GH-119462: Enforce invariants of type versioning (GH-120731)

* Remove uses of Py_TPFLAGS_VALID_VERSION_TAG

1 files changed, 56 insertions, 92 deletions

diff --git a/Objects/typeobject.c b/Objects/typeobject.c
index 0dcf1d3..1cc6ca7 100644
--- a/Objects/typeobject.c
+++ b/Objects/typeobject.c
@@ -974,43 +974,37 @@ PyType_Unwatch(int watcher_id, PyObject* obj)
     return 0;
 }
 
-#ifdef Py_GIL_DISABLED
-
 static void
-type_modification_starting_unlocked(PyTypeObject *type)
+set_version_unlocked(PyTypeObject *tp, unsigned int version)
 {
     ASSERT_TYPE_LOCK_HELD();
-
-    /* Clear version tags on all types, but leave the valid
-       version tag intact.  This prepares for a modification so that
-       any concurrent readers of the type cache will not see invalid
-       values.
-     */
-    if (!_PyType_HasFeature(type, Py_TPFLAGS_VALID_VERSION_TAG)) {
-        return;
+#ifndef Py_GIL_DISABLED
+    PyInterpreterState *interp = _PyInterpreterState_GET();
+    // lookup the old version and set to null
+    if (tp->tp_version_tag != 0) {
+        PyTypeObject **slot =
+            interp->types.type_version_cache
+            + (tp->tp_version_tag % TYPE_VERSION_CACHE_SIZE);
+        *slot = NULL;
     }
-
-    PyObject *subclasses = lookup_tp_subclasses(type);
-    if (subclasses != NULL) {
-        assert(PyDict_CheckExact(subclasses));
-
-        Py_ssize_t i = 0;
-        PyObject *ref;
-        while (PyDict_Next(subclasses, &i, NULL, &ref)) {
-            PyTypeObject *subclass = type_from_ref(ref);
-            if (subclass == NULL) {
-                continue;
-            }
-            type_modification_starting_unlocked(subclass);
-            Py_DECREF(subclass);
-        }
+    if (version) {
+        tp->tp_versions_used++;
+    }
+#else
+    if (version) {
+        _Py_atomic_add_uint16(&tp->tp_versions_used, 1);
     }
-
-    /* 0 is not a valid version tag */
-    _PyType_SetVersion(type, 0);
-}
-
 #endif
+    FT_ATOMIC_STORE_UINT32_RELAXED(tp->tp_version_tag, version);
+#ifndef Py_GIL_DISABLED
+    if (version != 0) {
+        PyTypeObject **slot =
+            interp->types.type_version_cache
+            + (version % TYPE_VERSION_CACHE_SIZE);
+        *slot = tp;
+    }
+#endif
+}
 
 static void
 type_modified_unlocked(PyTypeObject *type)
@@ -1021,16 +1015,16 @@ type_modified_unlocked(PyTypeObject *type)
 
        Invariants:
 
-       - before Py_TPFLAGS_VALID_VERSION_TAG can be set on a type,
+       - before tp_version_tag can be set on a type,
          it must first be set on all super types.
 
-       This function clears the Py_TPFLAGS_VALID_VERSION_TAG of a
+       This function clears the tp_version_tag of a
        type (so it must first clear it on all subclasses).  The
-       tp_version_tag value is meaningless unless this flag is set.
+       tp_version_tag value is meaningless when equal to zero.
        We don't assign new version tags eagerly, but only as
        needed.
      */
-    if (!_PyType_HasFeature(type, Py_TPFLAGS_VALID_VERSION_TAG)) {
+    if (type->tp_version_tag == 0) {
         return;
     }
 
@@ -1070,8 +1064,7 @@ type_modified_unlocked(PyTypeObject *type)
         }
     }
 
-    type->tp_flags &= ~Py_TPFLAGS_VALID_VERSION_TAG;
-    _PyType_SetVersion(type, 0); /* 0 is not a valid version tag */
+    set_version_unlocked(type, 0); /* 0 is not a valid version tag */
     if (PyType_HasFeature(type, Py_TPFLAGS_HEAPTYPE)) {
         // This field *must* be invalidated if the type is modified (see the
         // comment on struct _specialization_cache):
@@ -1083,7 +1076,7 @@ void
 PyType_Modified(PyTypeObject *type)
 {
     // Quick check without the lock held
-    if (!_PyType_HasFeature(type, Py_TPFLAGS_VALID_VERSION_TAG)) {
+    if (type->tp_version_tag == 0) {
         return;
     }
 
