PyBytes_FromStringAndSize() uses a global cache for 1-byte bytes:
https://github.com/python/cpython/blob/v3.10.0a4/Objects/bytesobject.c#L147

    if (size == 1 && str != NULL) {
        struct _Py_bytes_state *state = get_bytes_state();
        op = state->characters[*str & UCHAR_MAX];
        if (op != NULL) {
            Py_INCREF(op);
            return (PyObject *)op;
        }
    }

_PyBytes_Resize() will raise an error when (refcount != 1):
https://github.com/python/cpython/blob/v3.10.0a4/Objects/bytesobject.c#L3029

Then this code will raise an exception:

    obj1 = PyBytes_FromStringAndSize("a", 1);
    obj2 = PyBytes_FromStringAndSize("a", 1);
    ret = _PyBytes_Resize(&obj2, 2);  // ret is -1

BTW, 0-byte bytes comes from a global singleton, but _PyBytes_Resize() processes it before checking refcount:
https://github.com/python/cpython/blob/v3.10.0a4/Objects/bytesobject.c#L3021

    if (Py_SIZE(v) == 0) {
        if (newsize == 0) {
            return 0;
        }
        *pv = _PyBytes_FromSize(newsize, 0);
        Py_DECREF(v);
        return (*pv == NULL) ? -1 : 0;
    }
    if (Py_REFCNT(v) != 1) {
        goto error;
    }

_PyBytes_Resize() doc:
    
    Only use this to build up a brand new bytes object; don’t use this if the
    bytes may already be known in other parts of the code. It is an error to
    call this function if the refcount on the input bytes object is not one.