TensorFlow 0.10 (CUDA) on OSX segfaults on python import

Question

我一直在尝试在我的 El Capitan Macbook Pro（2013 年末，GeForce GT 750M）上启动并运行 tensorflow 0.10，但到目前为止没有成功。我尝试了official tensorflow documentation's instructions 和其他一些人的方法，包括this one 和this one。

作为参考，我正在尝试在 OSX 10.11.5 上使用 Python3、CUDA 7.5 和 tensorflow 0.10。

我已经安装了 CUDA，它可以识别我的 GPU。我可以在/Developer/NVIDIA/CUDA-7.5/samples/1_Utilities/deviceQuery 中成功编译deviceQuery 示例。它运行时的输出是：

./deviceQuery Starting...

 CUDA Device Query (Runtime API) version (CUDART static linking)

Detected 1 CUDA Capable device(s)

Device 0: "GeForce GT 750M"
  CUDA Driver Version / Runtime Version          7.5 / 7.5
  CUDA Capability Major/Minor version number:    3.0
  Total amount of global memory:                 2048 MBytes (2147024896 bytes)
  ( 2) Multiprocessors, (192) CUDA Cores/MP:     384 CUDA Cores
  GPU Max Clock rate:                            926 MHz (0.93 GHz)
  Memory Clock rate:                             2508 Mhz
  Memory Bus Width:                              128-bit
  L2 Cache Size:                                 262144 bytes
  Maximum Texture Dimension Size (x,y,z)         1D=(65536), 2D=(65536, 65536), 3D=(4096, 4096, 4096)
  Maximum Layered 1D Texture Size, (num) layers  1D=(16384), 2048 layers
  Maximum Layered 2D Texture Size, (num) layers  2D=(16384, 16384), 2048 layers
  Total amount of constant memory:               65536 bytes
  Total amount of shared memory per block:       49152 bytes
  Total number of registers available per block: 65536
  Warp size:                                     32
  Maximum number of threads per multiprocessor:  2048
  Maximum number of threads per block:           1024
  Max dimension size of a thread block (x,y,z): (1024, 1024, 64)
  Max dimension size of a grid size    (x,y,z): (2147483647, 65535, 65535)
  Maximum memory pitch:                          2147483647 bytes
  Texture alignment:                             512 bytes
  Concurrent copy and kernel execution:          Yes with 1 copy engine(s)
  Run time limit on kernels:                     Yes
  Integrated GPU sharing Host Memory:            No
  Support host page-locked memory mapping:       Yes
  Alignment requirement for Surfaces:            Yes
  Device has ECC support:                        Disabled
  Device supports Unified Addressing (UVA):      Yes
  Device PCI Domain ID / Bus ID / location ID:   0 / 1 / 0
  Compute Mode:
     < Default (multiple host threads can use ::cudaSetDevice() with device simultaneously) >

deviceQuery, CUDA Driver = CUDART, CUDA Driver Version = 7.5, CUDA Runtime Version = 7.5, NumDevs = 1, Device0 = GeForce GT 750M
Result = PASS

我还下载了 cudnn-7.5 库和头文件，并将这些文件放在 /usr/local/cuda/lib 和 include 的正确位置。

在 python3 交互式 REPL 中，如果我输入 import tensorflow，我会得到以下输出：

Python 3.5.2 (v3.5.2:4def2a2901a5, Jun 26 2016, 10:47:25) 
[GCC 4.2.1 (Apple Inc. build 5666) (dot 3)] on darwin
Type "help", "copyright", "credits" or "license" for more information.
>>> import tensorflow
I tensorflow/stream_executor/dso_loader.cc:108] successfully opened CUDA library libcublas.7.5.dylib locally
I tensorflow/stream_executor/dso_loader.cc:108] successfully opened CUDA library libcudnn.5.dylib locally
I tensorflow/stream_executor/dso_loader.cc:108] successfully opened CUDA library libcufft.7.5.dylib locally
Segmentation fault: 11

我的问题是，我需要做什么才能成功导入没有段错误的模块？如果有帮助，我在 python3 REPL here 中发布了运行该命令的dtruss 输出的要点，以及带有堆栈跟踪here 的诊断（崩溃）报告的要点。

Answer 1

此评论中描述了问题：https://github.com/tensorflow/tensorflow/issues/2940#issuecomment-238952433

“加载 libcuda.dylib 时存在错误 - 默认 cuda 安装会创建 libcuda.dylib，但 tensorflow 会尝试加载 libcuda.1.dylib 。这失败了，使用 LD_LIBRARY_PATH 如果 NULL 崩溃。如果你复制libcuda.dylib 到 libcuda.1.dylib 加载正常。”

使用拉取请求为其他所有人修复崩溃非常容易——即使用-c dbg 编译以准确查看哪一行正在尝试使用空值，并将类似的内容添加到代码中

if (mystring == NULL) { return; }