diff options
| author | Eric Snow <ericsnowcurrently@gmail.com> | 2023-03-14 20:01:35 (GMT) |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2023-03-14 20:01:35 (GMT) |
| commit | cdb21ba74d933e262bc1696b6ce78b50cb5a4443 (patch) | |
| tree | bfdf1ff0a564d9b7bc3f7288f115ccd3bc9c3d12 /Python/pystate.c | |
| parent | 80abd62647b2a36947a11a6a8e395061be6f0c61 (diff) | |
| download | cpython-cdb21ba74d933e262bc1696b6ce78b50cb5a4443.zip cpython-cdb21ba74d933e262bc1696b6ce78b50cb5a4443.tar.gz cpython-cdb21ba74d933e262bc1696b6ce78b50cb5a4443.tar.bz2 | |
gh-102660: Handle m_copy Specially for the sys and builtins Modules (gh-102661)
It doesn't make sense to use multi-phase init for these modules. Using a per-interpreter "m_copy" (instead of PyModuleDef.m_base.m_copy) makes this work okay. (This came up while working on gh-101660.)
Note that we might instead end up disallowing re-load for sys/builtins since they are so special.
https://github.com/python/cpython/issues/102660
Diffstat (limited to 'Python/pystate.c')
| -rw-r--r-- | Python/pystate.c | 1 |
1 files changed, 1 insertions, 0 deletions
diff --git a/Python/pystate.c b/Python/pystate.c index 28606e4..3a2966c 100644 --- a/Python/pystate.c +++ b/Python/pystate.c @@ -805,6 +805,7 @@ interpreter_clear(PyInterpreterState *interp, PyThreadState *tstate) assert(interp->imports.importlib == NULL); assert(interp->imports.import_func == NULL); + Py_CLEAR(interp->sysdict_copy); Py_CLEAR(interp->builtins_copy); Py_CLEAR(interp->dict); #ifdef HAVE_FORK |