Free tstate on python 3.7+ on finalize_interpreter

Author: bstaletic

include/pybind11/detail/internals.h

@@ -25,6 +25,7 @@ inline PyObject *make_object_base_type(PyTypeObject *metaclass);
 #    define PYBIND11_TLS_GET_VALUE(key) PyThread_tss_get((key))
 #    define PYBIND11_TLS_REPLACE_VALUE(key, value) PyThread_tss_set((key), (value))
 #    define PYBIND11_TLS_DELETE_VALUE(key) PyThread_tss_set((key), nullptr)
+#    define PYBIND11_TLS_FREE(key) PyThread_tss_free(key)
 #else
     // Usually an int but a long on Cygwin64 with Python 3.x
 #    define PYBIND11_TLS_KEY_INIT(var) decltype(PyThread_create_key()) var = 0
@@ -43,6 +44,7 @@ inline PyObject *make_object_base_type(PyTypeObject *metaclass);
 #        define PYBIND11_TLS_REPLACE_VALUE(key, value)                       \
              PyThread_set_key_value((key), (value))
 #    endif
+#    define PYBIND11_TLS_FREE(key) (void)key
 #endif
 
 // Python loads modules by default with dlopen with the RTLD_LOCAL flag; under libc++ and possibly

include/pybind11/embed.h

@@ -158,6 +158,11 @@ inline void finalize_interpreter() {
     if (builtins.contains(id) && isinstance<capsule>(builtins[id]))
         internals_ptr_ptr = capsule(builtins[id]);
 
+#if defined(WITH_THREAD)
+    if (internals_ptr_ptr && *internals_ptr_ptr)
+        PYBIND11_TLS_FREE((*internals_ptr_ptr)->tstate);
+#endif
+
     Py_Finalize();
 
     if (internals_ptr_ptr) {