FazBrowse GitHub Viewer | Trending |
URL:
| Home
Tools: [Download Repo ZIP]   [Original HTTPS Page]

gh-119109: functool.partial vectorcall supports pto->kw & fallback to tp_call removed by dg-pb · Pull Request #120783 · python/cpython · GitHub

/ cpython Public

gh-119109: functool.partial vectorcall supports pto->kw & fallback to tp_call removed #120783

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.

Sign up for GitHub

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

Closed
dg-pb wants to merge 2 commits into from
Conversation 7 Commits 2 Checks 0 Files changed
Closed

gh-119109: functool.partial vectorcall supports pto->kw & fallback to tp_call removed #120783

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
40d38a4
initial implementation
dg-pb Jun 20, 2024
80535f3
remove unnecessary code
dg-pb Jun 20, 2024
File filter

Filter by extension

Filter by extension .c  (1) All 1 file type selected
Viewed files
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Unified
Split
Hide whitespace
Diff view
Unified
Split
Hide whitespace
132 changes: 87 additions & 45 deletions Modules/_functoolsmodule.c
Show comments
View file
Open in desktop
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters. Learn more about bidirectional Unicode characters
Show hidden characters
Original file line number Diff line number Diff line change
Expand Up @@ -198,69 +198,104 @@ partial_dealloc(partialobject *pto)
}


/* Merging keyword arguments using the vectorcall convention is messy, so
* if we would need to do that, we stop using vectorcall and fall back
* to using partial_call() instead. */
Py_NO_INLINE static PyObject *
partial_vectorcall_fallback(PyThreadState *tstate, partialobject *pto,
PyObject *const *args, size_t nargsf,
PyObject *kwnames)
{
pto->vectorcall = NULL;
Py_ssize_t nargs = PyVectorcall_NARGS(nargsf);
return _PyObject_MakeTpCall(tstate, (PyObject *)pto,
args, nargs, kwnames);
}

static PyObject *
partial_vectorcall(partialobject *pto, PyObject *const *args,
size_t nargsf, PyObject *kwnames)
{
PyThreadState *tstate = _PyThreadState_GET();

/* pto->kw is mutable, so need to check every time */
if (PyDict_GET_SIZE(pto->kw)) {
return partial_vectorcall_fallback(tstate, pto, args, nargsf, kwnames);
}

/* Sizes */
Py_ssize_t pto_nargs = PyTuple_GET_SIZE(pto->args);
Py_ssize_t pto_nkwds = PyDict_GET_SIZE(pto->kw);
Py_ssize_t nargs = PyVectorcall_NARGS(nargsf);
Py_ssize_t nargs_total = nargs;
Py_ssize_t nkwds = 0;
if (kwnames != NULL) {
nargs_total += PyTuple_GET_SIZE(kwnames);
nkwds = PyTuple_GET_SIZE(kwnames);
}

/* Containers */
PyObject **pto_args = _PyTuple_ITEMS(pto->args);
Py_ssize_t pto_nargs = PyTuple_GET_SIZE(pto->args);

/* Fast path if we're called without arguments */
if (nargs_total == 0) {
return _PyObject_VectorcallTstate(tstate, pto->fn,
pto_args, pto_nargs, NULL);
}

/* Fast path using PY_VECTORCALL_ARGUMENTS_OFFSET to prepend a single
* positional argument */
if (pto_nargs == 1 && (nargsf & PY_VECTORCALL_ARGUMENTS_OFFSET)) {
PyObject **newargs = (PyObject **)args - 1;
PyObject *tmp = newargs[0];
newargs[0] = pto_args[0];
PyObject *ret = _PyObject_VectorcallTstate(tstate, pto->fn,
newargs, nargs + 1, kwnames);
newargs[0] = tmp;
if (pto_nkwds == 0) {
Py_ssize_t nargs_total = nargs + nkwds;

/* Fast path if we're called without arguments */
if (nargs_total == 0) {
return _PyObject_VectorcallTstate(tstate, pto->fn,
pto_args, pto_nargs, NULL);
}

