1 ------------------------------------------------------------------------------
3 -- GNAT COMPILER COMPONENTS --
9 -- Copyright (C) 1992-2006, Free Software Foundation, Inc. --
11 -- GNAT is free software; you can redistribute it and/or modify it under --
12 -- terms of the GNU General Public License as published by the Free Soft- --
13 -- ware Foundation; either version 2, or (at your option) any later ver- --
14 -- sion. GNAT is distributed in the hope that it will be useful, but WITH- --
15 -- OUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY --
16 -- or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License --
17 -- for more details. You should have received a copy of the GNU General --
18 -- Public License distributed with GNAT; see file COPYING. If not, write --
19 -- to the Free Software Foundation, 51 Franklin Street, Fifth Floor, --
20 -- Boston, MA 02110-1301, USA. --
22 -- GNAT was originally developed by the GNAT team at New York University. --
23 -- Extensive contributions were provided by Ada Core Technologies Inc. --
25 ------------------------------------------------------------------------------
27 -- Package containing routines used to deal with runtime checks. These
28 -- routines are used both by the semantics and by the expander. In some
29 -- cases, checks are enabled simply by setting flags for gigi, and in
30 -- other cases the code for the check is expanded.
32 -- The approach used for range and length checks, in regards to suppressed
33 -- checks, is to attempt to detect at compilation time that a constraint
34 -- error will occur. If this is detected a warning or error is issued and the
35 -- offending expression or statement replaced with a constraint error node.
36 -- This always occurs whether checks are suppressed or not. Dynamic range
37 -- checks are, of course, not inserted if checks are suppressed.
39 with Types; use Types;
40 with Uintp; use Uintp;
45 -- Called for each new main source program, to initialize internal
46 -- variables used in the package body of the Checks unit.
48 function Access_Checks_Suppressed (E : Entity_Id) return Boolean;
49 function Accessibility_Checks_Suppressed (E : Entity_Id) return Boolean;
50 function Alignment_Checks_Suppressed (E : Entity_Id) return Boolean;
51 function Discriminant_Checks_Suppressed (E : Entity_Id) return Boolean;
52 function Division_Checks_Suppressed (E : Entity_Id) return Boolean;
53 function Elaboration_Checks_Suppressed (E : Entity_Id) return Boolean;
54 function Index_Checks_Suppressed (E : Entity_Id) return Boolean;
55 function Length_Checks_Suppressed (E : Entity_Id) return Boolean;
56 function Overflow_Checks_Suppressed (E : Entity_Id) return Boolean;
57 function Range_Checks_Suppressed (E : Entity_Id) return Boolean;
58 function Storage_Checks_Suppressed (E : Entity_Id) return Boolean;
59 function Tag_Checks_Suppressed (E : Entity_Id) return Boolean;
60 function Validity_Checks_Suppressed (E : Entity_Id) return Boolean;
61 -- These functions check to see if the named check is suppressed, either
62 -- by an active scope suppress setting, or because the check has been
63 -- specifically suppressed for the given entity. If no entity is relevant
64 -- for the current check, then Empty is used as an argument. Note: the
65 -- reason we insist on specifying Empty is to force the caller to think
66 -- about whether there is any relevant entity that should be checked.
68 -- General note on following checks. These checks are always active if
69 -- Expander_Active and not Inside_A_Generic. They are inactive and have
70 -- no effect Inside_A_Generic. In the case where not Expander_Active
71 -- and not Inside_A_Generic, most of them are inactive, but some of them
72 -- operate anyway since they may generate useful compile time warnings.
74 procedure Apply_Access_Check (N : Node_Id);
75 -- Determines whether an expression node requires a runtime access
76 -- check and if so inserts the appropriate run-time check.
78 procedure Apply_Accessibility_Check (N : Node_Id; Typ : Entity_Id);
79 -- Given a name N denoting an access parameter, emits a run-time
80 -- accessibility check (if necessary), checking that the level of
81 -- the object denoted by the access parameter is not deeper than the
82 -- level of the type Typ. Program_Error is raised if the check fails.
