Bug report

Bug description:

Python'sceval.c contains the function

PyObject*_Py_CallBuiltinClass_StackRefSteal(_PyStackRefcallable,_PyStackRef*arguments,inttotal_args){PyObject*res;STACKREFS_TO_PYOBJECTS(arguments,total_args,args_o);if (CONVERSION_FAILED(args_o)) {res=NULL;        gotocleanup;    }PyTypeObject*tp= (PyTypeObject*)PyStackRef_AsPyObjectBorrow(callable);res=tp->tp_vectorcall((PyObject*)tp,args_o,total_args,NULL);/* <----- relevant line */   ....}

The call totp_vectorcall does not declarePY_VECTORCALL_ARGUMENTS_OFFSET despiteSTACKREFS_TO_PYOBJECTS (via_PyObjectArray_FromStackRefArray) always allocating anargs_o that has an extra scratch space at position -1.

Strictly speaking, this isn't incorrect. But it can force the caller to make a redundant copy of the arguments, which is inefficient. Best practice to allocate the extra entry and then communicate its presence to the vector-callee.

This inefficiency is not restricted to Python-internal types. It also affects custom heap types that are themselves vector-callable. In this case, Python's adaptive optimizing interpreter replacesCALL_NON_PY_GENERAL withCALL_BUILTIN_CLASS that then goes through this code path.

cc@markshannon@vstinner

CPython versions tested on:

3.14

Operating systems tested on:

macOS

Linked PRs

• gh-143361: Pass PY_VECTORCALL_ARGUMENTS_OFFSET in _Py_CallBuiltinClass_StackRefSteal #143367