Backporting METH_FASTCALL from Python 3.7 (#317)

* dict copy speedup

refer to bpo-31179 or python/cpython@boa7a037b8fde

* __build_class__() uses METH_FASTCALL

refer python/cpython@69de71b255
refer python/cpython@773dc6dd06

a single test related to __prepare__ fails.

* type_prepare uses METH_FASTCALL

refer python/cpython@d526cfe546
refer python/cpython@80ab22fa2c

the prepare-related test still fails.  It's just related to the error
message format though.

* separate into ParseStack and ParseStackAndKeywords

refer python/cpython@6518a93cb1
refer python/cpython@3e1fad6913
refer python/cpython@c0083fc47d

* Add _PyArg_NoStackKeywords

refer python/cpython@29d39cc8f5

* _PyStack_UnpackDict now returns int

refer python/cpython@998c20962c

* METH_FASTCALL changes to .inc files

done via python's Argument Clinic tool,
refer python/cpython@259f0e4437

* Added _PyArg_UnpackStack

refer python/cpython@fe54dda08

* Argument Clinic FASTCALL again

refer python/cpython@0c4a828ca

* Argument Clinic for ordered dictionary object

refer python/cpython@b05cbac052

* speed up getargs

refer python/cpython@1741441649

* FASTCALL for sorted, next, and getattr

refer python/cpython@5a60ecaa7a
refer python/cpython@fda6d0acf0
refer python/cpython@84b388bb80

* Optimize methoddescr_call

refer python/cpython@2a1b676d1f
refer python/cpython@c52572319c
refer python/cpython@35ecebe165
refer python/cpython@8128d5a491

* cleanup _PyMethodDef_RawFastCallDict

refer python/cpython@0a2e46835d
refer python/cpython@98ccba8344
refer python/cpython@c89ef828cf
refer python/cpython@250e4b0063

* print now uses METH_FASTCALL

refer python/cpython@c3858bd7c6
refer python/cpython@bd584f169f
refer python/cpython@06d34393c2

* _struct module now uses Argument Clinic

refer python/cpython@3f2d10132d

* make deque methods faster

refer python/cpython@dd407d5006

* recursive calls in PyObject_Call

refer python/cpython@7399a05965

only partially ported, because RawFastCallKeywords hasn't been ported

* add macros

refer python/cpython@68a001dd59

* all tests pass in MODE=dbg

* convert some internal functions to FASTCALL

__import__ might need to be changed later, if it is possible to backport
the METH_FASTCALL | METH_KEYWORDS flag distinction later.

* speed up unpickling

refer python/cpython@bee09aecc2

* added _PyMethodDef_RawFastCallKeywords

refer python/cpython@7399a05965

* PyCFunction_Call performance

refer python/cpython@12c5838dae

* avoid PyMethodObject in slots

main change in python/cpython@516b98161a
test_exceptions changed in python/cpython@331bbe6aaa
type_settattro changed in python/cpython@193f7e094f
_PyObject_CallFunctionVa changed in python/cpython@fe4ff83049

* fix refcount error found in MODE=dbg

all tests now pass in MODE=dbg

third_party/python/Python/clinic/bltinmodule.inc

@@ -1,11 +1,4 @@
-/*-*- mode:c;indent-tabs-mode:nil;c-basic-offset:4;tab-width:8;coding:utf-8 -*-│
-│vi: set net ft=c ts=4 sts=4 sw=4 fenc=utf-8                                :vi│
-╞══════════════════════════════════════════════════════════════════════════════╡
-│ Python 3                                                                     │
-│ https://docs.python.org/3/license.html                                       │
-╚─────────────────────────────────────────────────────────────────────────────*/
 /* clang-format off */
-
 /*[clinic input]
 preserve
 [clinic start generated code]*/
@@ -90,22 +83,26 @@ PyDoc_STRVAR(builtin_format__doc__,
 "details.");
 