@@ -1147,8 +1140,7 @@ type_mro_modified(PyTypeObject *type, PyObject *bases) {
 
  clear:
     assert(!(type->tp_flags & _Py_TPFLAGS_STATIC_BUILTIN));
-    type->tp_flags &= ~Py_TPFLAGS_VALID_VERSION_TAG;
-    _PyType_SetVersion(type, 0); /* 0 is not a valid version tag */
+    set_version_unlocked(type, 0); /* 0 is not a valid version tag */
     if (PyType_HasFeature(type, Py_TPFLAGS_HEAPTYPE)) {
         // This field *must* be invalidated if the type is modified (see the
         // comment on struct _specialization_cache):
@@ -1168,25 +1160,10 @@ This is similar to func_version_cache.
 void
 _PyType_SetVersion(PyTypeObject *tp, unsigned int version)
 {
-#ifndef Py_GIL_DISABLED
-    PyInterpreterState *interp = _PyInterpreterState_GET();
-    // lookup the old version and set to null
-    if (tp->tp_version_tag != 0) {
-        PyTypeObject **slot =
-            interp->types.type_version_cache
-            + (tp->tp_version_tag % TYPE_VERSION_CACHE_SIZE);
-        *slot = NULL;
-    }
-#endif
-    FT_ATOMIC_STORE_UINT32_RELAXED(tp->tp_version_tag, version);
-#ifndef Py_GIL_DISABLED
-    if (version != 0) {
-        PyTypeObject **slot =
-            interp->types.type_version_cache
-            + (version % TYPE_VERSION_CACHE_SIZE);
-        *slot = tp;
-    }
-#endif
+
+    BEGIN_TYPE_LOCK()
+    set_version_unlocked(tp, version);
+    END_TYPE_LOCK()
 }
 
 PyTypeObject *
@@ -1221,12 +1198,11 @@ assign_version_tag(PyInterpreterState *interp, PyTypeObject *type)
 {
     ASSERT_TYPE_LOCK_HELD();
 
-    /* Ensure that the tp_version_tag is valid and set
-       Py_TPFLAGS_VALID_VERSION_TAG.  To respect the invariant, this
-       must first be done on all super classes.  Return 0 if this
-       cannot be done, 1 if Py_TPFLAGS_VALID_VERSION_TAG.
+    /* Ensure that the tp_version_tag is valid.
+     * To respect the invariant, this must first be done on all super classes.
+     * Return 0 if this cannot be done, 1 if tp_version_tag is set.
     */
-    if (_PyType_HasFeature(type, Py_TPFLAGS_VALID_VERSION_TAG)) {
+    if (type->tp_version_tag != 0) {
         return 1;
     }
     if (!_PyType_HasFeature(type, Py_TPFLAGS_READY)) {
@@ -1235,14 +1211,22 @@ assign_version_tag(PyInterpreterState *interp, PyTypeObject *type)
     if (type->tp_versions_used >= MAX_VERSIONS_PER_CLASS) {
         return 0;
     }
-    type->tp_versions_used++;
+
+    PyObject *bases = lookup_tp_bases(type);
+    Py_ssize_t n = PyTuple_GET_SIZE(bases);
+    for (Py_ssize_t i = 0; i < n; i++) {
+        PyObject *b = PyTuple_GET_ITEM(bases, i);
+        if (!assign_version_tag(interp, _PyType_CAST(b))) {
+            return 0;
+        }
+    }
     if (type->tp_flags & Py_TPFLAGS_IMMUTABLETYPE) {
         /* static types */
         if (NEXT_GLOBAL_VERSION_TAG > _Py_MAX_GLOBAL_TYPE_VERSION_TAG) {
             /* We have run out of version numbers */
             return 0;
         }
-        _PyType_SetVersion(type, NEXT_GLOBAL_VERSION_TAG++);
+        set_version_unlocked(type, NEXT_GLOBAL_VERSION_TAG++);
         assert (type->tp_version_tag <= _Py_MAX_GLOBAL_TYPE_VERSION_TAG);
     }
     else {
@@ -1251,18 +1235,9 @@ assign_version_tag(PyInterpreterState *interp, PyTypeObject *type)
             /* We have run out of version numbers */
             return 0;
         }
-        _PyType_SetVersion(type, NEXT_VERSION_TAG(interp)++);
+        set_version_unlocked(type, NEXT_VERSION_TAG(interp)++);
         assert (type->tp_version_tag != 0);
     }
-
-    PyObject *bases = lookup_tp_bases(type);
-    Py_ssize_t n = PyTuple_GET_SIZE(bases);
-    for (Py_ssize_t i = 0; i < n; i++) {
-        PyObject *b = PyTuple_GET_ITEM(bases, i);
-        if (!assign_version_tag(interp, _PyType_CAST(b)))
-            return 0;
-    }
-    type->tp_flags |= Py_TPFLAGS_VALID_VERSION_TAG;
     return 1;
 }
 