84 procedure Apply_Address_Clause_Check (E : Entity_Id; N : Node_Id);
85 -- E is the entity for an object which has an address clause. If checks
86 -- are enabled, then this procedure generates a check that the specified
87 -- address has an alignment consistent with the alignment of the object,
88 -- raising PE if this is not the case. The resulting check (if one is
89 -- generated) is inserted before node N. check is also made for the case of
90 -- a clear overlay situation that the size of the overlaying object is not
91 -- larger than the overlaid object.
93 procedure Apply_Array_Size_Check (N : Node_Id; Typ : Entity_Id);
94 -- N is the node for an object declaration that declares an object of
95 -- array type Typ. This routine generates, if necessary, a check that
96 -- the size of the array is not too large, raising Storage_Error if so.
98 procedure Apply_Arithmetic_Overflow_Check (N : Node_Id);
99 -- Given a binary arithmetic operator (+ - *) expand a software integer
100 -- overflow check using range checks on a larger checking type or a call
101 -- to an appropriate runtime routine. This is used for all three operators
102 -- for the signed integer case, and for +/- in the fixed-point case. The
103 -- check is expanded only if Software_Overflow_Checking is enabled and
104 -- Do_Overflow_Check is set on node N. Note that divide is handled
105 -- separately using Apply_Arithmetic_Divide_Overflow_Check.
107 procedure Apply_Constraint_Check
110 No_Sliding : Boolean := False);
111 -- Top-level procedure, calls all the others depending on the class of Typ.
112 -- Checks that expression N verifies the constraint of type Typ. No_Sliding
113 -- is only relevant for constrained array types, id set to true, it
114 -- checks that indexes are in range.
116 procedure Apply_Discriminant_Check
119 Lhs : Node_Id := Empty);
120 -- Given an expression N of a discriminated type, or of an access type
121 -- whose designated type is a discriminanted type, generates a check to
122 -- ensure that the expression can be converted to the subtype given as
123 -- the second parameter. Lhs is empty except in the case of assignments,
124 -- where the target object may be needed to determine the subtype to
125 -- check against (such as the cases of unconstrained formal parameters
126 -- and unconstrained aliased objects). For the case of unconstrained
127 -- formals, the check is peformed only if the corresponding actual is
128 -- constrained, i.e., whether Lhs'Constrained is True.
130 function Build_Discriminant_Checks
134 -- Subsidiary routine for Apply_Discriminant_Check. Builds the expression
135 -- that compares discriminants of the expression with discriminants of the
136 -- type. Also used directly for membership tests (see Exp_Ch4.Expand_N_In).
138 procedure Apply_Divide_Check (N : Node_Id);
139 -- The node kind is N_Op_Divide, N_Op_Mod, or N_Op_Rem. An appropriate
140 -- check is generated to ensure that the right operand is non-zero. In
141 -- the divide case, we also check that we do not have the annoying case
142 -- of the largest negative number divided by minus one.
144 procedure Apply_Type_Conversion_Checks (N : Node_Id);
145 -- N is an N_Type_Conversion node. A type conversion actually involves
146 -- two sorts of checks. The first check is the checks that ensures that
147 -- the operand in the type conversion fits onto the base type of the
148 -- subtype it is being converted to (see RM 4.6 (28)-(50)). The second
149 -- check is there to ensure that once the operand has been converted to
150 -- a value of the target type, this converted value meets the
151 -- constraints imposed by the target subtype (see RM 4.6 (51)).
153 procedure Apply_Universal_Integer_Attribute_Checks (N : Node_Id);
154 -- The argument N is an attribute reference node intended for processing
155 -- by gigi. The attribute is one that returns a universal integer, but
156 -- the attribute reference node is currently typed with the expected
157 -- result type. This routine deals with range and overflow checks needed
158 -- to make sure that the universal result is in range.
160 procedure Determine_Range
165 -- N is a node for a subexpression. If N is of a discrete type with no
166 -- error indications, and no other peculiarities (e.g. missing type
167 -- fields), then OK is True on return, and Lo and Hi are set to a
168 -- conservative estimate of the possible range of values of N. Thus if OK
169 -- is True on return, the value of the subexpression N is known to like in
170 -- the range Lo .. Hi (inclusive). If the expression is not of a discrete
171 -- type, or some kind of error condition is detected, then OK is False on
172 -- exit, and Lo/Hi are set to No_Uint. Thus the significance of OK being
173 -- False on return is that no useful information is available on the range
174 -- of the expression.
