> reference counting is not intrusive in CPython (meaning the reference counting structures are outside the PyObject, last I checked) That's wrong. That was ne

> reference counting is not intrusive in CPython (meaning the reference counting structures are outside the PyObject, last I checked) That's wrong. That was never the case. Recent CPython: https://github.com/python/cpython/blob/6d419db10c84cacbb3862e2adc940342175bfce9/Include/object.h#L219 https://github.com/python/cpython/blob/6d419db10c84cacbb3862... CPython 2.0: https://github.com/python/cpython/blob/2a9b0a93091b9ef7350a94bb3d3f1c43725b7a8c/Include/object.h#L66 https://github.com/python/cpython/blob/2a9b0a93091b9ef7350a9... CPython 0.9.8: https://github.com/python/cpython/blob/dd104400dc551dd4098f35e12072e12c45822adc/Include/object.h#L91 https://github.com/python/cpython/blob/dd104400dc551dd4098f3... Regarding multiprocessing.Pool, that would not work as I said. I was thinking more about a plain fork, like this: def parent(): juicy_variable = ... def workerfunc(x): # I can access juicy_variable ... childs = [] for i in [1, 2, 3]: child = fork() if child == 0: workerfunc(i) sys.exit() childs.append(child) # Wait for and cleanup childs. # Communicate somehow with childs to get back results. ...