Normilize queue and device + change copying from numpy in call_origin

Author: densmirn

dpnp/dpnp_container.py

@@ -43,6 +43,7 @@
 import dpctl.tensor as dpt
 from dpctl.tensor._device import normalize_queue_device
 
+
 if config.__DPNP_OUTPUT_DPCTL__:
     try:
         """

dpnp/dpnp_utils/dpnp_algo_utils.pyx

@@ -97,8 +97,27 @@ def convert_list_args(input_list):
 
     return result_list
 
+def copy_from_origin(src, device=None, usm_type="device", sycl_queue=None):
+    """Copy result from origin."""
+    if not isinstance(src, numpy.ndarray):
+        return src
+
+    if src.size == 0:
+        return dpnp_container.empty(src.shape,
+                                    dtype=src.dtype,
+                                    device=device,
+                                    usm_type=usm_type,
+                                    sycl_queue=sycl_queue)
+
+    array_obj = dpctl.tensor._copy_utils.from_numpy(src,
+                                                    device=device,
+                                                    usm_type=usm_type,
+                                                    sycl_queue=sycl_queue)
 
-def copy_from_origin(dst, src):
+    return dpnp.dpnp_array.dpnp_array(src.shape, buffer=array_obj)
+
+
+def copy_from_origin_into(dst, src):
     """Copy origin result to output result."""
     if config.__DPNP_OUTPUT_DPCTL__ and hasattr(dst, "__sycl_usm_array_interface__"):
         if src.size:
@@ -145,7 +164,7 @@ def call_origin(function, *args, **kwargs):
         if args and args_new:
             arg, arg_new = args[0], args_new[0]
             if isinstance(arg_new, numpy.ndarray):
-                copy_from_origin(arg, arg_new)
+                copy_from_origin_into(arg, arg_new)
             elif isinstance(arg_new, list):
                 for i, val in enumerate(arg_new):
                     arg[i] = val
@@ -157,20 +176,18 @@ def call_origin(function, *args, **kwargs):
             if (kwargs_dtype is not None):
                 result_dtype = kwargs_dtype
 
-            result = dpnp_container.empty(result_origin.shape, dtype=result_dtype)
+            result = copy_from_origin(result_origin)
         else:
             result = kwargs_out
-
-        copy_from_origin(result, result_origin)
+            copy_from_origin_into(result, result_origin)
 
     elif isinstance(result, tuple):
         # convert tuple(fallback_array) to tuple(result_array)
         result_list = []
         for res_origin in result:
             res = res_origin
             if isinstance(res_origin, numpy.ndarray):
-                res = dpnp_container.empty(res_origin.shape, dtype=res_origin.dtype)
-                copy_from_origin(res, res_origin)
+                res = copy_from_origin(res_origin)
             result_list.append(res)
 
         result = tuple(result_list)