 #define BUILTIN_FORMAT_METHODDEF    \
-    {"format", (PyCFunction)builtin_format, METH_VARARGS, builtin_format__doc__},
+    {"format", (PyCFunction)builtin_format, METH_FASTCALL, builtin_format__doc__},
 
 static PyObject *
 builtin_format_impl(PyObject *module, PyObject *value, PyObject *format_spec);
 
 static PyObject *
-builtin_format(PyObject *module, PyObject *args)
+builtin_format(PyObject *module, PyObject **args, Py_ssize_t nargs, PyObject *kwnames)
 {
     PyObject *return_value = NULL;
     PyObject *value;
     PyObject *format_spec = NULL;
 
-    if (!PyArg_ParseTuple(args, "O|U:format",
+    if (!_PyArg_ParseStack(args, nargs, "O|U:format",
         &value, &format_spec)) {
         goto exit;
     }
+
+    if (!_PyArg_NoStackKeywords("format", kwnames)) {
+        goto exit;
+    }
     return_value = builtin_format_impl(module, value, format_spec);
 
 exit:
@@ -127,9 +124,16 @@ builtin_chr_impl(PyObject *module, int i);
 static PyObject *
 builtin_chr(PyObject *module, PyObject *arg)
 {
+    PyObject *return_value = NULL;
     int i;
-    if (!PyArg_Parse(arg, "i:chr", &i)) return 0;
-    return builtin_chr_impl(module, i);
+
+    if (!PyArg_Parse(arg, "i:chr", &i)) {
+        goto exit;
+    }
+    return_value = builtin_chr_impl(module, i);
+
+exit:
+    return return_value;
 }
 
 PyDoc_STRVAR(builtin_compile__doc__,
@@ -171,7 +175,7 @@ builtin_compile(PyObject *module, PyObject **args, Py_ssize_t nargs, PyObject *k
     int dont_inherit = 0;
     int optimize = -1;
 
-    if (!_PyArg_ParseStack(args, nargs, kwnames, &_parser,
+    if (!_PyArg_ParseStackAndKeywords(args, nargs, kwnames, &_parser,
         &source, PyUnicode_FSDecoder, &filename, &mode, &flags, &dont_inherit, &optimize)) {
         goto exit;
     }
@@ -188,23 +192,27 @@ PyDoc_STRVAR(builtin_divmod__doc__,
 "Return the tuple (x//y, x%y).  Invariant: div*y + mod == x.");
 
 #define BUILTIN_DIVMOD_METHODDEF    \
-    {"divmod", (PyCFunction)builtin_divmod, METH_VARARGS, builtin_divmod__doc__},
+    {"divmod", (PyCFunction)builtin_divmod, METH_FASTCALL, builtin_divmod__doc__},
 
 static PyObject *
 builtin_divmod_impl(PyObject *module, PyObject *x, PyObject *y);
 
 static PyObject *
-builtin_divmod(PyObject *module, PyObject *args)
+builtin_divmod(PyObject *module, PyObject **args, Py_ssize_t nargs, PyObject *kwnames)
 {
     PyObject *return_value = NULL;
     PyObject *x;
     PyObject *y;
 
-    if (!PyArg_UnpackTuple(args, "divmod",
+    if (!_PyArg_UnpackStack(args, nargs, "divmod",
         2, 2,
         &x, &y)) {
         goto exit;
     }
+
+    if (!_PyArg_NoStackKeywords("divmod", kwnames)) {
+        goto exit;
+    }
     return_value = builtin_divmod_impl(module, x, y);
 
 exit:
@@ -224,25 +232,29 @@ PyDoc_STRVAR(builtin_eval__doc__,
 "If only globals is given, locals defaults to it.");
 
 #define BUILTIN_EVAL_METHODDEF    \
-    {"eval", (PyCFunction)builtin_eval, METH_VARARGS, builtin_eval__doc__},
+    {"eval", (PyCFunction)builtin_eval, METH_FASTCALL, builtin_eval__doc__},
 
 static PyObject *
 builtin_eval_impl(PyObject *module, PyObject *source, PyObject *globals,
                   PyObject *locals);
 
 static PyObject *
-builtin_eval(PyObject *module, PyObject *args)
+builtin_eval(PyObject *module, PyObject **args, Py_ssize_t nargs, PyObject *kwnames)
 {
     PyObject *return_value = NULL;
     PyObject *source;
     PyObject *globals = Py_None;
     PyObject *locals = Py_None;
 
