-
-
Notifications
You must be signed in to change notification settings - Fork 31k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Micro-optimize list index range checks #9784
Changes from all commits
File filter
Filter by extension
Conversations
Jump to
Diff view
Diff view
There are no files selected for viewing
Original file line number | Diff line number | Diff line change |
---|---|---|
|
@@ -208,6 +208,19 @@ PyList_Size(PyObject *op) | |
return Py_SIZE(op); | ||
} | ||
|
||
static inline int | ||
valid_index(Py_ssize_t i, Py_ssize_t limit) | ||
{ | ||
/* The cast to size_t lets us use just a single comparison | ||
to check whether i is in the range: 0 <= i < limit. | ||
|
||
See: Section 14.2 "Bounds Checking" in the Agner Fog | ||
optimization manual found at: | ||
https://www.agner.org/optimize/optimizing_cpp.pdf | ||
*/ | ||
return (size_t) i < (size_t) limit; | ||
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. I'm not sure that the behaviour is well defined in C. I fear that it's Undefined Behaviour. @benjaminp @gpshead: What do you think ? There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. If it's well defined, why should we hack such micro optimization? Why compilers would not implement the optimization themself? There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. I think because they don't know that There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. It is well defined. It is used for example in the STL implementations. But there was not found any difference in microbenchmark results on 64-bit platforms in previous discussion in bpo-28397. |
||
} | ||
|
||
static PyObject *indexerr = NULL; | ||
|
||
PyObject * | ||
|
@@ -217,7 +230,7 @@ PyList_GetItem(PyObject *op, Py_ssize_t i) | |
PyErr_BadInternalCall(); | ||
return NULL; | ||
} | ||
if (i < 0 || i >= Py_SIZE(op)) { | ||
if (!valid_index(i, Py_SIZE(op))) { | ||
if (indexerr == NULL) { | ||
indexerr = PyUnicode_FromString( | ||
"list index out of range"); | ||
|
@@ -240,7 +253,7 @@ PyList_SetItem(PyObject *op, Py_ssize_t i, | |
PyErr_BadInternalCall(); | ||
return -1; | ||
} | ||
if (i < 0 || i >= Py_SIZE(op)) { | ||
if (!valid_index(i, Py_SIZE(op))) { | ||
Py_XDECREF(newitem); | ||
PyErr_SetString(PyExc_IndexError, | ||
"list assignment index out of range"); | ||
|
@@ -426,7 +439,7 @@ list_contains(PyListObject *a, PyObject *el) | |
static PyObject * | ||
list_item(PyListObject *a, Py_ssize_t i) | ||
{ | ||
if (i < 0 || i >= Py_SIZE(a)) { | ||
if (!valid_index(i, Py_SIZE(a))) { | ||
if (indexerr == NULL) { | ||
indexerr = PyUnicode_FromString( | ||
"list index out of range"); | ||
|
@@ -749,7 +762,7 @@ list_inplace_repeat(PyListObject *self, Py_ssize_t n) | |
static int | ||
list_ass_item(PyListObject *a, Py_ssize_t i, PyObject *v) | ||
{ | ||
if (i < 0 || i >= Py_SIZE(a)) { | ||
if (!valid_index(i, Py_SIZE(a))) { | ||
PyErr_SetString(PyExc_IndexError, | ||
"list assignment index out of range"); | ||
return -1; | ||
|
@@ -996,7 +1009,7 @@ list_pop_impl(PyListObject *self, Py_ssize_t index) | |
} | ||
if (index < 0) | ||
index += Py_SIZE(self); | ||
if (index < 0 || index >= Py_SIZE(self)) { | ||
if (!valid_index(index, Py_SIZE(self))) { | ||
PyErr_SetString(PyExc_IndexError, "pop index out of range"); | ||
return NULL; | ||
} | ||
|
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Why not just define this as taking two
size_t
parameters instead of doing the casting below. The casts then happen implicitly at all call sites.