bdk <[email protected]> added the comment: I applied a reduced version in 3b06320cf630 that only resets mainthreadident to at least allow signal() after fork from a thread (should fix the example bad.py). Leaving this open for discussion about the other ThreadLocal behaviors after fork (couldn't replicate these parts of the test on CPython).
________________________________________ PyPy bug tracker <[email protected]> <https://bugs.pypy.org/issue1289> ________________________________________ _______________________________________________ pypy-issue mailing list [email protected] http://mail.python.org/mailman/listinfo/pypy-issue
