sinfo.ads: Documentation update.

2014-01-24  Ed Schonberg  <schonberg@adacore.com>

	* sinfo.ads: Documentation update.

2014-01-24  Ed Schonberg  <schonberg@adacore.com>

	* sem_ch3.adb (Constant_Redeclaration): New declaration is
	illegal if previous one has an initial expression that is an
	aggregate expanded into assignments.

2014-01-24  Ed Schonberg  <schonberg@adacore.com>

	* sem_ch5.adb (Analyze_Loop_Parameter_Specification): Small
	code reorganization to remove spurious warning on a loop with
	an array element iterator that has a null range.

From-SVN: r207041

4 files changed, 96 insertions, 22 deletions

diff --git a/gcc/ada/ChangeLog b/gcc/ada/ChangeLog
index 41a0ed0..59072a6 100644
--- a/gcc/ada/ChangeLog
+++ b/gcc/ada/ChangeLog
@@ -1,3 +1,35 @@
+2014-01-24  Ed Schonberg  <schonberg@adacore.com>
+
+	* sinfo.ads: Documentation update.
+
+2014-01-24  Ed Schonberg  <schonberg@adacore.com>
+
+	* sem_ch3.adb (Constant_Redeclaration): New declaration is
+	illegal if previous one has an initial expression that is an
+	aggregate expanded into assignments.
+
+2014-01-24  Ed Schonberg  <schonberg@adacore.com>
+
+	* sem_ch5.adb (Analyze_Loop_Parameter_Specification): Small
+	code reorganization to remove spurious warning on a loop with
+	an array element iterator that has a null range.
+
+2014-01-24  Vincent Celier  <celier@adacore.com>
+
+	* make.adb (Binding_Phase): When setting the Ada paths, use
+	the library ALI dir, not the object dir in libraries.
+
+2014-01-24  Yannick Moy  <moy@adacore.com>
+
+	* sinfo.ads: Add documentation of check flag fields.
+
+2014-01-24  Ed Schonberg  <schonberg@adacore.com>
+
+	* sem_res.adb (Resolve_Actuals): If an actual is a view
+	conversion of a discriminated object, and the formal type is
+	discriminated and constrained, apply a discriminant check to
+	the object itself.
+
 2014-01-24  Robert Dewar  <dewar@adacore.com>
 
 	* prj.adb, prj-env.adb, back_end.adb: Add comment, minor code clean ups.
diff --git a/gcc/ada/sem_ch3.adb b/gcc/ada/sem_ch3.adb
index cdfc062..014105b 100644
--- a/gcc/ada/sem_ch3.adb
+++ b/gcc/ada/sem_ch3.adb
@@ -11104,10 +11104,13 @@ package body Sem_Ch3 is
       --  If previous full declaration or a renaming declaration exists, or if
       --  a homograph is present, let Enter_Name handle it, either with an
       --  error or with the removal of an overridden implicit subprogram.
+      --  The previous one is a full declaration if it has an expression
+      --  (which in the case of an aggregate is indicated by the Init flag).
 
       if Ekind (Prev) /= E_Constant
         or else Nkind (Parent (Prev)) = N_Object_Renaming_Declaration
         or else Present (Expression (Parent (Prev)))
+        or else Has_Init_Expression (Parent (Prev))
         or else Present (Full_View (Prev))
       then
          Enter_Name (Id);
diff --git a/gcc/ada/sem_ch5.adb b/gcc/ada/sem_ch5.adb
index a5be290..1d64709 100644
--- a/gcc/ada/sem_ch5.adb
+++ b/gcc/ada/sem_ch5.adb
@@ -2457,9 +2457,11 @@ package body Sem_Ch5 is
 
       --  Check for null or possibly null range and issue warning. We suppress
       --  such messages in generic templates and instances, because in practice
-      --  they tend to be dubious in these cases.
+      --  they tend to be dubious in these cases. The check applies as well to
+      --  rewritten array element loops where a null range may be detected
+      --  statically.
 
-      if Nkind (DS) = N_Range and then Comes_From_Source (N) then
+      if Nkind (DS) = N_Range then
          declare
             L : constant Node_Id := Low_Bound  (DS);
             H : constant Node_Id := High_Bound (DS);
@@ -2481,21 +2483,23 @@ package body Sem_Ch5 is
                   if Compile_Time_Compare
                        (L, H, Assume_Valid => False) = GT
                   then
-                     Error_Msg_N
-                       ("??loop range is null, loop will not execute", DS);
-
                      --  Since we know the range of the loop is null, set the
                      --  appropriate flag to remove the loop entirely during
                      --  expansion.
 