@@ -3318,7 +3293,7 @@ mro_internal_unlocked(PyTypeObject *type, int initial, PyObject **p_old_mro)
     else {
         /* For static builtin types, this is only called during init
            before the method cache has been populated. */
-        assert(_PyType_HasFeature(type, Py_TPFLAGS_VALID_VERSION_TAG));
+        assert(type->tp_version_tag);
     }
 
     if (p_old_mro != NULL)
@@ -5463,7 +5438,7 @@ _PyType_LookupRef(PyTypeObject *type, PyObject *name)
 #else
     if (entry->version == type->tp_version_tag &&
         entry->name == name) {
-        assert(_PyType_HasFeature(type, Py_TPFLAGS_VALID_VERSION_TAG));
+        assert(type->tp_version_tag);
         OBJECT_STAT_INC_COND(type_cache_hits, !is_dunder_name(name));
         OBJECT_STAT_INC_COND(type_cache_dunder_hits, is_dunder_name(name));
         Py_XINCREF(entry->value);
@@ -5486,7 +5461,6 @@ _PyType_LookupRef(PyTypeObject *type, PyObject *name)
     if (MCACHE_CACHEABLE_NAME(name)) {
         has_version = assign_version_tag(interp, type);
         version = type->tp_version_tag;
-        assert(!has_version || _PyType_HasFeature(type, Py_TPFLAGS_VALID_VERSION_TAG));
     }
     END_TYPE_LOCK()
 
@@ -5770,16 +5744,6 @@ type_setattro(PyObject *self, PyObject *name, PyObject *value)
         return -1;
     }
 
-#ifdef Py_GIL_DISABLED
-    // In free-threaded builds readers can race with the lock-free portion
-    // of the type cache and the assignment into the dict.  We clear all of the
-    // versions initially so no readers will succeed in the lock-free case.
-    // They'll then block on the type lock until the update below is done.
-    type_modification_starting_unlocked(type);
-#endif
-
-    res = _PyDict_SetItem_LockHeld((PyDictObject *)dict, name, value);
-
     /* Clear the VALID_VERSION flag of 'type' and all its
         subclasses.  This could possibly be unified with the
         update_subclasses() recursion in update_slot(), but carefully:
@@ -5787,6 +5751,8 @@ type_setattro(PyObject *self, PyObject *name, PyObject *value)
         recursing into subclasses. */
     type_modified_unlocked(type);
 
+    res = _PyDict_SetItem_LockHeld((PyDictObject *)dict, name, value);
+
     if (res == 0) {
         if (is_dunder_name(name)) {
             res = update_slot(type, name);
@@ -5898,7 +5864,6 @@ fini_static_type(PyInterpreterState *interp, PyTypeObject *type,
 
     if (final) {
         type->tp_flags &= ~Py_TPFLAGS_READY;
-        type->tp_flags &= ~Py_TPFLAGS_VALID_VERSION_TAG;
         _PyType_SetVersion(type, 0);
     }
 
@@ -8516,12 +8481,11 @@ init_static_type(PyInterpreterState *interp, PyTypeObject *self,
 
         assert(NEXT_GLOBAL_VERSION_TAG <= _Py_MAX_GLOBAL_TYPE_VERSION_TAG);
         _PyType_SetVersion(self, NEXT_GLOBAL_VERSION_TAG++);
-        self->tp_flags |= Py_TPFLAGS_VALID_VERSION_TAG;
     }
     else {
         assert(!initial);
         assert(self->tp_flags & _Py_TPFLAGS_STATIC_BUILTIN);
-        assert(self->tp_flags & Py_TPFLAGS_VALID_VERSION_TAG);
+        assert(self->tp_version_tag != 0);
     }
 
     managed_static_type_state_init(interp, self, isbuiltin, initial);