+2014-08-01 Robert Dewar <dewar@adacore.com>
+
+ * sem_ch8.adb, opt.ads Minor comment updates.
+
+2014-08-01 Hristian Kirtchev <kirtchev@adacore.com>
+
+ * exp_intr.adb (Expand_Unc_Deallocation): Request a renaming
+ from the side effects removal machinery.
+ * exp_util.adb (Duplicate_Subexpr): Add formal parameter
+ Renaming_Req. Update the nested call to Remove_Side_Effects.
+ (Duplicate_Subexpr_No_Checks): Add formal parameter
+ Renaming_Req. Update the nested call to Remove_Side_Effects.
+ (Duplicate_Subexpr_Move_Checks): Add formal parameter
+ Renaming_Req. Update the nested call to Remove_Side_Effects.
+ (Remove_Side_Effects): Add formal parameter Renaming_Req. Generate
+ an object renaming declaration when the caller requests it.
+ * exp_util.ads (Duplicate_Subexpr): Add formal
+ parameter Renaming_Req. Update comment on usage.
+ (Duplicate_Subexpr_No_Checks): Add formal parameter Renaming_Req.
+ (Duplicate_Subexpr_Move_Checks): Add formal parameter
+ Renaming_Req.
+
2014-08-01 Bob Duff <duff@adacore.com>
* gnat_ugn.texi: Minor updates.
end if;
end if;
- -- Normal processing for non-controlled types
+ -- Normal processing for non-controlled types. The argument to free is
+ -- a renaming rather than a constant to ensure that the original context
+ -- is always set to null after the deallocation takes place.
- Free_Arg := Duplicate_Subexpr_No_Checks (Arg);
+ Free_Arg := Duplicate_Subexpr_No_Checks (Arg, Renaming_Req => True);
Free_Node := Make_Free_Statement (Loc, Empty);
Append_To (Stmts, Free_Node);
Set_Storage_Pool (Free_Node, Pool);
-----------------------
function Duplicate_Subexpr
- (Exp : Node_Id;
- Name_Req : Boolean := False) return Node_Id
+ (Exp : Node_Id;
+ Name_Req : Boolean := False;
+ Renaming_Req : Boolean := False) return Node_Id
is
begin
- Remove_Side_Effects (Exp, Name_Req);
+ Remove_Side_Effects (Exp, Name_Req, Renaming_Req);
return New_Copy_Tree (Exp);
end Duplicate_Subexpr;
---------------------------------
function Duplicate_Subexpr_No_Checks
- (Exp : Node_Id;
- Name_Req : Boolean := False) return Node_Id
+ (Exp : Node_Id;
+ Name_Req : Boolean := False;
+ Renaming_Req : Boolean := False) return Node_Id
is
New_Exp : Node_Id;
+
begin
- Remove_Side_Effects (Exp, Name_Req);
+ Remove_Side_Effects (Exp, Name_Req, Renaming_Req);
New_Exp := New_Copy_Tree (Exp);
Remove_Checks (New_Exp);
return New_Exp;
-----------------------------------
function Duplicate_Subexpr_Move_Checks
- (Exp : Node_Id;
- Name_Req : Boolean := False) return Node_Id
+ (Exp : Node_Id;
+ Name_Req : Boolean := False;
+ Renaming_Req : Boolean := False) return Node_Id
is
New_Exp : Node_Id;
+
begin
- Remove_Side_Effects (Exp, Name_Req);
+ Remove_Side_Effects (Exp, Name_Req, Renaming_Req);
New_Exp := New_Copy_Tree (Exp);
Remove_Checks (Exp);
return New_Exp;
procedure Remove_Side_Effects
(Exp : Node_Id;
Name_Req : Boolean := False;
+ Renaming_Req : Boolean := False;
Variable_Ref : Boolean := False)
is
Loc : constant Source_Ptr := Sloc (Exp);
Set_Analyzed (Prefix (Exp), False);
end if;
- E :=
- Make_Object_Declaration (Loc,
- Defining_Identifier => Def_Id,
- Object_Definition => New_Occurrence_Of (Exp_Type, Loc),
- Constant_Present => True,
- Expression => Relocate_Node (Exp));
+ -- Generate:
+ -- Rnn : Exp_Type renames Expr;
+
+ if Renaming_Req then
+ E :=
+ Make_Object_Renaming_Declaration (Loc,
+ Defining_Identifier => Def_Id,
+ Subtype_Mark => New_Occurrence_Of (Exp_Type, Loc),
+ Name => Relocate_Node (Exp));
+
+ -- Generate:
+ -- Rnn : constant Exp_Type := Expr;
+
+ else
+ E :=
+ Make_Object_Declaration (Loc,
+ Defining_Identifier => Def_Id,
+ Object_Definition => New_Occurrence_Of (Exp_Type, Loc),
+ Constant_Present => True,
+ Expression => Relocate_Node (Exp));
+
+ Set_Assignment_OK (E);
+ end if;
- Set_Assignment_OK (E);
Insert_Action (Exp, E);
-- If the expression has the form v.all then we can just capture the
-- be the earliest point at which they are used.
function Duplicate_Subexpr
- (Exp : Node_Id;
- Name_Req : Boolean := False) return Node_Id;
+ (Exp : Node_Id;
+ Name_Req : Boolean := False;
+ Renaming_Req : Boolean := False) return Node_Id;
-- Given the node for a subexpression, this function makes a logical copy
-- of the subexpression, and returns it. This is intended for use when the
-- expansion of an expression needs to repeat part of it. For example,
-- expression and the returned result then become references to this saved
-- value. Exp must be analyzed on entry. On return, Exp is analyzed, but
-- the caller is responsible for analyzing the returned copy after it is
- -- attached to the tree. The Name_Req flag is set to ensure that the result
- -- is suitable for use in a context requiring name (e.g. the prefix of an
- -- attribute reference).