-    if (!PyArg_UnpackTuple(args, "eval",
+    if (!_PyArg_UnpackStack(args, nargs, "eval",
         1, 3,
         &source, &globals, &locals)) {
         goto exit;
     }
+
+    if (!_PyArg_NoStackKeywords("eval", kwnames)) {
+        goto exit;
+    }
     return_value = builtin_eval_impl(module, source, globals, locals);
 
 exit:
@@ -262,25 +274,29 @@ PyDoc_STRVAR(builtin_exec__doc__,
 "If only globals is given, locals defaults to it.");
 
 #define BUILTIN_EXEC_METHODDEF    \
-    {"exec", (PyCFunction)builtin_exec, METH_VARARGS, builtin_exec__doc__},
+    {"exec", (PyCFunction)builtin_exec, METH_FASTCALL, builtin_exec__doc__},
 
 static PyObject *
 builtin_exec_impl(PyObject *module, PyObject *source, PyObject *globals,
                   PyObject *locals);
 
 static PyObject *
-builtin_exec(PyObject *module, PyObject *args)
+builtin_exec(PyObject *module, PyObject **args, Py_ssize_t nargs, PyObject *kwnames)
 {
     PyObject *return_value = NULL;
     PyObject *source;
     PyObject *globals = Py_None;
     PyObject *locals = Py_None;
 
-    if (!PyArg_UnpackTuple(args, "exec",
+    if (!_PyArg_UnpackStack(args, nargs, "exec",
         1, 3,
         &source, &globals, &locals)) {
         goto exit;
     }
+
+    if (!_PyArg_NoStackKeywords("exec", kwnames)) {
+        goto exit;
+    }
     return_value = builtin_exec_impl(module, source, globals, locals);
 
 exit:
@@ -317,23 +333,27 @@ PyDoc_STRVAR(builtin_hasattr__doc__,
 "This is done by calling getattr(obj, name) and catching AttributeError.");
 
 #define BUILTIN_HASATTR_METHODDEF    \
-    {"hasattr", (PyCFunction)builtin_hasattr, METH_VARARGS, builtin_hasattr__doc__},
+    {"hasattr", (PyCFunction)builtin_hasattr, METH_FASTCALL, builtin_hasattr__doc__},
 
 static PyObject *
 builtin_hasattr_impl(PyObject *module, PyObject *obj, PyObject *name);
 
 static PyObject *
-builtin_hasattr(PyObject *module, PyObject *args)
+builtin_hasattr(PyObject *module, PyObject **args, Py_ssize_t nargs, PyObject *kwnames)
 {
     PyObject *return_value = NULL;
     PyObject *obj;
     PyObject *name;
 
-    if (!PyArg_UnpackTuple(args, "hasattr",
+    if (!_PyArg_UnpackStack(args, nargs, "hasattr",
         2, 2,
         &obj, &name)) {
         goto exit;
     }
+
+    if (!_PyArg_NoStackKeywords("hasattr", kwnames)) {
+        goto exit;
+    }
     return_value = builtin_hasattr_impl(module, obj, name);
 
 exit:
@@ -361,25 +381,29 @@ PyDoc_STRVAR(builtin_setattr__doc__,
 "setattr(x, \'y\', v) is equivalent to ``x.y = v\'\'");
 
 #define BUILTIN_SETATTR_METHODDEF    \
-    {"setattr", (PyCFunction)builtin_setattr, METH_VARARGS, builtin_setattr__doc__},
+    {"setattr", (PyCFunction)builtin_setattr, METH_FASTCALL, builtin_setattr__doc__},
 
 static PyObject *
 builtin_setattr_impl(PyObject *module, PyObject *obj, PyObject *name,
                      PyObject *value);
 
 static PyObject *
-builtin_setattr(PyObject *module, PyObject *args)
+builtin_setattr(PyObject *module, PyObject **args, Py_ssize_t nargs, PyObject *kwnames)
 {
     PyObject *return_value = NULL;
     PyObject *obj;
     PyObject *name;
     PyObject *value;
 
