The problem comes from the construction of Stream operations, which
happens at the point a tagged type is frozen. Streams need to see the
full view of types, so that for example the Read attribute for an array
can be expanded into a loop over the Read attribute for the component
type. Now if during that expansion we have a private type we may need
to retrieve the full view of the type to find its structure. And the
corresponding tree must be compiled in the context of the package that
defines the type, which might not be the current package. The problem
arises when both array and component types are private, then the
described mechanism is invoked twice.
Tested on x86_64-pc-linux-gnu, committed on trunk
gcc/ada/
* exp_attr.adb (Compile_Stream_Body_In_Scope): Do not install
package if array type and element type come from the same
package, and the original array type is private.diff --git a/gcc/ada/exp_attr.adb b/gcc/ada/exp_attr.adb
--- a/gcc/ada/exp_attr.adb
+++ b/gcc/ada/exp_attr.adb
@@ -888,6 +888,11 @@ package body Exp_Attr is
-- special stream-processing operations for that type (for example
-- Unbounded_String and its wide varieties).
+ -- We don't install the package either if array type and element
+ -- type come from the same package, and the original array type is
+ -- private, because in this case the underlying type Arr is
+ -- itself a full view, which carries the full view of the component.
+
Scop := Scope (C_Type);
if Is_Private_Type (C_Type)
@@ -896,7 +901,15 @@ package body Exp_Attr is
and then Ekind (Scop) = E_Package
and then No (Get_Stream_Convert_Pragma (C_Type))
then
- Install := True;
+ if Scope (Arr) = Scope (C_Type)
+ and then Is_Private_Type (Etype (Prefix (N)))
+ and then Full_View (Etype (Prefix (N))) = Arr
+ then
+ null;
+
+ else
+ Install := True;
+ end if;
end if;
end if;
