy.cmake:6 (cmake_minimum_required):
Compatibility with CMake < 2.8.12 will be removed from a future version of
CMake.

Update the VERSION argument value or use a ... suffix to tell
CMake that the project does not need compatibility with older versions.
Call Stack (most recent call first):
cmake/external/python.cmake:72 (FIND_PACKAGE)
cmake/third_party.cmake:230 (include)
CMakeLists.txt:307 (include)

Found NumPy: C:/Anaconda3/Lib/site-packages/numpy/core/include
Download dependence[cudaerror] from http://paddlepaddledeps.bj.bcebos.com/cudaErrorMessage.tar.gz
Looking for UINT64_MAX
Looking for UINT64_MAX - found
Looking for sys/types.h
Looking for sys/types.h - found
Looking for stdint.h
Looking for stdint.h - found
Looking for stddef.h
Looking for stddef.h - found
Check size of pthread_spinlock_t
Check size of pthread_spinlock_t - failed
Check size of pthread_barrier_t
Check size of pthread_barrier_t - failed
Paddle version is 2.1.3
Found CUDA: C:/PPOCR/CUDA/V110 (found version "11.0")
Cannot find CUPTI, GPU Profiling is incorrect.
Enable Intel OpenMP with E:/Paddle/build/third_party/install/mklml/lib/libiomp5md.lib
CMake Warning at CMakeLists.txt:371 (message):
On inference mode, will take place some specific optimization. Turn on the
ON_INFER flag when building inference_lib only.

commit: 06d47ff
branch: HEAD
WITH_DLNNE:
MESSAGE: This is just a message for publishing release.
You are building AVX version without NOAVX core.
So the wheel package may fail on NOAVX machine.
You can add -DNOAVX_CORE_FILE=/path/to/your/core_noavx.* in cmake command
to get a full wheel package to resolve this warning.
While, this version will still work on local machine.
Configuring done