-    if (!PyArg_UnpackTuple(args, "setattr",
+    if (!_PyArg_UnpackStack(args, nargs, "setattr",
         3, 3,
         &obj, &name, &value)) {
         goto exit;
     }
+
+    if (!_PyArg_NoStackKeywords("setattr", kwnames)) {
+        goto exit;
+    }
     return_value = builtin_setattr_impl(module, obj, name, value);
 
 exit:
@@ -395,23 +419,27 @@ PyDoc_STRVAR(builtin_delattr__doc__,
 "delattr(x, \'y\') is equivalent to ``del x.y\'\'");
 
 #define BUILTIN_DELATTR_METHODDEF    \
-    {"delattr", (PyCFunction)builtin_delattr, METH_VARARGS, builtin_delattr__doc__},
+    {"delattr", (PyCFunction)builtin_delattr, METH_FASTCALL, builtin_delattr__doc__},
 
 static PyObject *
 builtin_delattr_impl(PyObject *module, PyObject *obj, PyObject *name);
 
 static PyObject *
-builtin_delattr(PyObject *module, PyObject *args)
+builtin_delattr(PyObject *module, PyObject **args, Py_ssize_t nargs, PyObject *kwnames)
 {
     PyObject *return_value = NULL;
     PyObject *obj;
     PyObject *name;
 
-    if (!PyArg_UnpackTuple(args, "delattr",
+    if (!_PyArg_UnpackStack(args, nargs, "delattr",
         2, 2,
         &obj, &name)) {
         goto exit;
     }
+
+    if (!_PyArg_NoStackKeywords("delattr", kwnames)) {
+        goto exit;
+    }
     return_value = builtin_delattr_impl(module, obj, name);
 
 exit:
@@ -504,24 +532,28 @@ PyDoc_STRVAR(builtin_pow__doc__,
 "invoked using the three argument form.");
 
 #define BUILTIN_POW_METHODDEF    \
-    {"pow", (PyCFunction)builtin_pow, METH_VARARGS, builtin_pow__doc__},
+    {"pow", (PyCFunction)builtin_pow, METH_FASTCALL, builtin_pow__doc__},
 
 static PyObject *
 builtin_pow_impl(PyObject *module, PyObject *x, PyObject *y, PyObject *z);
 
 static PyObject *
-builtin_pow(PyObject *module, PyObject *args)
+builtin_pow(PyObject *module, PyObject **args, Py_ssize_t nargs, PyObject *kwnames)
 {
     PyObject *return_value = NULL;
     PyObject *x;
     PyObject *y;
     PyObject *z = Py_None;
 
-    if (!PyArg_UnpackTuple(args, "pow",
+    if (!_PyArg_UnpackStack(args, nargs, "pow",
         2, 3,
         &x, &y, &z)) {
         goto exit;
     }
+
+    if (!_PyArg_NoStackKeywords("pow", kwnames)) {
+        goto exit;
+    }
     return_value = builtin_pow_impl(module, x, y, z);
 
 exit:
@@ -541,22 +573,26 @@ PyDoc_STRVAR(builtin_input__doc__,
 "On *nix systems, readline is used if available.");
 
 #define BUILTIN_INPUT_METHODDEF    \
-    {"input", (PyCFunction)builtin_input, METH_VARARGS, builtin_input__doc__},
+    {"input", (PyCFunction)builtin_input, METH_FASTCALL, builtin_input__doc__},
 
 static PyObject *
 builtin_input_impl(PyObject *module, PyObject *prompt);
 
 static PyObject *
-builtin_input(PyObject *module, PyObject *args)
+builtin_input(PyObject *module, PyObject **args, Py_ssize_t nargs, PyObject *kwnames)
 {
     PyObject *return_value = NULL;
     PyObject *prompt = NULL;
 
-    if (!PyArg_UnpackTuple(args, "input",
+    if (!_PyArg_UnpackStack(args, nargs, "input",
         0, 1,
         &prompt)) {
         goto exit;
     }
+
+    if (!_PyArg_NoStackKeywords("input", kwnames)) {
+        goto exit;
+    }
     return_value = builtin_input_impl(module, prompt);
 
 exit:
@@ -585,23 +621,27 @@ PyDoc_STRVAR(builtin_sum__doc__,
 "reject non-numeric types.");
 
