diff --git a/gcc/ada/exp_ch7.adb b/gcc/ada/exp_ch7.adb
--- a/gcc/ada/exp_ch7.adb
+++ b/gcc/ada/exp_ch7.adb
@@ -43,6 +43,7 @@ with Exp_Prag;  use Exp_Prag;
 with Exp_Tss;   use Exp_Tss;
 with Exp_Util;  use Exp_Util;
 with Freeze;    use Freeze;
+with GNAT_CUDA; use GNAT_CUDA;
 with Lib;       use Lib;
 with Nlists;    use Nlists;
 with Nmake;     use Nmake;
@@ -5756,6 +5757,13 @@ package body Exp_Ch7 is
             Build_Static_Dispatch_Tables (N);
          end if;
 
+         --  If procedures marked with CUDA_Global have been defined within N,
+         --  we need to register them with the CUDA runtime at program startup.
+         --  This requires multiple declarations and function calls which need
+         --  to be appended to N's declarations.
+
+         Build_And_Insert_CUDA_Initialization (N);
+
          Build_Task_Activation_Call (N);
 
          --  Verify the run-time semantics of pragma Initial_Condition at the


diff --git a/gcc/ada/sem_ch7.adb b/gcc/ada/sem_ch7.adb
--- a/gcc/ada/sem_ch7.adb
+++ b/gcc/ada/sem_ch7.adb
@@ -40,7 +40,6 @@ with Exp_Dist;  use Exp_Dist;
 with Exp_Dbug;  use Exp_Dbug;
 with Freeze;    use Freeze;
 with Ghost;     use Ghost;
-with GNAT_CUDA; use GNAT_CUDA;
 with Lib;       use Lib;
 with Lib.Xref;  use Lib.Xref;
 with Namet;     use Namet;
@@ -1000,13 +999,6 @@ package body Sem_Ch7 is
          Analyze_List (Declarations (N));
       end if;
 
-      --  If procedures marked with CUDA_Global have been defined within N, we
-      --  need to register them with the CUDA runtime at program startup. This
-      --  requires multiple declarations and function calls which need to be
-      --  appended to N's declarations.
-
-      Build_And_Insert_CUDA_Initialization (N);
-
       HSS := Handled_Statement_Sequence (N);
 
       if Present (HSS) then