176 procedure Install_Null_Excluding_Check (N : Node_Id);
177 -- Determines whether an access node requires a runtime access check and
178 -- if so inserts the appropriate run-time check.
180 -------------------------------------------------------
181 -- Control and Optimization of Range/Overflow Checks --
182 -------------------------------------------------------
184 -- Range checks are controlled by the Do_Range_Check flag. The front end
185 -- is responsible for setting this flag in relevant nodes. Originally
186 -- the back end generated all corresponding range checks. But later on
187 -- we decided to generate all range checks in the front end. We are now
188 -- in the transitional phase where some of these checks are still done
189 -- by the back end, but many are done by the front end.
191 -- Overflow checks are similarly controlled by the Do_Overflow_Check flag.
192 -- The difference here is that if Backend_Overflow_Checks is is
193 -- (Backend_Overflow_Checks_On_Target set False), then the actual overflow
194 -- checks are generated by the front end, but if back end overflow checks
195 -- are active (Backend_Overflow_Checks_On_Target set True), then the back
196 -- end does generate the checks.
198 -- The following two routines are used to set these flags, they allow
199 -- for the possibility of eliminating checks. Checks can be eliminated
200 -- if an identical check has already been performed.
202 procedure Enable_Overflow_Check (N : Node_Id);
203 -- First this routine determines if an overflow check is needed by doing
204 -- an appropriate range check. If a check is not needed, then the call
205 -- has no effect. If a check is needed then this routine sets the flag
206 -- Set Do_Overflow_Check in node N to True, unless it can be determined
207 -- that the check is not needed. The only condition under which this is
208 -- the case is if there was an identical check earlier on.
210 procedure Enable_Range_Check (N : Node_Id);
211 -- Set Do_Range_Check flag in node N True, unless it can be determined
212 -- that the check is not needed. The only condition under which this is
213 -- the case is if there was an identical check earlier on. This routine
214 -- is not responsible for doing range analysis to determine whether or
215 -- not such a check is needed -- the caller is expected to do this. The
216 -- one other case in which the request to set the flag is ignored is
217 -- when Kill_Range_Check is set in an N_Unchecked_Conversion node.
219 -- The following routines are used to keep track of processing sequences
220 -- of statements (e.g. the THEN statements of an IF statement). A check
221 -- that appears within such a sequence can eliminate an identical check
222 -- within this sequence of statements. However, after the end of the
223 -- sequence of statements, such a check is no longer of interest, since
224 -- it may not have been executed.
226 procedure Conditional_Statements_Begin;
227 -- This call marks the start of processing of a sequence of statements.
228 -- Every call to this procedure must be followed by a matching call to
229 -- Conditional_Statements_End.
231 procedure Conditional_Statements_End;
232 -- This call removes from consideration all saved checks since the
233 -- corresponding call to Conditional_Statements_Begin. These two
234 -- procedures operate in a stack like manner.
236 -- The mechanism for optimizing checks works by remembering checks
237 -- that have already been made, but certain conditions, for example
238 -- an assignment to a variable involved in a check, may mean that the
239 -- remembered check is no longer valid, in the sense that if the same
240 -- expression appears again, another check is required because the
241 -- value may have changed.
243 -- The following routines are used to note conditions which may render
244 -- some or all of the stored and remembered checks to be invalidated.
246 procedure Kill_Checks (V : Entity_Id);
247 -- This procedure records an assignment or other condition that causes
248 -- the value of the variable to be changed, invalidating any stored
249 -- checks that reference the value. Note that all such checks must
250 -- be discarded, even if they are not in the current statement range.
252 procedure Kill_All_Checks;
253 -- This procedure kills all remembered checks
255 -----------------------------
256 -- Length and Range Checks --
257 -----------------------------
259 -- In the following procedures, there are three arguments which have
260 -- a common meaning as follows:
262 -- Expr The expression to be checked. If a check is required,
263 -- the appropriate flag will be placed on this node. Whether
264 -- this node is further examined depends on the setting of
265 -- the parameter Source_Typ, as described below.
267 -- ??? Apply_Length_Check and Apply_Range_Check do not have an Expr
270 -- ??? Apply_Length_Check and Apply_Range_Check have a Ck_Node formal
271 -- which is undocumented, is it the same as Expr?
