commit 226a012d1cd61f42ecd3056c554922f359a1a35d
author Brandt Bucher <brandtbucher@gmail.com> Fri Dec 04 19:45:57 2020 -0800
committer GitHub <noreply@github.com> Fri Dec 04 19:45:57 2020 -0800
tree 86407049a5d2c22b0ce8626407ca106eb3c26afd
parent 2de5097ba4c50eba90df55696a7b2e74c93834d4

bpo-42536: GC track recycled tuples (GH-23623)

Several built-in and standard library types now ensure that their internal result tuples are always tracked by the garbage collector:

- collections.OrderedDict.items
- dict.items
- enumerate
- functools.reduce
- itertools.combinations
- itertools.combinations_with_replacement
- itertools.permutations
- itertools.product
- itertools.zip_longest
- zip

Previously, they could have become untracked by a prior garbage collection.

Modules/itertoolsmodule.c

diff --git a/Modules/itertoolsmodule.c b/Modules/itertoolsmodule.c
index 7144856..293735a 100644
--- a/Modules/itertoolsmodule.c
+++ b/Modules/itertoolsmodule.c
@@ -3,6 +3,7 @@
 #define PY_SSIZE_T_CLEAN
 #include "Python.h"
 #include "pycore_long.h"          // _PyLong_GetZero()
+#include "pycore_object.h"        // _PyObject_GC_TRACK()
 #include "pycore_tuple.h"         // _PyTuple_ITEMS()
 #include <stddef.h>               // offsetof()
 
@@ -2378,6 +2379,11 @@ product_next(productobject *lz)
             lz->result = result;
             Py_DECREF(old_result);
         }
+        // bpo-42536: The GC may have untracked this result tuple. Since we're
+        // recycling it, make sure it's tracked again:
+        else if (!_PyObject_GC_IS_TRACKED(result)) {
+            _PyObject_GC_TRACK(result);
+        }
         /* Now, we've got the only copy so we can update it in-place */
         assert (npools==0 || Py_REFCNT(result) == 1);
 
@@ -2701,6 +2707,11 @@ combinations_next(combinationsobject *co)
             co->result = result;
             Py_DECREF(old_result);
         }
+        // bpo-42536: The GC may have untracked this result tuple. Since we're
+        // recycling it, make sure it's tracked again:
+        else if (!_PyObject_GC_IS_TRACKED(result)) {
+            _PyObject_GC_TRACK(result);
+        }
         /* Now, we've got the only copy so we can update it in-place
          * CPython's empty tuple is a singleton and cached in
          * PyTuple's freelist.
@@ -3035,6 +3046,11 @@ cwr_next(cwrobject *co)
             co->result = result;
             Py_DECREF(old_result);
         }
+        // bpo-42536: The GC may have untracked this result tuple. Since we're
+        // recycling it, make sure it's tracked again:
+        else if (!_PyObject_GC_IS_TRACKED(result)) {
+            _PyObject_GC_TRACK(result);
+        }
         /* Now, we've got the only copy so we can update it in-place CPython's
            empty tuple is a singleton and cached in PyTuple's freelist. */
         assert(r == 0 || Py_REFCNT(result) == 1);
@@ -3379,6 +3395,11 @@ permutations_next(permutationsobject *po)
             po->result = result;
             Py_DECREF(old_result);
         }
+        // bpo-42536: The GC may have untracked this result tuple. Since we're
+        // recycling it, make sure it's tracked again:
+        else if (!_PyObject_GC_IS_TRACKED(result)) {
+            _PyObject_GC_TRACK(result);
+        }
         /* Now, we've got the only copy so we can update it in-place */
         assert(r == 0 || Py_REFCNT(result) == 1);
 
@@ -4649,6 +4670,11 @@ zip_longest_next(ziplongestobject *lz)
             PyTuple_SET_ITEM(result, i, item);
             Py_DECREF(olditem);
         }
+        // bpo-42536: The GC may have untracked this result tuple. Since we're
+        // recycling it, make sure it's tracked again:
+        if (!_PyObject_GC_IS_TRACKED(result)) {
+            _PyObject_GC_TRACK(result);
+        }
     } else {
         result = PyTuple_New(tuplesize);
         if (result == NULL)