/* Fast path using PY_VECTORCALL_ARGUMENTS_OFFSET to prepend a single
* positional argument */
if (pto_nargs == 1 && (nargsf & PY_VECTORCALL_ARGUMENTS_OFFSET)) {
PyObject **newargs = (PyObject **)args - 1;
PyObject *tmp = newargs[0];
newargs[0] = pto_args[0];
PyObject *ret = _PyObject_VectorcallTstate(tstate, pto->fn,
newargs, nargs + 1,
kwnames);
newargs[0] = tmp;
return ret;
}
Py_ssize_t new_nargs_total = pto_nargs + pto_nkwds + nargs_total;
PyObject *small_stack[_PY_FASTCALL_SMALL_STACK];
PyObject *ret;
PyObject **stack;

if (new_nargs_total <= (Py_ssize_t)Py_ARRAY_LENGTH(small_stack)) {
stack = small_stack;
}
else {
stack = PyMem_Malloc(new_nargs_total * sizeof(PyObject *));
if (stack == NULL) {
PyErr_NoMemory();
return NULL;
}
}

/* Copy to new stack, using borrowed references */
memcpy(stack, pto_args, pto_nargs * sizeof(PyObject*));
memcpy(stack + pto_nargs, args, nargs_total * sizeof(PyObject*));

ret = _PyObject_VectorcallTstate(tstate, pto->fn,
stack, pto_nargs + nargs, kwnames);
if (stack != small_stack) {
PyMem_Free(stack);
}
return ret;
}

Py_ssize_t newnargs_total = pto_nargs + nargs_total;
/* Merge keywords */
PyObject *new_kw, *key, *val;
new_kw = PyDict_Copy(pto->kw);
if (new_kw == NULL) {
return NULL;
}
if (nkwds != 0) {
for (Py_ssize_t i=0; i < nkwds; i++) {
key = PyTuple_GET_ITEM(kwnames, i);
if (PyDict_SetItem(new_kw, key, args[nargs + i]) != 0) {
Py_DECREF(new_kw);
return NULL;
}
}
}
Py_ssize_t new_nkwds = PyDict_GET_SIZE(new_kw);
PyObject *new_kwnames = PyTuple_New(new_nkwds);
if (new_kwnames == NULL) {
return NULL;
}

Py_ssize_t new_nargs = pto_nargs + nargs;

/* Arg Stack */
PyObject *small_stack[_PY_FASTCALL_SMALL_STACK];
PyObject *ret;
PyObject **stack;

if (newnargs_total <= (Py_ssize_t)Py_ARRAY_LENGTH(small_stack)) {
if (new_nargs + new_nkwds <= (Py_ssize_t)Py_ARRAY_LENGTH(small_stack)) {
stack = small_stack;
}
else {
stack = PyMem_Malloc(newnargs_total * sizeof(PyObject *));
stack = PyMem_Malloc((new_nargs + new_nkwds) * sizeof(PyObject *));
if (stack == NULL) {
PyErr_NoMemory();
return NULL;
Expand All @@ -269,13 +304,20 @@ partial_vectorcall(partialobject *pto, PyObject *const *args,

/* Copy to new stack, using borrowed references */
memcpy(stack, pto_args, pto_nargs * sizeof(PyObject*));
memcpy(stack + pto_nargs, args, nargs_total * sizeof(PyObject*));
memcpy(stack + pto_nargs, args, nargs * sizeof(PyObject*));
Py_ssize_t pos = 0, i = 0;
while (PyDict_Next(new_kw, &pos, &key, &val)) {
PyTuple_SET_ITEM(new_kwnames, i, Py_NewRef(key));
stack[new_nargs + i] = val;
i++;
}

ret = _PyObject_VectorcallTstate(tstate, pto->fn,
stack, pto_nargs + nargs, kwnames);
ret = _PyObject_VectorcallTstate(tstate, pto->fn, stack, new_nargs, new_kwnames);
if (stack != small_stack) {
PyMem_Free(stack);
}
Py_XDECREF(new_kw);
Py_XDECREF(new_kwnames);
return ret;
}

Expand Down
Back | FazBrowse Home | New Git URL