273 -- Target_Typ The target type on which the check is to be based. For
274 -- example, if we have a scalar range check, then the check
275 -- is that we are in range of this type.
277 -- Source_Typ Normally Empty, but can be set to a type, in which case
278 -- this type is used for the check, see below.
280 -- The checks operate in one of two modes:
282 -- If Source_Typ is Empty, then the node Expr is examined, at the very
283 -- least to get the source subtype. In addition for some of the checks,
284 -- the actual form of the node may be examined. For example, a node of
285 -- type Integer whose actual form is an Integer conversion from a type
286 -- with range 0 .. 3 can be determined to have a value in range 0 .. 3.
288 -- If Source_Typ is given, then nothing can be assumed about the Expr,
289 -- and indeed its contents are not examined. In this case the check is
290 -- based on the assumption that Expr can be an arbitrary value of the
293 -- Currently, the only case in which a Source_Typ is explicitly supplied
294 -- is for the case of Out and In_Out parameters, where, for the conversion
295 -- on return (the Out direction), the types must be reversed. This is
296 -- handled by the caller.
298 procedure Apply_Length_Check
300 Target_Typ : Entity_Id;
301 Source_Typ : Entity_Id := Empty);
302 -- This procedure builds a sequence of declarations to do a length check
303 -- that checks if the lengths of the two arrays Target_Typ and source type
304 -- are the same. The resulting actions are inserted at Node using a call
305 -- to Insert_Actions.
307 -- For access types, the Directly_Designated_Type is retrieved and
308 -- processing continues as enumerated above, with a guard against null
311 -- Note: calls to Apply_Length_Check currently never supply an explicit
312 -- Source_Typ parameter, but Apply_Length_Check takes this parameter and
313 -- processes it as described above for consistency with the other routines
316 procedure Apply_Range_Check
318 Target_Typ : Entity_Id;
319 Source_Typ : Entity_Id := Empty);
320 -- For a Node of kind N_Range, constructs a range check action that tests
321 -- first that the range is not null and then that the range is contained in
322 -- the Target_Typ range.
324 -- For scalar types, constructs a range check action that first tests that
325 -- the expression is contained in the Target_Typ range. The difference
326 -- between this and Apply_Scalar_Range_Check is that the latter generates
327 -- the actual checking code in gigi against the Etype of the expression.
329 -- For constrained array types, construct series of range check actions
330 -- to check that each Expr range is properly contained in the range of
333 -- For a type conversion to an unconstrained array type, constructs a range
334 -- check action to check that the bounds of the source type are within the
335 -- constraints imposed by the Target_Typ.
337 -- For access types, the Directly_Designated_Type is retrieved and
338 -- processing continues as enumerated above, with a guard against null
341 -- The source type is used by type conversions to unconstrained array
342 -- types to retrieve the corresponding bounds.
344 procedure Apply_Static_Length_Check
346 Target_Typ : Entity_Id;
347 Source_Typ : Entity_Id := Empty);
348 -- Tries to determine statically whether the two array types source type
349 -- and Target_Typ have the same length. If it can be determined at compile
350 -- time that they do not, then an N_Raise_Constraint_Error node replaces
351 -- Expr, and a warning message is issued.
353 procedure Apply_Scalar_Range_Check
355 Target_Typ : Entity_Id;
356 Source_Typ : Entity_Id := Empty;
357 Fixed_Int : Boolean := False);
358 -- For scalar types, determines whether an expression node should be
359 -- flagged as needing a runtime range check. If the node requires such a
360 -- check, the Do_Range_Check flag is turned on. The Fixed_Int flag if set
361 -- causes any fixed-point values to be treated as though they were discrete
362 -- values (i.e. the underlying integer value is used).
364 type Check_Result is private;
365 -- Type used to return result of Range_Check call, for later use in
366 -- call to Insert_Range_Checks procedure.
368 procedure Append_Range_Checks
369 (Checks : Check_Result;
371 Suppress_Typ : Entity_Id;
372 Static_Sloc : Source_Ptr;
373 Flag_Node : Node_Id);
374 -- Called to append range checks as returned by a call to Range_Check.
375 -- Stmts is a list to which either the dynamic check is appended or the
376 -- raise Constraint_Error statement is appended (for static checks).
