pycore_pystate.h no longer redefines PyThreadState_GET() (GH-28921)

Redefining the PyThreadState_GET() macro in pycore_pystate.h is
useless since it doesn't affect files not including it. Either use
_PyThreadState_GET() directly, or don't use pycore_pystate.h internal
C API. For example, the _testcapi extension don't use the internal C
API, but use the public PyThreadState_Get() function instead.

Replace PyThreadState_Get() with _PyThreadState_GET(). The
_PyThreadState_GET() macro is more efficient than PyThreadState_Get()
and PyThreadState_GET() function calls which call fail with a fatal
Python error.

posixmodule.c and _ctypes extension now include <windows.h> before
pycore header files (like pycore_call.h).

_PyTraceback_Add() now uses _PyErr_Fetch()/_PyErr_Restore() instead
of PyErr_Fetch()/PyErr_Restore().

The _decimal and _xxsubinterpreters extensions are now built with the
Py_BUILD_CORE_MODULE macro defined to get access to the internal C
API.

Python/bltinmodule.c

@@ -213,7 +213,7 @@ builtin___build_class__(PyObject *self, PyObject *const *args, Py_ssize_t nargs,
         goto error;
     }
     PyFrameConstructor *f =  PyFunction_AS_FRAME_CONSTRUCTOR(func);
-    PyThreadState *tstate = PyThreadState_GET();
+    PyThreadState *tstate = _PyThreadState_GET();
     cell = _PyEval_Vector(tstate, f, ns, NULL, 0, NULL);
     if (cell != NULL) {
         if (bases != orig_bases) {

Python/ceval.c

@@ -1064,7 +1064,7 @@ static int unpack_iterable(PyThreadState *, PyObject *, int, int, PyObject **);
 PyObject *
 PyEval_EvalCode(PyObject *co, PyObject *globals, PyObject *locals)
 {
-    PyThreadState *tstate = PyThreadState_GET();
+    PyThreadState *tstate = _PyThreadState_GET();
     if (locals == NULL) {
         locals = globals;
     }

Python/context.c

@@ -728,7 +728,7 @@ static int
 contextvar_set(PyContextVar *var, PyObject *val)
 {
     var->var_cached = NULL;
-    PyThreadState *ts = PyThreadState_Get();
+    PyThreadState *ts = _PyThreadState_GET();
 
     PyContext *ctx = context_get();
     if (ctx == NULL) {

Python/pylifecycle.c

@@ -472,7 +472,7 @@ interpreter_update_config(PyThreadState *tstate, int only_update_path_config)
 int
 _PyInterpreterState_SetConfig(const PyConfig *src_config)
 {
-    PyThreadState *tstate = PyThreadState_Get();
+    PyThreadState *tstate = _PyThreadState_GET();
     int res = -1;
 
     PyConfig config;

Python/pystate.c

@@ -1671,8 +1671,11 @@ _check_xidata(PyThreadState *tstate, _PyCrossInterpreterData *data)
 int
 _PyObject_GetCrossInterpreterData(PyObject *obj, _PyCrossInterpreterData *data)
 {
-    // PyThreadState_Get() aborts if tstate is NULL.
-    PyThreadState *tstate = PyThreadState_Get();
+    PyThreadState *tstate = _PyThreadState_GET();
+#ifdef Py_DEBUG
+    // The caller must hold the GIL
+    _Py_EnsureTstateNotNULL(tstate);
+#endif
     PyInterpreterState *interp = tstate->interp;
 
     // Reset data before re-populating.

Python/traceback.c

@@ -6,11 +6,13 @@
 #include "code.h"                 // PyCode_Addr2Line etc
 #include "pycore_interp.h"        // PyInterpreterState.gc
 #include "frameobject.h"          // PyFrame_GetBack()
-#include "pycore_frame.h"         // _PyFrame_GetCode()
-#include "pycore_pyarena.h"       // _PyArena_Free()
 #include "pycore_ast.h"           // asdl_seq_*
 #include "pycore_compile.h"       // _PyAST_Optimize
+#include "pycore_frame.h"         // _PyFrame_GetCode()
 #include "pycore_parser.h"        // _PyParser_ASTFromString
+#include "pycore_pyarena.h"       // _PyArena_Free()
+#include "pycore_pyerrors.h"      // _PyErr_Fetch()
+#include "pycore_pystate.h"       // _PyThreadState_GET()
 #include "../Parser/pegen.h"      // _PyPegen_byte_offset_to_character_offset()
 #include "structmember.h"         // PyMemberDef
 #include "osdefs.h"               // SEP
@@ -267,11 +269,12 @@ void _PyTraceback_Add(const char *funcname, const char *filename, int lineno)
     PyCodeObject *code;
     PyFrameObject *frame;
     PyObject *exc, *val, *tb;
+    PyThreadState *tstate = _PyThreadState_GET();
 
     /* Save and clear the current exception. Python functions must not be
        called with an exception set. Calling Python functions happens when
        the codec of the filesystem encoding is implemented in pure Python. */
-    PyErr_Fetch(&exc, &val, &tb);
+    _PyErr_Fetch(tstate, &exc, &val, &tb);
 
     globals = PyDict_New();
     if (!globals)
@@ -281,14 +284,14 @@ void _PyTraceback_Add(const char *funcname, const char *filename, int lineno)
         Py_DECREF(globals);
         goto error;
     }
-    frame = PyFrame_New(PyThreadState_Get(), code, globals, NULL);
+    frame = PyFrame_New(tstate, code, globals, NULL);
     Py_DECREF(globals);
     Py_DECREF(code);
     if (!frame)
         goto error;
     frame->f_lineno = lineno;
 
-    PyErr_Restore(exc, val, tb);
+    _PyErr_Restore(tstate, exc, val, tb);
     PyTraceBack_Here(frame);
     Py_DECREF(frame);
     return;