Updated doc

doc/library/compile/shared.txt

@@ -28,9 +28,17 @@
                internally to store the shared variable.
        :type return_internal_type: bool
 
-       Changes to this value will be visible to all functions using this SharedVariable.
+       By default, return a copy of the data. If ``borrow=True`` (and
+       ``return_internal_type=False``), maybe it will return a copy.
+       For tensor, it will always return a ndarray by default, so if
+       the data is on the GPU, it will return a copy, but if the data
+       is on the CPU, it will return the original data.  If you do
+       ``borrow=True`` and ``return_internal_type=True``, it will
+       always return the original data, not a copy, but this can be a
+       GPU object.
     
     .. method:: set_value(self, new_value, borrow=False)
+
        :param new_value: The new value.
        :type new_value: A compatible type for this shared variable.
 
@@ -38,8 +46,7 @@
            related to aliased memory.
        :type borrow: bool
 
-       Changes to this value will be visible to all functions using this SharedVariable.
-        
+       The new value will be seen by all functions using this SharedVariable.
 
     .. method:: __init__(self, name, type, value, strict, container=None)