377 -- Static_Sloc is the Sloc at which the raise CE node points, Flag_Node is
378 -- used as the node at which to set the Has_Dynamic_Check flag. Checks_On
379 -- is a boolean value that says if range and index checking is on or not.
381 procedure Insert_Range_Checks
382 (Checks : Check_Result;
384 Suppress_Typ : Entity_Id;
385 Static_Sloc : Source_Ptr := No_Location;
386 Flag_Node : Node_Id := Empty;
387 Do_Before : Boolean := False);
388 -- Called to insert range checks as returned by a call to Range_Check.
389 -- Node is the node after which either the dynamic check is inserted or
390 -- the raise Constraint_Error statement is inserted (for static checks).
391 -- Suppress_Typ is the type to check to determine if checks are suppressed.
392 -- Static_Sloc, if passed, is the Sloc at which the raise CE node points,
393 -- otherwise Sloc (Node) is used. The Has_Dynamic_Check flag is normally
394 -- set at Node. If Flag_Node is present, then this is used instead as the
395 -- node at which to set the Has_Dynamic_Check flag. Normally the check is
396 -- inserted after, if Do_Before is True, the check is inserted before
401 Target_Typ : Entity_Id;
402 Source_Typ : Entity_Id := Empty;
403 Warn_Node : Node_Id := Empty)
405 -- Like Apply_Range_Check, except it does not modify anything. Instead
406 -- it returns an encapsulated result of the check operations for later
407 -- use in a call to Insert_Range_Checks. If Warn_Node is non-empty, its
408 -- Sloc is used, in the static case, for the generated warning or error.
409 -- Additionally, it is used rather than Expr (or Low/High_Bound of Expr)
410 -- in constructing the check.
412 -----------------------
413 -- Expander Routines --
414 -----------------------
416 -- Some of the earlier processing for checks results in temporarily setting
417 -- the Do_Range_Check flag rather than actually generating checks. Now we
418 -- are moving the generation of such checks into the front end for reasons
419 -- of efficiency and simplicity (there were difficutlies in handling this
420 -- in the back end when side effects were present in the expressions being
423 -- Probably we could eliminate the Do_Range_Check flag entirely and
424 -- generate the checks earlier, but this is a delicate area and it
425 -- seemed safer to implement the following routines, which are called
426 -- late on in the expansion process. They check the Do_Range_Check flag
427 -- and if it is set, generate the actual checks and reset the flag.
429 procedure Generate_Range_Check
431 Target_Type : Entity_Id;
432 Reason : RT_Exception_Code);
433 -- This procedure is called to actually generate and insert a range check.
434 -- A check is generated to ensure that the value of N lies within the range
435 -- of the target type. Note that the base type of N may be different from
436 -- the base type of the target type. This happens in the conversion case.
437 -- The Reason parameter is the exception code to be used for the exception
440 -- Note on the relation of this routine to the Do_Range_Check flag. Mostly
441 -- for historical reasons, we often set the Do_Range_Check flag and then
442 -- later we call Generate_Range_Check if this flag is set. Most probably we
443 -- could eliminate this intermediate setting of the flag (historically the
444 -- back end dealt with range checks, using this flag to indicate if a check
445 -- was required, then we moved checks into the front end).
447 procedure Generate_Index_Checks (N : Node_Id);
448 -- This procedure is called to generate index checks on the subscripts for
449 -- the indexed component node N. Each subscript expression is examined, and
450 -- if the Do_Range_Check flag is set, an appropriate index check is
451 -- generated and the flag is reset.
453 -- Similarly, we set the flag Do_Discriminant_Check in the semantic
454 -- analysis to indicate that a discriminant check is required for selected
455 -- component of a discriminated type. The following routine is called from
456 -- the expander to actually generate the call.
458 procedure Generate_Discriminant_Check (N : Node_Id);
459 -- N is a selected component for which a discriminant check is required to
460 -- make sure that the discriminants have appropriate values for the
461 -- selection. This is done by calling the appropriate discriminant checking
462 -- routine for the selector.
464 -----------------------
465 -- Validity Checking --
466 -----------------------
468 -- In (RM 13.9.1(9-11)) we have the following rules on invalid values
470 -- If the representation of a scalar object does not represent value of
471 -- the object's subtype (perhaps because the object was not initialized),
472 -- the object is said to have an invalid representation. It is a bounded
473 -- error to evaluate the value of such an object. If the error is
474 -- detected, either Constraint_Error or Program_Error is raised.
