the program quit without any error

[nuclear-missile]

I compile on my own computer with env RTX5060+cudnn9.13+cuda13.0
it compile perfect but cannot run

the env list:


path: C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v13.0\bin\x64;C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v13.0\bin;C:\Windows\system32;C:\Windows;C:\Windows\System32\Wbem;C:\Windows\System32\WindowsPowerShell\v1.0;C:\Windows\System32\OpenSSH;C:\ProgramData\miniconda3;C:\ProgramData\miniconda3\Scripts;C:\ProgramData\miniconda3\Library\bin;C:\Program Files\dotnet;C:\Program Files\NVIDIA Corporation\Nsight Compute 2025.2.0;C:\Program Files (x86)\NVIDIA Corporation\PhysX\Common;C:\Program Files\Git\cmd;C:\Program Files (x86)\Windows Kits\10\Windows Performance Toolkit;C:\Program Files\CMake\bin;C:\Users\Administrator\AppData\Local\Microsoft\WindowsApps;C:\Users\Administrator\AppData\Local\Programs\Microsoft VS Code\bin;C:\Program Files\NVIDIA\CUDNN\v9.13\bin\13.0;

I change the CMakeLists.txt:
......
if(USE_BACKEND STREQUAL "CUDA")
message(STATUS "-DUSE_BACKEND=CUDA, using CUDA backend.")

enable_language(CUDA)

set(CUDA_STANDARD 11)
set(NEURALNET_BACKEND_SOURCES
neuralnet/cudabackend.cpp
neuralnet/cudautils.cpp
neuralnet/cudahelpers.cu
)

https://en.wikipedia.org/wiki/CUDA#GPUs_supported

set(CMAKE_CUDA_ARCHITECTURES 90)

if(CMAKE_CUDA_COMPILER_VERSION VERSION_GREATER_EQUAL 12.8)

set(CMAKE_CUDA_ARCHITECTURES 50 52 53 60 61 62 70 72 75 80 86 87 90 120)

set(CMAKE_CUDA_FLAGS "-Wno-deprecated-gpu-targets")

elseif(CMAKE_CUDA_COMPILER_VERSION VERSION_GREATER_EQUAL 12.0)

set(CMAKE_CUDA_ARCHITECTURES 50 52 53 60 61 62 70 72 75 80 86 87 90)

set(CMAKE_CUDA_FLAGS "-Wno-deprecated-gpu-targets")

elseif(CMAKE_CUDA_COMPILER_VERSION VERSION_GREATER_EQUAL 11.8)

set(CMAKE_CUDA_ARCHITECTURES 35 37 50 52 53 60 61 62 70 72 75 80 86 87 90)

set(CMAKE_CUDA_FLAGS "-Wno-deprecated-gpu-targets")

elseif(CMAKE_CUDA_COMPILER_VERSION VERSION_GREATER_EQUAL 11.5)

set(CMAKE_CUDA_ARCHITECTURES 35 37 50 52 53 60 61 62 70 72 75 80 86 87)

set(CMAKE_CUDA_FLAGS "-Wno-deprecated-gpu-targets")

elseif(CMAKE_CUDA_COMPILER_VERSION VERSION_GREATER_EQUAL 11.1)

set(CMAKE_CUDA_ARCHITECTURES 35 37 50 52 53 60 61 62 70 72 75 80 86)

set(CMAKE_CUDA_FLAGS "-Wno-deprecated-gpu-targets")

elseif(CMAKE_CUDA_COMPILER_VERSION VERSION_GREATER_EQUAL 11.0)

set(CMAKE_CUDA_ARCHITECTURES 35 37 50 52 53 60 61 62 70 72 75 80)

set(CMAKE_CUDA_FLAGS "-Wno-deprecated-gpu-targets")

elseif (CMAKE_CUDA_COMPILER_VERSION VERSION_GREATER_EQUAL 10.2)

set(CMAKE_CUDA_ARCHITECTURES 30 35 37 50 52 53 60 61 62 70 72 75)

set(CMAKE_CUDA_FLAGS "-Wno-deprecated-gpu-targets")

else()

message(WARNING "CUDA 10.2 or greater is recommended, but attempting to build anyways")

set(CMAKE_CUDA_ARCHITECTURES 30 37 53 70)

endif()

cmake output:
Selecting Windows SDK version 10.0.26100.0 to target Windows 10.0.22631.
The C compiler identification is MSVC 19.44.35217.0
The CXX compiler identification is MSVC 19.44.35217.0
Detecting C compiler ABI info
Detecting C compiler ABI info - done
Check for working C compiler: C:/Program Files/Microsoft Visual Studio/2022/Community/VC/Tools/MSVC/14.44.35207/bin/Hostx64/x64/cl.exe - skipped
Detecting C compile features
Detecting C compile features - done
Detecting CXX compiler ABI info
Detecting CXX compiler ABI info - done
Check for working CXX compiler: C:/Program Files/Microsoft Visual Studio/2022/Community/VC/Tools/MSVC/14.44.35207/bin/Hostx64/x64/cl.exe - skipped
Detecting CXX compile features
Detecting CXX compile features - done
Building 'katago' executable for GTP engine and other tools.
-DUSE_BACKEND=CUDA, using CUDA backend.
The CUDA compiler identification is NVIDIA 13.0.88 with host compiler MSVC 19.44.35217.0
Detecting CUDA compiler ABI info
Detecting CUDA compiler ABI info - done
Check for working CUDA compiler: C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v13.0/bin/nvcc.exe - skipped
Detecting CUDA compile features
Detecting CUDA compile features - done
Including Git revision in the compiled executable, specify -DNO_GIT_REVISION=1 to disable
CMake Warning at CMakeLists.txt:478 (message):
WARNING: libzip library was NOT found. KataGo should still work for
GTP/matches/analysis if everything else is good, but selfplay for writing
training data will not be possible.

Setting up build for Windows.
Setting up build for MSVC.
Configuring done (16.0s)
Generating done (0.0s)

run output:
"compile\katago.exe" gtp -model "weights\b18c384nbt-optimisticv13-s5971M.bin.gz" -config "katago-v1.16.3-cuda12.8-cudnn9.8.0-windows-x64+bs50\analysis_example.cfg" 0<test_commands.txt
KataGo v1.16.3
Using Chinese rules initially, unless GTP/GUI overrides this

D:\2023-06-15-windows64+katago>pause
Press any key to continue . . .

[lightvector]

If you're on windows, you might be failing to find the relevant DLLs (despite compiling against those libs, windows might not know at runtime where to find the DLLs). Try copying the DLLs from the microsoft redistributable, as well as zlib, and the DLLs for CUDA and CUDNN into the exe directory and see if that fixes it?