fix(lite): support set data by copy on device tensor

GitOrigin-RevId: 88b7f73d364d63bb3cad95ec063f283048895f94

lite/pylite/megenginelite/tensor.py

@@ -407,7 +407,7 @@ class LiteTensor(object):
 
     def set_data_by_copy(self, data, data_length=0, layout=None):
         """
-        copy the data to the tensor
+        copy the data to the tensor, the memory of the tensor must be continue
         param data: the data to copy to tensor, it should be list,
         numpy.ndarraya or ctypes with length
         """
@@ -415,37 +415,34 @@ class LiteTensor(object):
             self.layout = layout
 
         assert self.is_continue, "set_data_by_copy can only apply in continue tensor."
-        assert (
-            self.is_pinned_host or self.device_type == LiteDeviceType.LITE_CPU
-        ), "set_data_by_copy can only apply in cpu tensor or pinned tensor."
 
         c_type = _lite_dtypes_to_ctype[LiteDataType(self._layout.data_type)]
 
-        tensor_memory = c_void_p()
+        cpu_tensor = LiteTensor(self._layout)
+        tensor_length = self.nbytes
 
         if type(data) == list:
             length = len(data)
-            self._api.LITE_get_tensor_memory(self._tensor, byref(tensor_memory))
-            tensor_length = self.nbytes
             assert (
                 length * sizeof(c_type) <= tensor_length
             ), "the length of input data to set to the tensor is too large."
-            arr = (c_type * length)(*data)
-            memmove(tensor_memory, arr, sizeof(c_type) * length)
+            cdata = (c_type * length)(*data)
+            self._api.LITE_reset_tensor_memory(cpu_tensor._tensor, cdata, tensor_length)
+            self.copy_from(cpu_tensor)
 
         elif type(data) == np.ndarray:
-            if self.nbytes != data.nbytes:
-                self.layout = LiteLayout(data.shape, data.dtype)
-            arr = data.ctypes.data_as(POINTER(c_type))
-            self._api.LITE_get_tensor_memory(self._tensor, byref(tensor_memory))
-            assert self.nbytes == data.nbytes
-            memmove(tensor_memory, arr, self.nbytes)
+            self.layout = LiteLayout(data.shape, data.dtype)
+            cpu_tensor.layout = LiteLayout(data.shape, data.dtype)
+            cdata = data.ctypes.data_as(POINTER(c_type))
+            self._api.LITE_reset_tensor_memory(cpu_tensor._tensor, cdata, self.nbytes)
+            self.copy_from(cpu_tensor)
+
         else:
             assert (
                 data_length == self.nbytes or layout is not None
             ), "when input data is ctypes, the length of input data or layout must set"
-            self._api.LITE_get_tensor_memory(self._tensor, byref(tensor_memory))
-            memmove(tensor_memory, data, data_length)
+            self._api.LITE_reset_tensor_memory(cpu_tensor._tensor, data, tensor_length)
+            self.copy_from(cpu_tensor)
 
     def get_data_by_share(self):
         """
@@ -454,6 +451,7 @@ class LiteTensor(object):
         the tensor memory is write again, such as LiteNetwok forward next time.
         """
 
+        self.update()
         buffer = c_void_p()
         self._api.LITE_get_tensor_memory(self._tensor, byref(buffer))
         buffer = self.np_array_type.from_address(buffer.value)

lite/pylite/test/test_tensor.py

@@ -323,3 +323,27 @@ def test_tensor_get_memory_by_share():
     tensor.set_data_by_copy(arr)
     assert test_data[1][18] == 5
     assert test_data[3][7] == 345
+
+
+@require_cuda
+def test_tensor_set_data_device():
+    layout = LiteLayout([2, 16], "int8")
+    tensor = LiteTensor(layout, device_type=LiteDeviceType.LITE_CUDA)
+    assert tensor.nbytes == 2 * 16
+
+    data = [i for i in range(32)]
+    tensor.set_data_by_copy(data)
+    real_data = tensor.to_numpy()
+    for i in range(32):
+        assert real_data[i // 16][i % 16] == i
+
+    arr = np.ones([2, 16], "int8")
+    tensor.set_data_by_copy(arr)
+    real_data = tensor.to_numpy()
+    for i in range(32):
+        assert real_data[i // 16][i % 16] == 1
+
+    tensor.set_data_by_copy(list(range(32)))
+    real_data = tensor.to_numpy()
+    for i in range(32):
+        assert real_data[i // 16][i % 16] == i