475 -- Otherwise, execution continues using the invalid representation. The
476 -- rules of the language outside this subclause assume that all objects
477 -- have valid representations. The semantics of operations on invalid
478 -- representations are as follows:
480 -- 10 If the representation of the object represents a value of the
481 -- object's type, the value of the type is used.
483 -- 11 If the representation of the object does not represent a value
484 -- of the object's type, the semantics of operations on such
485 -- representations is implementation-defined, but does not by
486 -- itself lead to erroneous or unpredictable execution, or to
487 -- other objects becoming abnormal.
489 -- We quote the rules in full here since they are quite delicate. Most
490 -- of the time, we can just compute away with wrong values, and get a
491 -- possibly wrong result, which is well within the range of allowed
492 -- implementation defined behavior. The two tricky cases are subscripted
493 -- array assignments, where we don't want to do wild stores, and case
494 -- statements where we don't want to do wild jumps.
496 -- In GNAT, we control validity checking with a switch -gnatV that can take
497 -- three parameters, n/d/f for None/Default/Full. These modes have the
498 -- following meanings:
500 -- None (no validity checking)
502 -- In this mode, there is no specific checking for invalid values
503 -- and the code generator assumes that all stored values are always
504 -- within the bounds of the object subtype. The consequences are as
507 -- For case statements, an out of range invalid value will cause
508 -- Constraint_Error to be raised, or an arbitrary one of the case
509 -- alternatives will be executed. Wild jumps cannot result even
510 -- in this mode, since we always do a range check
512 -- For subscripted array assignments, wild stores will result in
513 -- the expected manner when addresses are calculated using values
514 -- of subscripts that are out of range.
516 -- It could perhaps be argued that this mode is still conformant with
517 -- the letter of the RM, since implementation defined is a rather
518 -- broad category, but certainly it is not in the spirit of the
519 -- RM requirement, since wild stores certainly seem to be a case of
520 -- erroneous behavior.
522 -- Default (default standard RM-compatible validity checking)
524 -- In this mode, which is the default, minimal validity checking is
525 -- performed to ensure no erroneous behavior as follows:
527 -- For case statements, an out of range invalid value will cause
528 -- Constraint_Error to be raised.
530 -- For subscripted array assignments, invalid out of range
531 -- subscript values will cause Constraint_Error to be raised.
533 -- Full (Full validity checking)
535 -- In this mode, the protections guaranteed by the standard mode are
536 -- in place, and the following additional checks are made:
538 -- For every assignment, the right side is checked for validity
540 -- For every call, IN and IN OUT parameters are checked for validity
542 -- For every subscripted array reference, both for stores and loads,
543 -- all subscripts are checked for validity.
545 -- These checks are not required by the RM, but will in practice
546 -- improve the detection of uninitialized variables, particularly
547 -- if used in conjunction with pragma Normalize_Scalars.
549 -- In the above description, we talk about performing validity checks,
550 -- but we don't actually generate a check in a case where the compiler
551 -- can be sure that the value is valid. Note that this assurance must
552 -- be achieved without assuming that any uninitialized value lies within
553 -- the range of its type. The following are cases in which values are
554 -- known to be valid. The flag Is_Known_Valid is used to keep track of
555 -- some of these cases.
557 -- If all possible stored values are valid, then any uninitialized
558 -- value must be valid.
560 -- Literals, including enumeration literals, are clearly always valid
562 -- Constants are always assumed valid, with a validity check being
563 -- performed on the initializing value where necessary to ensure that
566 -- For variables, the status is set to known valid if there is an
567 -- initializing expression. Again a check is made on the initializing
568 -- value if necessary to ensure that this assumption is valid. The
569 -- status can change as a result of local assignments to a variable.
570 -- If a known valid value is unconditionally assigned, then we mark
571 -- the left side as known valid. If a value is assigned that is not
572 -- known to be valid, then we mark the left side as invalid. This
573 -- kind of processing does NOT apply to non-local variables since we
574 -- are not following the flow graph (more properly the flow of actual
575 -- processing only corresponds to the flow graph for local assignments).
