python - Anaconda Acclerate/NumbaPro CUDA 链接错误 OSX

Question

总体目标是使用NumbaPro在 GPU 上运行某些功能(在 OSX 10.8.3 上)。

在开始之前，我只是想把一切都设置好。根据this page我安装了 CUDA，注册为 CUDA 开发人员，下载了编译器 SDK 并设置了 NUMBAPRO_NVVM=/path/to/libnvvm.dylib 环境变量。

但是，运行这个基本测试函数:

from numbapro import autojit

@autojit(target='gpu')
def my_function(x):
    if x == 0.0:
        return 1.0
    else:
        return x*x*x

print my_function(4.4)
exit()

出现此错误:

File ".../anaconda/lib/python2.7/site-packages/numba/decorators.py", line 207, in compile_function
    compiled_function = dec(f)
File "...lib/python2.7/site-packages/numbapro/cudapipeline/decorators.py", line 35, in _jit_decorator
File "...lib/python2.7/site-packages/numbapro/cudapipeline/decorators.py", line 128, in __init__
File "...lib/python2.7/site-packages/numbapro/cudapipeline/environment.py", line 31, in generate_ptx
File "...lib/python2.7/site-packages/numbapro/cudapipeline/environment.py", line 186, in _link_llvm_math_intrinsics
KeyError: 1

我尝试过 @vectorize'ing 而不是 autojit，同样的错误。@autojit 本身没有目标就可以正常工作。

有什么想法吗？

Answer 1

为了子孙后代的利益，我询问了 Continuum 支持。他们回应:

It seems that you are running a CUDA GPU with compute capability 1.x. NVVM only supports CC2.0 and above. We definitely should have a better error reporting and make it clear in the NumbaPro documentation for the supported compute capability.