Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

set_backend(AF_BACKEND_CUDA): Failed to load dynamic library on Windows #214

Open
4lrdyD opened this issue Aug 1, 2018 · 1 comment
Open

Comments

@4lrdyD
Copy link
Contributor

4lrdyD commented Aug 1, 2018

I got the lib folder in the system path, however:

julia> set_backend(AF_BACKEND_CUDA)
ERROR: ArrayFire Error (501) : Failed to load dynamic library.
Error in af_err __cdecl unified::AFSymbolManager::setBackend(af_backend)
In file src\api\unified\symbol_manager.cpp:232
Failed to load dynamic library. See http://www.arrayfire.com/docs/unifiedbackend.htm for instructions to set up environment for Unified backend.

Stacktrace:
 [1] _error(::UInt32) at C:\Users\user\.julia\v0.6\ArrayFIre\src\util.jl:81
 [2] set_backend(::UInt32) at C:\Users\user\.julia\v0.6\ArrayFIre\src\wrap.jl:865
julia> versioninfo()
Julia Version 0.6.3
Commit d55cadc350* (2018-05-28 20:20 UTC)
Platform Info:
  OS: Windows (x86_64-w64-mingw32)
  CPU: Intel(R) Core(TM) i7-3630QM CPU @ 2.40GHz
  WORD_SIZE: 64
  BLAS: libopenblas (USE64BITINT DYNAMIC_ARCH NO_AFFINITY Sandybridge)
  LAPACK: libopenblas64_
  LIBM: libopenlibm
  LLVM: libLLVM-3.9.1 (ORCJIT, ivybridge)
julia> afinfo()
ArrayFire v3.6.1 (OpenCL, 64-bit Windows, build b443e14)
[0] NVIDIA: GeForce GT 630M, 2048 MB
-1- INTEL: Intel(R) HD Graphics 4000, 1400 MB
-2- INTEL: Intel(R) Core(TM) i7-3630QM CPU @ 2.40GHz, 6037 MB
@4lrdyD
Copy link
Contributor Author

4lrdyD commented Dec 14, 2018

I tried to use ArrayFire with Visual C++, and when I use the helloworld sample with af.lib or afopencl.lib I get expected results with the info of the device:

ArrayFire v3.6.1 (OpenCL, 64-bit Windows, build b443e14)
[0] NVIDIA: GeForce GT 630M, 2048 MB
-1- INTEL: Intel(R) HD Graphics 4000, 1400 MB
-2- INTEL: Intel(R) Core(TM) i7-3630QM CPU @ 2.40GHz, 6037 MB

but when I use afcuda.lib i get the following:

ArrayFire Exception (Internal error:998):
In function __cdecl cuda::DeviceManager::DeviceManager(void)
In file src\backend\cuda\platform.cpp:509
CUDA Error (35): CUDA driver version is insufficient for CUDA runtime version


In function void __cdecl af::setDevice(const int)
In file src\api\cpp\device.cpp:92

I tried to find newer driver version, but I have the latest for my device (v391.35), I guess the issue is for this reason and it won't be fixed unless I work with another device since mine must be somewhat deprecated, or maybe wait for another version of the driver, if it comes out.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant