fix (and test) bisecting into code object cache

Cython/Compiler/Nodes.py

@@ -8711,12 +8711,12 @@ static void __Pyx_AddTraceback(const char *funcname, int %(CLINENO)s,
     PyObject *py_globals = 0;
     PyFrameObject *py_frame = 0;
 
-    py_code = %(FINDCODEOBJECT)s(%(CLINENO)s);
+    py_code = %(FINDCODEOBJECT)s(%(CLINENO)s ? %(CLINENO)s : %(LINENO)s);
     if (!py_code) {
         py_code = __Pyx_CreateCodeObjectForTraceback(
             funcname, %(CLINENO)s, %(LINENO)s, %(FILENAME)s);
         if (!py_code) goto bad;
-        %(INSERTCODEOBJECT)s(%(CLINENO)s, py_code);
+        %(INSERTCODEOBJECT)s(%(CLINENO)s ? %(CLINENO)s : %(LINENO)s, py_code);
     }
     py_globals = PyModule_GetDict(%(GLOBALS)s);
     if (!py_globals) goto bad;

Cython/Utility/ModuleSetupCode.c

@@ -234,6 +234,7 @@ struct __Pyx_CodeObjectCache {
 
 static struct __Pyx_CodeObjectCache __pyx_code_cache = {0,0,NULL};
 
+static int __pyx_bisect_code_objects(__Pyx_CodeObjectCacheEntry* entries, int count, int code_line);
 static void __pyx_clear_code_object_cache(void);
 static PyCodeObject *__pyx_find_code_object(int code_line);
 static void __pyx_insert_code_object(int code_line, PyCodeObject* code_object);
@@ -258,25 +259,25 @@ static void __pyx_clear_code_object_cache(void) {
     PyMem_Free(entries);
 }
 
-static int __pyx_bisect_code_objects(int code_line) {
-    int start, end, mid;
-    __Pyx_CodeObjectCacheEntry* entries = __pyx_code_cache.entries;
-    start = 0;
-    end = __pyx_code_cache.count - 1;
+static int __pyx_bisect_code_objects(__Pyx_CodeObjectCacheEntry* entries, int count, int code_line) {
+    int start = 0, mid = 0, end = count - 1;
+    if (end >= 0 && code_line > entries[end].code_line) {
+        return count;
+    }
     while (start < end) {
         mid = (start + end) / 2;
         if (code_line < entries[mid].code_line) {
-            end = mid - 1;
+            end = mid;
         } else if (code_line > entries[mid].code_line) {
              start = mid + 1;
         } else {
             return mid;
         }
     }
-    if (start == end || code_line <= entries[start].code_line) {
-        return start;
+    if (code_line <= entries[mid].code_line) {
+        return mid;
     } else {
-        return end;
+        return mid + 1;
     }
 }
 
@@ -286,7 +287,7 @@ static PyCodeObject *__pyx_find_code_object(int code_line) {
     if (unlikely(!code_line) || unlikely(!__pyx_code_cache.entries)) {
         return NULL;
     }
-    pos = __pyx_bisect_code_objects(code_line);
+    pos = __pyx_bisect_code_objects(__pyx_code_cache.entries, __pyx_code_cache.count, code_line);
     if (unlikely(pos >= __pyx_code_cache.count) || unlikely(__pyx_code_cache.entries[pos].code_line != code_line)) {
         return NULL;
     }
@@ -313,7 +314,7 @@ static void __pyx_insert_code_object(int code_line, PyCodeObject* code_object) {
         }
         return;
     }
-    pos = __pyx_bisect_code_objects(code_line);
+    pos = __pyx_bisect_code_objects(__pyx_code_cache.entries, __pyx_code_cache.count, code_line);
     if ((pos < __pyx_code_cache.count) && unlikely(__pyx_code_cache.entries[pos].code_line == code_line)) {
         PyCodeObject* tmp = entries[pos].code_object;
         entries[pos].code_object = code_object;

tests/run/code_object_cache.pyx

+# mode: run
+# tag: except
+
+# test the code object cache that is being used in exception raising
+
+### low level tests
+
+cdef extern from *:
+    # evil hack to access the internal utility function
+    ctypedef struct PyCodeObject
+    ctypedef struct __Pyx_CodeObjectCacheEntry:
+        int code_line
+        PyCodeObject* code_object
+    int __pyx_bisect_code_objects(__Pyx_CodeObjectCacheEntry* entries, int count, int code_line)
+
+def test_lowlevel_bisect2(*indices):
+    """
+    >>> test_lowlevel_bisect2(1, 2, 3, 4, 5, 6)
+    [0, 0, 1, 1, 2, 2]
+    """
+    cdef __Pyx_CodeObjectCacheEntry* cache = [
+        __Pyx_CodeObjectCacheEntry(2, NULL),
+        __Pyx_CodeObjectCacheEntry(4, NULL),
+        ]
+    return [ __pyx_bisect_code_objects(cache, 2, i)
+             for i in indices ]
+
+def test_lowlevel_bisect5(*indices):
+    """
+    >>> test_lowlevel_bisect5(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11)
+    [0, 1, 2, 2, 2, 3, 3, 3, 4, 5, 5]
+    """
+    cdef __Pyx_CodeObjectCacheEntry* cache = [
+        __Pyx_CodeObjectCacheEntry(1, NULL),
+        __Pyx_CodeObjectCacheEntry(2, NULL),
+        __Pyx_CodeObjectCacheEntry(5, NULL),
+        __Pyx_CodeObjectCacheEntry(8, NULL),
+        __Pyx_CodeObjectCacheEntry(9, NULL),
+        ]
+    return [ __pyx_bisect_code_objects(cache, 5, i)
+             for i in indices ]
+
+def test_lowlevel_bisect6(*indices):
+    """
+    >>> test_lowlevel_bisect6(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13)
+    [0, 0, 1, 2, 2, 2, 3, 3, 4, 5, 5, 5, 6]
+    """
+    cdef __Pyx_CodeObjectCacheEntry* cache = [
+        __Pyx_CodeObjectCacheEntry(2, NULL),
+        __Pyx_CodeObjectCacheEntry(3, NULL),
+        __Pyx_CodeObjectCacheEntry(6, NULL),
+        __Pyx_CodeObjectCacheEntry(8, NULL),
+        __Pyx_CodeObjectCacheEntry(9, NULL),
+        __Pyx_CodeObjectCacheEntry(12, NULL),
+        ]
+    return [ __pyx_bisect_code_objects(cache, 6, i)
+             for i in indices ]
+
+### Python level tests
+
+import sys
+
+def tb():
+    return sys.exc_info()[-1]
+
+def raise_keyerror():
+    raise KeyError
+
+def check_code_object_identity_recursively(tb1, tb2):
+    if tb1 is None or tb2 is None:
+        return
+    code1, code2 = tb1.tb_frame.f_code, tb2.tb_frame.f_code
+    if code1 is not code2:
+        print('%s != %s' % (code1, code2))
+    check_code_object_identity_recursively(tb1.tb_next, tb2.tb_next)
+
+def assert_simple_code_object_reuse():
+    """
+    >>> try: assert_simple_code_object_reuse()
+    ... except KeyError: t1 = tb()
+    >>> try: assert_simple_code_object_reuse()
+    ... except KeyError: t2 = tb()
+    >>> check_code_object_identity_recursively(t1.tb_next, t2.tb_next)
+    """
+    raise KeyError
+
+def assert_multi_step_code_object_reuse(recursions=0):
+    """
+    >>> for depth in range(5):
+    ...     try: assert_multi_step_code_object_reuse(depth)
+    ...     except KeyError: t1 = tb()
+    ...     try: assert_multi_step_code_object_reuse(depth)
+    ...     except KeyError: t2 = tb()
+    ...     check_code_object_identity_recursively(t1.tb_next, t2.tb_next)
+    """
+    if recursions:
+        assert_multi_step_code_object_reuse(recursions-1)
+    else:
+        raise_keyerror()