576 -- For non-local variables, we preserve the current setting, i.e. a
577 -- validity check is performed when assigning to a knonwn valid global.
579 -- Note: no validity checking is required if range checks are suppressed
580 -- regardless of the setting of the validity checking mode.
582 -- The following procedures are used in handling validity checking
584 procedure Apply_Subscript_Validity_Checks (Expr : Node_Id);
585 -- Expr is the node for an indexed component. If validity checking and
586 -- range checking are enabled, all subscripts for this indexed component
587 -- are checked for validity.
589 procedure Check_Valid_Lvalue_Subscripts (Expr : Node_Id);
590 -- Expr is a lvalue, i.e. an expression representing the target of an
591 -- assignment. This procedure checks for this expression involving an
592 -- assignment to an array value. We have to be sure that all the subscripts
593 -- in such a case are valid, since according to the rules in (RM
594 -- 13.9.1(9-11)) such assignments are not permitted to result in erroneous
595 -- behavior in the case of invalid subscript values.
597 procedure Ensure_Valid (Expr : Node_Id; Holes_OK : Boolean := False);
598 -- Ensure that Expr represents a valid value of its type. If this type
599 -- is not a scalar type, then the call has no effect, since validity
600 -- is only an issue for scalar types. The effect of this call is to
601 -- check if the value is known valid, if so, nothing needs to be done.
602 -- If this is not known, then either Expr is set to be range checked,
603 -- or specific checking code is inserted so that an exception is raised
604 -- if the value is not valid.
606 -- The optional argument Holes_OK indicates whether it is necessary to
607 -- worry about enumeration types with non-standard representations leading
608 -- to "holes" in the range of possible representations. If Holes_OK is
609 -- True, then such values are assumed valid (this is used when the caller
610 -- will make a separate check for this case anyway). If Holes_OK is False,
611 -- then this case is checked, and code is inserted to ensure that Expr is
612 -- valid, raising Constraint_Error if the value is not valid.
614 function Expr_Known_Valid (Expr : Node_Id) return Boolean;
615 -- This function tests it the value of Expr is known to be valid in the
616 -- sense of RM 13.9.1(9-11). In the case of GNAT, it is only discrete types
617 -- which are a concern, since for non-discrete types we simply continue
618 -- computation with invalid values, which does not lead to erroneous
619 -- behavior. Thus Expr_Known_Valid always returns True if the type of Expr
620 -- is non-discrete. For discrete types the value returned is True only if
621 -- it can be determined that the value is Valid. Otherwise False is
624 procedure Insert_Valid_Check (Expr : Node_Id);
625 -- Inserts code that will check for the value of Expr being valid, in
626 -- the sense of the 'Valid attribute returning True. Constraint_Error
627 -- will be raised if the value is not valid.
629 procedure Null_Exclusion_Static_Checks (N : Node_Id);
630 -- Ada 2005 (AI-231): Check bad usages of the null-exclusion issue
632 procedure Remove_Checks (Expr : Node_Id);
633 -- Remove all checks from Expr except those that are only executed
634 -- conditionally (on the right side of And Then/Or Else. This call
635 -- removes only embedded checks (Do_Range_Check, Do_Overflow_Check).
637 procedure Validity_Check_Range (N : Node_Id);
638 -- If N is an N_Range node, then Ensure_Valid is called on its bounds,
639 -- if validity checking of operands is enabled.
643 type Check_Result is array (Positive range 1 .. 2) of Node_Id;
644 -- There are two cases for the result returned by Range_Check:
646 -- For the static case the result is one or two nodes that should cause
647 -- a Constraint_Error. Typically these will include Expr itself or the
648 -- direct descendents of Expr, such as Low/High_Bound (Expr)). It is the
649 -- responsibility of the caller to rewrite and substitute the nodes with
650 -- N_Raise_Constraint_Error nodes.
652 -- For the non-static case a single N_Raise_Constraint_Error node with a
653 -- non-empty Condition field is returned.
655 -- Unused entries in Check_Result, if any, are simply set to Empty For
656 -- external clients, the required processing on this result is achieved
657 -- using the Insert_Range_Checks routine.
659 pragma Inline (Apply_Length_Check);
660 pragma Inline (Apply_Range_Check);
661 pragma Inline (Apply_Static_Length_Check);
projects / gcc.git / blob