                      Set_Is_Null_Loop (Loop_Nod);
 
-                  --  Here is where the loop could execute because of invalid
-                  --  values, so issue appropriate message and in this case we
-                  --  do not set the Is_Null_Loop flag since the loop may
-                  --  execute.
+                     if Comes_From_Source (N) then
+                        Error_Msg_N
+                          ("??loop range is null, loop will not execute", DS);
+                     end if;
 
-                  else
+                     --  Here is where the loop could execute because of
+                     --  invalid values, so issue appropriate message and in
+                     --  this case we do not set the Is_Null_Loop flag since
+                     --  the loop may execute.
+
+                  elsif Comes_From_Source (N) then
                      Error_Msg_N
                        ("??loop range may be null, loop may not execute",
                         DS);
diff --git a/gcc/ada/sinfo.ads b/gcc/ada/sinfo.ads
index cc11f4f..f71fdd0 100644
--- a/gcc/ada/sinfo.ads
+++ b/gcc/ada/sinfo.ads
@@ -516,12 +516,12 @@ package Sinfo is
    --  expansion is performed and the analysis must generate a tree in a
    --  form that meets additional requirements.
 
-   --  This light expansion does two transformations of the tree, that cannot
-   --  be postponed after the frontend semantic analysis:
+   --  This light expansion does two transformations of the tree that cannot
+   --  be postponed after semantic analysis:
 
    --    1. Replace object renamings by renamed object. This requires the
-   --       introdtion of temporaries at the point of the renaming, which must
-   --       be properly analyzed.
+   --       introduction of temporaries at the point of the renaming, which
+   --       must be properly analyzed.
 
    --    2. Fully qualify entity names. This is needed to generate suitable
    --       local effects and call-graphs in ALI files, with the completely
@@ -549,6 +549,39 @@ package Sinfo is
    --  not make sense from a user point-of-view, and that cross-references that
    --  do not lead to data dependences for subprograms can be safely ignored.
 
+   -----------------------
+   -- Check Flag Fields --
+   -----------------------
+
+   --  The following flag fields appear in expression nodes:
+
+   --  Do_Division_Check
+   --  Do_Overflow_Check
+   --  Do_Range_Check
+
+   --  These three flags are always set by the front end during semantic
+   --  analysis, on expression nodes that may trigger the corresponding
+   --  check. The front end then inserts or not the check during expansion.
+   --  In particular, these flags should also be correctly set in ASIS mode
+   --  and GNATprove mode.
+
+   --  Note that this accounts for all nodes that trigger the corresponding
+   --  checks, except for range checks on subtype_indications, which may be
+   --  required to check that a range_constraint is compatible with the given
+   --  subtype (RM 3.2.2(11)).
+
+   --  The following flag fields appear in various nodes:
+
+   --  Do_Accessibility_Check
+   --  Do_Discriminant_Check
+   --  Do_Length_Check
+   --  Do_Storage_Check
+   --  Do_Tag_Check
+
+   --  These flags are used in some specific cases by the front end, either
+   --  during semantic analysis or during expansion, and cannot be expected
+   --  to be set on all nodes that trigger the corresponding check.
+
    ------------------------
    -- Common Flag Fields --
    ------------------------
@@ -2430,12 +2463,14 @@ package Sinfo is
       --  Etype (Node5-Sem)
       --  Must_Not_Freeze (Flag8-Sem)
 
-      --  Note: Etype is a copy of the Etype field of the Subtype_Mark. The
-      --  reason for this redundancy is so that in a list of array index types,
-      --  the Etype can be uniformly accessed to determine the subscript type.
-      --  This means that no Itype is constructed for the actual subtype that
-      --  is created by the subtype indication. If such an Itype is required,
-      --  it is constructed in the context in which the indication appears.
+      --  Note: Depending on context, the Etype is either the entity of the
+      --  Subtype_Mark field, or it is an itype constructed to reify the
+      --  subtype indication. In particular, such itypes are created for a
+      --  subtype indication that appears in an array type declaration. This
+      --  simplifies constraint checking in indexed components.
+
+      --  For subtype indications that appear in scalar type and subtype
+      --  declarations, the Etype is the entity of the subtype mark.
 
       -------------------------
       -- 3.2.2  Subtype Mark --
@@ -7325,7 +7360,7 @@ package Sinfo is
       --  N_Expression_With_Actions has type Standard_Void_Type. However some
       --  backends do not support such expression-with-actions occurring
       --  outside of a proper (non-void) expression, so this should just be
-      --  used as an intermediate representation within the front-end. Also
+      --  used as an intermediate representation within the front end. Also
       --  note that this is really an irregularity (expressions and statements
       --  are not interchangeable, and in particular an N_Null_Statement is
       --  not a proper expression), and in the long term all cases of this
@@ -7746,7 +7781,7 @@ package Sinfo is
       --  e.g. involving unconstrained array types.
 
       --  For the case of the standard gigi backend, this means that all
-      --  checks are done in the front-end.
+      --  checks are done in the front end.
 
       --  However, in the case of specialized back-ends, notably the JVM
       --  backend for JGNAT, additional requirements and restrictions apply