6 Janus and threads

Where SWI-Prolog support native preemptively scheduled threads that exploit multiple cores, Python has a single interpreter that can switch between native threads.^{6Actually, you can create multiple Python interpreters. It is not yet clear to us whether that can help improving on concurrency.} Initially the Python interpreter is associated with the thread that created it which, for janus, is the first thread calling Python. Janus uses PyGILState_Ensure() and PyGILState_Release() around calls to e.g. py_call/2. In addition, the thread that created Python releases its interpreter after every call from Prolog on Python. As a result:

• Multiple Prolog threads can make calls to Python. The access to Python is serialized. If a Prolog thread does not want other threads to use Python it can use py_with_gil/1. When multiple Prolog threads make many calls to Python performance tends to drop significantly.

• Multiple Python threads can make calls to Prolog. As Python uses only one core at the same time, Prolog working on behalve of Python synchronously only uses one core. Prolog threads not related to Python can proceed on other cores.

• Python threading is cooperative. At the moment, a thread Python thread running Prolog will not allow other Python threads to make progress. Possibly that can be improved in the future.

• It appears to be possible to initialize Python in a thread. Python remains accessible from other threads after the initializing thread has terminated.