+ -- attached to the tree.
+ --
+ -- The Name_Req flag is set to ensure that the result is suitable for use
+ -- in a context requiring a name (for example, the prefix of an attribute
+ -- reference) (can't this just be a qualification in Ada 2012???).
+ --
+ -- The Renaming_Req flag is set to produce an object renaming declaration
+ -- rather than an object declaration. This is valid only if the expression
+ -- Exp designates a renamable object. This is used for example in the case
+ -- of an unchecked deallocation, to make sure the object gets set to null.
--
-- Note that if there are any run time checks in Exp, these same checks
-- will be duplicated in the returned duplicated expression. The two
-- following functions allow this behavior to be modified.
function Duplicate_Subexpr_No_Checks
- (Exp : Node_Id;
- Name_Req : Boolean := False) return Node_Id;
+ (Exp : Node_Id;
+ Name_Req : Boolean := False;
+ Renaming_Req : Boolean := False) return Node_Id;
-- Identical in effect to Duplicate_Subexpr, except that Remove_Checks
-- is called on the result, so that the duplicated expression does not
-- include checks. This is appropriate for use when Exp, the original
-- expression, so that there is no need to repeat any checks.
function Duplicate_Subexpr_Move_Checks
- (Exp : Node_Id;
- Name_Req : Boolean := False) return Node_Id;
+ (Exp : Node_Id;
+ Name_Req : Boolean := False;
+ Renaming_Req : Boolean := False) return Node_Id;
-- Identical in effect to Duplicate_Subexpr, except that Remove_Checks is
-- called on Exp after the duplication is complete, so that the original
-- expression does not include checks. In this case the result returned
procedure Remove_Side_Effects
(Exp : Node_Id;
Name_Req : Boolean := False;
+ Renaming_Req : Boolean := False;
Variable_Ref : Boolean := False);
-- Given the node for a subexpression, this function replaces the node if
-- necessary by an equivalent subexpression that is guaranteed to be side
-- to which Exp is attached. Exp must be analyzed and resolved before the
-- call and is analyzed and resolved on return. Name_Req may only be set to
-- True if Exp has the form of a name, and the effect is to guarantee that
- -- any replacement maintains the form of name. If Variable_Ref is set to
- -- TRUE, a variable is considered as side effect (used in implementing
- -- Force_Evaluation). Note: after call to Remove_Side_Effects, it is
- -- safe to call New_Copy_Tree to obtain a copy of the resulting expression.
+ -- any replacement maintains the form of name. If Renaming_Req is set to
+ -- TRUE, the routine produces an object renaming reclaration capturing the
+ -- expression. If Variable_Ref is set to TRUE, a variable is considered as
+ -- side effect (used in implementing Force_Evaluation). Note: after call to
+ -- Remove_Side_Effects, it is safe to call New_Copy_Tree to obtain a copy
+ -- of the resulting expression.
function Represented_As_Scalar (T : Entity_Id) return Boolean;
-- Returns True iff the implementation of this type in code generation
-- GNAT
-- True if compiling in GNAT system mode (-gnatg switch)
- -- Setting this switch has the following effects
+ -- Setting this switch has the following effects on the language that is
+ -- accepted. Note that several of the following have the effect of changing
+ -- an error to a warning. But warnings are usually treated as fatal errors
+ -- in -gnatg mode, so to actually take advantage of such a change, it is
+ -- necessary to add an explicit pragma Warnings (Off) in the source and
+ -- this requires clear documentation of why this is necessary.
-- The identifier character set is set to 'n' (7-bit ASCII)
-- Returning objects of limited types is allowed
- -- All entities are considered known to Known_But_Invisible
-
-- Non-static call in preelaborated unit give a warning, not an error
-- Warnings on possible elaboration errors are suppressed
- -- Warning about packing being ignored is suppressed
+ -- Warnings about packing being ignored are suppressed
-- Warnings in internal units are not suppressed (they normally are)
-- for that processing
function Known_But_Invisible (E : Entity_Id) return Boolean;
- -- This function determines whether the entity E (which is not
- -- visible) can reasonably be considered to be known to the writer
- -- of the reference. This is a heuristic test, used only for the
- -- purposes of figuring out whether we prefer to complain that an
- -- entity is undefined or invisible (and identify the declaration
- -- of the invisible entity in the latter case). The point here is
- -- that we don't want to complain that something is invisible and
- -- then point to something entirely mysterious to the writer.
+ -- This function determines whether a reference to the entity E, which
+ -- is not visible, can reasonably be considered to be known to the
+ -- writer of the reference. This is a heuristic test, used only for
+ -- the purposes of figuring out whether we prefer to complain that an
+ -- entity is undefined or invisible (and identify the declaration of
+ -- the invisible entity in the latter case). The point here is that we
+ -- don't want to complain that something is invisible and then point to
+ -- something entirely mysterious to the writer.
procedure Nvis_Messages;
-- Called if there are no visible entries for N, but there is at least
elsif not Comes_From_Source (E) then
return False;
- -- In gnat internal mode, we consider all entities known
+ -- In gnat internal mode, we consider all entities known. The
+ -- historical reason behind this discrepancy is not known??? But the
+ -- only effect is to modify the error message given, so it is not
+ -- critical. Since it only affects the exact wording of error
+ -- messages in illegal programs, we do not mention this as an
+ -- effect of -gnatg, since it is not a language modification.
elsif GNAT_Mode then
return True;
projects / gcc.git / commitdiff