使用pyopencl找不到OpenCL包含文件

问题描述

我正在尝试使用在python中运行并使用pyopencl的程序。

我将Windows与Anaconda和GeForce GTX 1080配合使用。我按照所有步骤安装和运行pyopencl，这是很多工作。

但是，当我运行完整的代码时，似乎缺少某些库，但是我不知道从何处获取缺少的库或如何链接它。这是运行时错误：

Traceback (most recent call last):

  File "D:\Deep_Physics\Protein\Mi3-GPU-master\mi3gpu\Mi3.py",line 1236,in <module>
    main(sys.argv[1:])

  File "D:\Deep_Physics\Protein\Mi3-GPU-master\mi3gpu\Mi3.py",line 1232,in main
    actions[kNown_args.action](args,remaining_args,print)

  File "D:\Deep_Physics\Protein\Mi3-GPU-master\mi3gpu\Mi3.py",line 411,in inverseIsing
    gpus = setup_GPUs(p,log)

  File "D:\Deep_Physics\Protein\Mi3-GPU-master\mi3gpu\Mi3.py",line 312,in setup_GPUs
    clinfo,gpudevs,cllog = setup_GPU_context(scriptPath,scriptfile,p,log)

  File "D:\Deep_Physics\Protein\Mi3-GPU-master\mi3gpu\mcmcGPU.py",line 903,in setup_GPU_context
    cl_prg = cl.Program(cl_ctx,src).build(optstr + extraopt)

  File "C:\ProgramData\Anaconda3\envs\mi3\lib\site-packages\pyopencl\__init__.py",line 513,in build
    self._build_and_catch_errors(

  File "C:\ProgramData\Anaconda3\envs\mi3\lib\site-packages\pyopencl\__init__.py",line 569,in _build_and_catch_errors
    raise err

RuntimeError: clBuildProgram Failed: BUILD_PROGRAM_FAILURE - clBuildProgram Failed: BUILD_PROGRAM_FAILURE - clBuildProgram Failed: BUILD_PROGRAM_FAILURE

Build on <pyopencl.Device 'GeForce GTX 1080' on 'NVIDIA CUDA' at 0x1bcb431db20>:

<kernel>:18:10: Fatal error: 'mwc64x/cl/mwc64x/mwc64xvec2_rng.cl' file not found
#include <mwc64x/cl/mwc64x/mwc64xvec2_rng.cl>
         ^

(options: -D q=4 -D L=93 -D NHIST=256 -D HISTWS=512 -D WGSIZE=512 -cl-nv-verbose -Werror -I D:Deep_PhysicsProteinMi3-GPU-mastermi3gpu -I C:\ProgramData\Anaconda3\envs\mi3\lib\site-packages\pyopencl\cl)

作为参考，这是导入pyopencl后运行命令printGPUS(print)时的输出：

Platform 0 'NVIDIA CUDA':
    vendor: NVIDIA Corporation
    Version: OpenCL 1.2 CUDA 10.1.152
    Extensions: cl_khr_global_int32_base_atomics
                cl_khr_global_int32_extended_atomics
                cl_khr_local_int32_base_atomics
                cl_khr_local_int32_extended_atomics cl_khr_fp64
                cl_khr_byte_addressable_store cl_khr_icd cl_khr_gl_sharing
                cl_nv_compiler_options cl_nv_device_attribute_query
                cl_nv_pragma_unroll cl_nv_d3d10_sharing cl_khr_d3d10_sharing
                cl_nv_d3d11_sharing cl_nv_copy_opts cl_nv_create_buffer
  Device 'GeForce GTX 1080':
    vendor: NVIDIA Corporation
    Version: OpenCL 1.2 CUDA
    Driver Version: 425.25
    Max Clock Frequency: 1733
    Max Compute Units: 20
    Max Work Group Size: 1024
    Global Mem Size: 8589934592
    Global Mem Cache Size: 327680
    Local Mem Size: 49152
    Max Constant Buffer Size: 65536

解决方法

暂无找到可以解决该程序问题的有效方法，小编努力寻找整理中！

如果你已经找到好的解决方法，欢迎将解决方案带上本链接一起发送给小编。

小编邮箱:dio#foxmail.com (将#修改为@）

opencl pyopencl python