 #define BUILTIN_SUM_METHODDEF    \
-    {"sum", (PyCFunction)builtin_sum, METH_VARARGS, builtin_sum__doc__},
+    {"sum", (PyCFunction)builtin_sum, METH_FASTCALL, builtin_sum__doc__},
 
 static PyObject *
 builtin_sum_impl(PyObject *module, PyObject *iterable, PyObject *start);
 
 static PyObject *
-builtin_sum(PyObject *module, PyObject *args)
+builtin_sum(PyObject *module, PyObject **args, Py_ssize_t nargs, PyObject *kwnames)
 {
     PyObject *return_value = NULL;
     PyObject *iterable;
     PyObject *start = NULL;
 
-    if (!PyArg_UnpackTuple(args, "sum",
+    if (!_PyArg_UnpackStack(args, nargs, "sum",
         1, 2,
         &iterable, &start)) {
         goto exit;
     }
+
+    if (!_PyArg_NoStackKeywords("sum", kwnames)) {
+        goto exit;
+    }
     return_value = builtin_sum_impl(module, iterable, start);
 
 exit:
@@ -619,24 +659,28 @@ PyDoc_STRVAR(builtin_isinstance__doc__,
 "or ...`` etc.");
 
 #define BUILTIN_ISINSTANCE_METHODDEF    \
-    {"isinstance", (PyCFunction)builtin_isinstance, METH_VARARGS, builtin_isinstance__doc__},
+    {"isinstance", (PyCFunction)builtin_isinstance, METH_FASTCALL, builtin_isinstance__doc__},
 
 static PyObject *
 builtin_isinstance_impl(PyObject *module, PyObject *obj,
                         PyObject *class_or_tuple);
 
 static PyObject *
-builtin_isinstance(PyObject *module, PyObject *args)
+builtin_isinstance(PyObject *module, PyObject **args, Py_ssize_t nargs, PyObject *kwnames)
 {
     PyObject *return_value = NULL;
     PyObject *obj;
     PyObject *class_or_tuple;
 
-    if (!PyArg_UnpackTuple(args, "isinstance",
+    if (!_PyArg_UnpackStack(args, nargs, "isinstance",
         2, 2,
         &obj, &class_or_tuple)) {
         goto exit;
     }
+
+    if (!_PyArg_NoStackKeywords("isinstance", kwnames)) {
+        goto exit;
+    }
     return_value = builtin_isinstance_impl(module, obj, class_or_tuple);
 
 exit:
@@ -654,27 +698,31 @@ PyDoc_STRVAR(builtin_issubclass__doc__,
 "or ...`` etc.");
 
 #define BUILTIN_ISSUBCLASS_METHODDEF    \
-    {"issubclass", (PyCFunction)builtin_issubclass, METH_VARARGS, builtin_issubclass__doc__},
+    {"issubclass", (PyCFunction)builtin_issubclass, METH_FASTCALL, builtin_issubclass__doc__},
 
 static PyObject *
 builtin_issubclass_impl(PyObject *module, PyObject *cls,
                         PyObject *class_or_tuple);
 
 static PyObject *
-builtin_issubclass(PyObject *module, PyObject *args)
+builtin_issubclass(PyObject *module, PyObject **args, Py_ssize_t nargs, PyObject *kwnames)
 {
     PyObject *return_value = NULL;
     PyObject *cls;
     PyObject *class_or_tuple;
 
-    if (!PyArg_UnpackTuple(args, "issubclass",
+    if (!_PyArg_UnpackStack(args, nargs, "issubclass",
         2, 2,
         &cls, &class_or_tuple)) {
         goto exit;
     }
+
+    if (!_PyArg_NoStackKeywords("issubclass", kwnames)) {
+        goto exit;
+    }
     return_value = builtin_issubclass_impl(module, cls, class_or_tuple);
 
 exit:
     return return_value;
 }
-/*[clinic end generated code: output=2ef82846acdfa0f5 input=a9049054013a1b77]*/
+/*[clinic end generated code: output=17fedd2dec148677 input=a9049054013a1b77]*/