Multiprocessing hangs on Mac OS X without MKL

This bug occurs on Mac OS X 10.9 (but does not seem to occur on Linux and Windows) with both Python 2.7 and Python 3.3 at least for numpy 1.8.1 not compiled with MKL.

The following piece of code reproduces the bug:

import multiprocessing as mp

import numpy as np


def compute(n):
    print('Enter')
    np.dot(np.eye(n), np.eye(n))
    print('Exit')


print('\nWithout multiprocessing:')
compute(1000)

print('\nWith multiprocessing:')
workers = mp.Pool(1)
results = workers.map(compute, (1000, 1000))

The above code hangs in the workers.map call when executing np.dot(np.eye(n), np.eye(n)).

If the sequential call to compute is ommitted, the code runs as it should. Also, if the argument n is reduced sufficiently (< 100 on the computer which the bug was detected on), the code runs as it should. Again, the bug is not present when numpy is compiled with MKL.

[beltergd]

Another option is to add the following: import os os.environ[“VECLIB_MAXIMUM_THREADS”] = “1”

[matthew-brett]

I just tried the binary above - it does not hang.